6.1.1 CORESE

• Name: COnceptual REsource Search Engine
• Keywords: Semantic Web, Search Engine, RDF, SPARQL
• Functional Description:

  Corese is a Semantic Web Factory, it implements W3C RDF, RDFS, OWL RL, SHACL, SPARQL 1 .1 Query and Update as well as RDF Inference Rules.

  Furthermore, Corese query language integrates original features such as approximate search and extended Property Path. It provides STTL: SPARQL Template Transformation Language for RDF graphs. It also provides LDScript: a Script Language for Linked Data. Corese provides distributed federated query processing.

• URL: http://project.inria.fr/corese
• Contact: Olivier Corby
• Participants: Erwan Demairy, Fabien Gandon, Fuqi Song, Olivier Corby, Olivier Savoie, Virginie Bottollier
• Partners: I3S, Mnemotix

6.1.2 DBpedia

• Name: DBpedia
• Keywords: RDF, SPARQL
• Functional Description: DBpedia is an international crowd-sourced community effort to extract structured information from Wikipedia and make this information available on the semantic Web as linked open data. The DBpedia triple stores then allow anyone to solve sophisticated queries against Wikipedia extracted data, and to link the different data sets on these data. The French chapter of DBpedia was created and deployed by Wimmics and is now an online running platform providing data to several projects such as: QAKIS, Izipedia, zone47, Sépage, HdA Lab., JocondeLab, etc.
• Release Contributions: The new release is based on updated Wikipedia dumps and the inclusion of the DBpedia history extraction of the pages.
• URL: http://wiki.dbpedia.org/
• Contact: Fabien Gandon
• Participants: Fabien Gandon, Elmahdi Korfed

6.1.3 Fuzzy labelling argumentation module

• Name: Fuzzy labelling algorithm for abstract argumentation
• Keywords: Artificial intelligence, Multi-agent, Knowledge representation, Algorithm
• Functional Description: The goal of the algorithm is to compute the fuzzy acceptability degree of a set of arguments in an abstract argumentation framework. The acceptability degree is computed from the trustworthiness associated with the sources of the arguments.
• Contact: Serena Villata
• Participant: Serena Villata

6.1.4 Qakis

• Name: Question-Answering wiki framework based system
• Keyword: Natural language
• Functional Description: The QAKiS system implements question answering over DBpedia. QAKiS allows end users to submit a query to an RDF triple store in English and to obtain the answer in the same language, hiding the complexity of the non-intuitive formal query languages involved in the resolution process. At the same time, the expressiveness of these standards is exploited to scale to the huge amounts of available semantic data. Its major novelty is to implement a relation-based match for question interpretation, to convert the user question into a query language (e.g. SPARQL). English, French and German DBpedia chapters are the RDF data sets to be queried using a natural language interface.
• URL: http://www.qakis.org/
• Contact: Elena Cabrio
• Participants: Alessio Palmero Aprosio, Amine Hallili, Elena Cabrio, Fabien Gandon, Julien Cojan, Serena Villata

6.1.5 Corese Server

• Name: Corese Server
• Keywords: Semantic Web, RDF, SPARQL
• Scientific Description: A Web server to interact with Corese via HTTP SPARQL endpoint, STTL display engine
• Contact: Olivier Corby
• Participants: Alban Gaignard, Fuqi Song, Olivier Corby
• Partner: I3S

6.1.6 CREEP semantic technology

• Keywords: Natural language processing, Machine learning, Artificial intelligence
• Scientific Description: The software provides a modular architecture specifically tailored at the classification of cyberbullying and offensive content on social media platforms. The system can use a variety of features (ngrams, different word embeddings, etc) and all the netwok parameters (number of hidden layers, dropout, etc) can be altered by using a configuration file.
• Functional Description: The software uses machine learning techniques to classify cyberbullying instances in social media interactions.
• Release Contributions: +Attention mechanism +Hyperparameters for emoji in config file +Predictions output +Streamlined labeling of arbitrary files
• Publications: hal-01906096v1, hal-01920266v1
• Contact: Michele Corazza
• Participants: Michele Corazza, Elena Cabrio, Serena Villata

6.1.7 Licentia

• Keywords: Right, License
• Scientific Description: In order to ensure the high quality of the data published on the Web of Data, part of the self-description of the data should consist in the licensing terms which specify the admitted use and re-use of the data by third parties. This issue is relevant both for data publication as underlined in the “Linked Data Cookbook” where it is required to specify an appropriate license for the data, and for the open data publication as expressing the constraints on the reuse of the data would encourage the publication of more open data. The main problem is that data producers and publishers often do not have extensive knowledge about the existing licenses, and the legal terminology used to express the terms of data use and reuse. To address this open issue, we present Licentia, a suite of services to support data producers and publishers in data licensing by means of a user-friendly interface that masks to the user the complexity of the legal reasoning process. In particular, Licentia offers two services: i) the user selects among a pre-defined list those terms of use and reuse (i.e., permissions, prohibitions, and obligations) she would assign to the data and the system returns the set of licenses meeting (some of) the selected requirements together with the machine readable licenses’ specifications, and ii) the user selects a license and she can verify whether a certain action is allowed on the data released under such license. Licentia relies on the dataset of machine-readable licenses (RDF, Turtle syntax, ODRL vocabulary and Creative Commons vocabulary) available at http://datahub.io/dataset/rdflicense. We rely on the deontic logic presented by Governatori et al. to address the problem of verifying the compatibility of the licensing terms in order to find the license compatible with the constraints selected by the user. The need for licensing compatibility checking is high, as shown by other similar services (e.g., Licensius5 or Creative Commons Choose service6 ). However, the advantage of Licentia with respect to these services is twofold: first, in these services compatibility is pre-calculated among a pre-defined and small set of licenses, while in Licentia compatibility is computed at runtime and we consider more than 50 heterogeneous licenses, second, Licentia provides a further service that is not considered by the others, i.e., it allows to select a license from our dataset and verify whether some selected actions are compatible with such license.
• Functional Description:

  Licentia is a web service application with the aim to support users in licensing data. Our goal is to provide a full suite of services to help in the process of choosing the most suitable license depending on the data to be licensed.

  The core technology used in our services is powered by the SPINdle Reasoner and the use of Defeasible Deontic Logic to reason over the licenses and conditions.

  The dataset of RDF licenses we use in Licentia is the RDF licenses dataset where the Creative Commons Vocabulary and Open Digital Rights Language (ODRL) Ontology are used to express the licenses.

• URL: http://licentia.inria.fr/
• Contact: Serena Villata
• Participant: Cristian Cardellino

6.1.8 lod2quiz

• Keywords: Semantic Web, Artificial intelligence, Web Application, E-learning
• Functional Description: Automatic quiz generator
• Release Contributions: Contains the core engine of lod2quiz, deployed as a web application that exposes a REST API with methods for the quiz generation.
• Publications: hal-01688798, hal-01811490, hal-01758737
• Contact: Oscar Rodriguez Rocha
• Participants: Oscar Rodriguez Rocha, Catherine Faron

6.1.9 SPARQL micro-services

• Name: SPARQL micro-services
• Keywords: Web API, SPARQL, Microservices, LOD - Linked open data, Data integration
• Functional Description: The approach leverages the micro-service architectural principles to define the SPARQL Micro-Service architecture, aimed at querying Web APIs using SPARQL. A SPARQL micro-service is a lightweight SPARQL endpoint that typically provides access to a small, resource-centric graph. Furthermore, this architecture can be used to dynamically assign dereferenceable URIs to Web API resources that do not have URIs beforehand, thus literally “bringing” Web APIs into the Web of Data. The implementation supports a large scope of JSON-based Web APIs, may they be RESTful or not.
• URL: https://github.com/frmichel/sparql-micro-service
• Publications: hal-02060966, hal-01722792, hal-01947589, hal-02168164
• Author: Franck Michel
• Contact: Franck Michel

6.1.10 ACTA

• Name: A Tool for Argumentative Clinical Trial Analysis
• Keywords: Artificial intelligence, Natural language processing, Argument mining
• Functional Description: Argumentative analysis of textual documents of various nature (e.g., persuasive essays, online discussion blogs, scientific articles) allows to detect the main argumentative components (i.e., premises and claims) present in the text and to predict whether these components are connected to each other by argumentative relations (e.g., support and attack), leading to the identification of (possibly complex) argumentative structures. Given the importance of argument-based decision making in medicine, ACTA is a tool for automating the argumentative analysis of clinical trials. The tool is designed to support doctors and clinicians in identifying the document(s) of interest about a certain disease, and in analyzing the main argumentative content and PICO elements.
• URL: http://ns.inria.fr/acta/
• Contact: Serena Villata

6.1.11 WebAudio tube guitar amp sims CLEAN, DISTO and METAL MACHINEs

• Name: Tube guitar amplifier simulators for Web Browser : CLEAN MACHINE, DISTO MACHINE and METAL MACHINE
• Keyword: Tube guitar amplifier simulator for web browser
• Scientific Description: This software is one of the only ones of its kind to work in a web browser. It uses "white box" simulation techniques combined with perceptual approximation methods to provide a quality of guitar playing in hand comparable to the best existing software in the native world.
• Functional Description: Software programs for creating real-time simulations of tube guitar amplifiers that behave most faithfully like real hardware amplifiers, and run in a web browser. In addition, the generated simulations can run within web-based digital audio workstations as plug-ins. The "CLEAN MACHINE" version specializes in the simulation of acoustic guitars when playing electric guitars. The DISTO machine specializes in classic rock tube amp simulations, and METAL MACHINE targets metal amp simulations. These programs are one of the results of the ANR WASABI project.
• Release Contributions: First stable version, delivered and integrated into the ampedstudio.com software. Two versions have been delivered: a limited free version and a commercial one.
• News of the Year: Best paper at WebAudio Conference 2020.
• Publications: hal-01721463, hal-01893681, hal-02337828, hal-03087768, hal-01721483, hal-01735478, hal-02366725, hal-02557901, hal-01589330, hal-03087763, hal-01893660, hal-01589229
• Contact: Michel Buffa
• Participant: Michel Buffa
• Partner: Amp Track Ltd, Finland

6.1.12 Morph-xR2RML

• Name: Morph-xR2RML
• Keywords: RDF, Semantic Web, LOD - Linked open data, MongoDB, SPARQL
• Functional Description:

  The xR2RML mapping language that enables the description of mappings from relational or non relational databases to RDF. It is an extension of R2RML and RML.

  Morph-xR2RML is an implementation of the xR2RML mapping language, targeted to translate data from the MongoDB database, as well as relational databases (MySQL, PostgreSQL, MonetDB). Two running modes are available: (1) the graph materialization mode creates all possible RDF triples at once, (2) the query rewriting mode translates a SPARQL 1.0 query into a target database query and returns a SPARQL answer. It can run as a SPARQL endpoint or as a stand-alone application.

  Morph-xR2RML was developed by the I3S laboratory as an extension of the Morph-RDB project which is an implementation of R2RML.

• URL: https://github.com/frmichel/morph-xr2rml/
• Publications: hal-01207828, hal-01330146, hal-01280951
• Author: Franck Michel
• Contact: Franck Michel

6.1.13 ARViz

• Name: Association Rules Visualization
• Keyword: Information visualization
• Functional Description: ARViz supports the exploration of thematic attributes describing association rules (e.g. confidence, interestingness, and symmetry) through a set of interactive, synchronized, and complementary visualisation techniques (i.e. a chord diagram, an association graph, and a scatter plot). Furthermore, the interface allows the user to recover the scientific publications related to rules of interest.
• Release Contributions: Visualization of association rules within the scientific literature of COVID-19.
• URL: http://covid19.i3s.unice.fr:8080/arviz/
• Contact: Marco Antonio Alba Winckler

6.1.14 MGExplorer

• Name: Multivariate Graph Explorer
• Keyword: Information visualization
• Functional Description: MGExplorer is an information visualization toolsuit that integrates many information visualization techniques aimed at supporting the exploration of multivariate graphs. MGExplorer allows users to choose and combine the information visualization techniques creating a graph that describes the exploratory path of dataset.
• Release Contributions: Visualization of data extracted from linked data datasets.
• URL: http://covid19.i3s.unice.fr:8080/
• Contact: Marco Antonio Alba Winckler
• Partner: Universidade Federal do Rio Grande do Sul

7.1.1 LinkedDataViz and MGExplorer

Participants: Marco Winckler, Aline Menin, Olivier Corby, Alain Giboin, Fabien Gandon.

Visualization techniques are useful tools to explore datasets, enabling the discovery of meaningful patterns and causal relationships. Nonetheless, the discovery process is often exploratory and requires multiple views to support analyzing different or complementary perspectives to the data. The analytical reasoning that guides the exploration processes based on multiple views can be represented by provenance between views. In this paper 71, we introduce the term ancillary search tasks to characterize multiple complementary search tasks (possibly run in parallel) that help users to achieve a complex search task. This concept has been extended to support chained views to describe the incremental exploration of large, multidimensional datasets through the combination of multiple chained visualization techniques and visual querying, and the representation of analytical provenance through a visual representation of the dependencies between views. As a proof-of-concept, we developed a visualization tool MGExplorer, which encompasses a sample of five visualization techniques (Node-Edge Diagram,ClusterVis, GlyphMatrix, Histogram, and IRIS) that are used to explore multivariate graphs datasets. Each view in MGExplorer supports visual querying techniques that enable the definition of subsets of the current dataset to be explored in another, chained view.

Linked Data Viz is a platform to provide graphic views of Linked Data. The platform is now generic in the sense that it can query any SPARQL endpoint. A specific service has been designed in order to submit a SPARQL query and the URL of a SPARQL endpoint. The outcomes of Linked Data Viz are used as entry point for visualizations created by the tool MGExplorer.

LinkedDataViz web site: http://covid19.i3s.unice.fr:8080/ldviz.

7.1.2 Visualization of geospatial Linked Data

Participants: Franck Michel, Marco Winckler, Olivier Corby.

By means of an Ubinet Master internship, we initiated a work meant to explore the cross-fertilization of geospatial data visualization and reasoning on linked data. This nascent work sheds light on interesting leads that we intend to push further.

7.2.1 Autonomous agents in a social and ubiquitous Web

Participants: Andrei Ciortea, Olivier Corby, Fabien Gandon, Franck Michel.

Recent W3C recommendations for the Web of Things (WoT) and the Social Web are turning hypermedia into a homogeneous information fabric that interconnects heterogeneous resources: devices, people, information resources, abstract concepts, etc. The integration of multi-agent systems with such hypermedia environments now provides a means to distribute autonomous behavior in worldwide pervasive systems. A central problem then is to enable autonomous agents to discover heterogeneous resources in world wide and dynamic hypermedia environments. This is a problem in particular in WoT environments that rely on open standards and evolve rapidly—thus requiring agents to adapt their behavior at runtime in pursuit of their design objectives. To this end, we developed a hypermedia search engine for the WoT that allows autonomous agents to perform approximate search queries in order to retrieve relevant resources in their environment in (weak) real time. The search engine crawls dynamic WoT environments to discover and index device metadata described with the W3C WoT Thing Description, and exposes a SPARQL endpoint that agents can use for approximate search. To demonstrate the feasibility of our approach, we implemented a prototype application for the maintenance of industrial robots in worldwide manufacturing systems. The prototype demonstrates that our semantic hypermedia search engine enhances the flexibility and agility of autonomous agents in a social and ubiquitous Web 9.

7.2.2 Multilingual Hate Speech Detection

Participants: Elena Cabrio, Serena Villata, Michele Corazza.

The increasing popularity of social media platforms like Twitter and Facebook has led to a rise in the presence of hate and aggressive speech on these platforms. Despite the number of approaches recently proposed in the Natural Language Processing research area for detecting these forms of abusive language, the issue of identifying hate speech at scale is still an unsolved problem. In this research activity, together with Sara Tonelli (FBK Trento) and Stefano Menini (FBK Trento), we have proposed a robust recurrent neural architecture which is shown to perform in a satisfactory way across different languages, namely English, Italian and German. We address an extensive analysis of the obtained experimental results over the three languages to gain a better understanding of the contribution of the different components employed in the system, both from the architecture point of view (i.e., Long Short Term Memory, Gated Recurrent Unit, and bidirectional Long Short Term Memory) and from the feature selection point of view (i.e., social network specific features, emotion lexica, emojis, embeddings). To address such in-depth analysis, we use three freely available datasets for hate speech detection on social media on English, Italian and German 10.

7.2.3 Supporting Fake News Identification through Stance Detection

Participants: Elena Cabrio, Serena Villata, Jérôme Delobelle.

This work is part of the DGA project RAPID CONFIRMA (COntre argumentation contre les Fausses InfoRMAtion) aiming to automatically detect fake news and limit their diffusion. In our work, we present a concrete application scenario where a fake news detection system is empowered with an argument mining model, to highlight and aid the analysis of the arguments put forward to support or oppose a given target topic in articles containing fake information 41. More precisely, we propose to extend a disinformation analysis tool with a stance detection module for arguments relying on pretrained language models, i.e., BERT, with the aim of obtaining a more effective analysis tool both for users and analysts. To evaluate the argument stance detection module in the disinformation context, we propose to annotate a new resource of fake news articles, where arguments are classified as being InFavor or Against towards a target topic. Our new annotated data set contains sentences about three topics currently attracting a lot of fake news around them, i.e., public health demands vaccination, white helmets provide essential services, and the risible impact of Covid-19. This data set collects 86 articles containing nearly 3000 sentences.

7.2.4 Aspect-based Sentiment Analysis in Polarized Contexts

Participants: Vorakit Vorakitphan, Elena Cabrio, Serena Villata.

Aspect-based Sentiment Analysis (ABSA) aims at capturing sentiment (i.e., positive, negative or neutral) expressed toward each aspect (i.e., attribute) of a target entity. The main interest is to capture sentiment nuances about different entities. However, in a context of opinion polarization, different groups of people can form strong convictions of competing opinions on such target entities, resulting in different (often opposite) evaluations of the same aspect. Compare, for example, the differences in the pro- and anti-Brexit discourses concerning the withdrawal of the United Kingdom from the European Union, aligning with contrasting attitudes toward the EU, the immigration and the country’s culture. Whilst in standard scenarios of sentiment analysis about specific entities and their aspects it is assumed that sentiment is consistent (e.g., a big screen is a desirable characteristics for a TV), this is not the case for polarized contexts. Hence, for example, a "clean Brexit" might be desirable to some, but not to others. Together with Marco Guerini (FBK Trento), we proposed a comprehensive framework for studying the interaction of ABSA with opinion polarization in newspapers and social media 38. We first trained a machine learning algorithm that detects the emotions and their intensities at sentence-level, and then we mapped emotion intensities into the Valence, Arousal, and Dominance (VAD) model. Later, we built a framework to assess whether and how VAD are connected to polarized contexts, by computing the VAD scores of a set of key-concepts that can be found on newspapers with opposite views. These key-concepts (e.g., "stop immigration") are built from a set of aspects ("immigration" in our example) combined with relevant verbs or adjectives that represent a clear polarized opinion toward the aspect (e.g., "stop"). To experiment with the proposed approach, we focussed on the Brexit scenario as it provided us with the required elements to carry out our study, because of the opinion divisions formed around one or more political positions or issues. In our experimental setting, we selected two British newspapers known to be polarized, i.e., either for or against Brexit. Results show that VAD are not absolute, but relative to the newspaper’s viewpoint on the key-concept. Our approach highlights that using the proposed key-concepts gives us fine-grained details about VAD elements that strongly interact with the polarized context. We showed that standard SA approaches can be deceptive in such polarized setting (considering only the word "Brexit" on both newspapers, the valence is almost identical), while our ABSA approach showed a clear-cut polarization.

7.2.5 Fuzzy Polarity Propagation for Multi-Domain Sentiment Analysis

Participants: Andrea Tettamanzi.

Together with Claude Pasquier and Célia da Costa Pereira of the I3S Laboratory, we studied how different domain-dependent polarities can be learned for the same concepts, in the context of multi-domain sentiment analysis. To this aim, we extend an existing approach based on the propagation of fuzzy polarities over a semantic graph capturing background linguistic knowledge to learn concept polarities with respect to various domains and their uncertainty from labeled datasets. In particular, we use POS tagging to refine the association between terms and concepts and word embedding to enhance the construction of the semantic graph. The proposed approach 34 was then evaluated on a standard benchmark, showing that the combined use of POS tagging and word embedding improves its performance. One particularly strong point of the proposed approach is its recall, which is always very close to 100%. In addition, it exhibits good cross-domain generalization capabilities.

7.2.6 Linking interactive WebAudio applications to the WASABI knowledge base

Participants: Michel Buffa.

7.2.7 Using Agent-Based Modeling to explore the role of socio-environmental interactions on Ancient Settlement Dynamics

Participants: Andrea Tettamanzi.

Within the framework of a mult-disciplinary project involving archaeologists, economists, geographers, and computer scientists, we used Agent-Based Modelling to explore the respective impacts of environmental and social factors on the settlement pattern and dynamics during the Roman period in South-Eastern France 52.

7.3.1 Publication of the Covid-on-the-Web dataset

Participants: Franck Michel, Fabien Gandon, Valentin Ah-Kane, Anna Bobasheva, Elena Cabrio, Olivier Corby, Raphaël Gazzotti, Alain Giboin, Santiago Marro, Tobias Mayer, Serena Villata, Marco Winckler.

The Covid-on-the-Web project aims to allow biomedical researchers to access, query and make sense of COVID-19 related literature. Launched in Mars 2020, it involved multiple skills of the team in knowledge representation, text, data and argument mining, as well as data visualization and exploration. Among the achievements of the projetct is the Covid-on-the-Web RDF dataset 48 that we genetared and published by processing, analyzing and enriching the “COVID-19 Open Research Dataset” (CORD-19) that gathers 100K+ full-text scientific articles related to the coronaviruses. The dataset produced comprises two main knowledge graphs: (1) named entities mentioned in the CORD-19 corpus and linked to DBpedia, Wikidata and other BioPortal vocabularies, and (2) arguments extracted using ACTA, a tool automating the extraction and visualization of argumentative graphs, meant to help clinicians analyze clinical trials and make decisions.

Web site https://github.com/Wimmics/CovidOnTheWeb.

7.3.2 Mining the Covid-on-the-Web Data

Participants: Lucie Cadorel, Andrea Tettamanzi.

As soon as the Covid-on-theWeb RDF dataset was published, we set out to exploit it to mine interesting associations from it. We thus proposed a method to discover interesting association rules from an RDF knowledge graph, by combining clustering, community detection, and dimensionality reduction, as well as criteria for filtering the discovered association rules in order to keep only the most interesting rules 21. Our results demonstrate the effectiveness and scalability of the proposed method and suggest several possible uses of the discovered rules, including (i) curating the knowledge graph by detecting errors, (ii) finding relevant and coherent collections of scientific articles, and (iii) suggesting novel hypotheses to biomedical researchers for further investigation.

7.3.3 Publication of the WASABI dataset

Participants: Franck Michel, Fabien Gandon, Elena Cabrio, Alain Giboin, Marco Winckler, Maroua Tikat, Michael Fell.

Since 2017, a two-million song database consisting of metadata collected from multiple open data sources and automatically extracted information has been constructed in the context of the WASABI project. The goal is to build a knowledge graph linking collected metadata (artists, discography, producers, dates, etc.) with metadata generated by the analysis of both the songs' lyrics (topics, places, emotions, structure, etc.) and audio signal (chords, sound, etc.). It relies on natural language processing and machine learning methods for extraction, and semantic Web frameworks for integration. The dataset describes more than 2 millions commercial songs, 200K albums and 77K artists. It can be exploited by music search engines, music professionals or scientists willing to analyze popular music published since 1950. It is available under an open license in multiple formats and is accompanied by online applications and open source software including an interactive navigator, a REST API and a SPARQL endpoint.

Web site https://github.com/micbuffa/WasabiDataset.

7.3.4 Semantic Web for Biodiversity

Participants: Franck Michel, Catherine Faron.

This activity addresses the challenges of exploiting knowledge representation and semantic web technologies to enable data sharing and integration in the biodiversity area. The collaboration with the ”Muséum National d'Histoire Naturelle” of Paris (MNHN) goes on along two main axes.

First, in 2019 the MNHN started using our SPARQL Micro-Services architecture and framework to help biologists in editing taxonomic information by confronting multiple, heterogeneous data sources  70. In 2020 this collaboration has been strengthened and the MNHN now heavily relies on those services for daily activities.

Second, we have kept on the work initiated whithin the Bioschemas.org W3C community group that seeks the definition and adoption of common biology-related markup terms. While a new term TaxonName was defined and we updated MNHN webpages accordingly, we have undertaken an "evangelization" action to promote this practice in the biodiversity community 30.

7.3.5 Enriching the WASABI Song Corpus with Lyrics Annotations.

Participants: Elena Cabrio, Michael Fell, Michel Buffa.

The WASABI Song Corpus is a large corpus of songs enriched with metadata extracted from music databases on the Web, and resulting from the processing of song lyrics and from audio analysis. Given that lyrics encode an important part of the semantics of a song, we have focused on the design and application of methods to extract relevant information from the lyrics, such as their structure segmentation, their topics, the explicitness of the lyrics content, the salient passages of a song and the emotions conveyed. So far, the corpus contains 1.73M songs with lyrics (1.41M unique lyrics) annotated at different levels with the output of the above mentioned methods. Such corpus labels and the provided methods can be exploited by music search engines and music professionals (e.g. journalists, radio presenters) to better handle large collections of lyrics, allowing an intelligent browsing, categorization and recommendation of songs.

7.3.6 Ontology alignment in the sourcing domain

Participants: Molka Dhouib, Catherine Faron, Andrea Tettamanzi.

7.3.7 A feature-based comparative analysis of legal ontologies

Participants: Serena Villata.

Ontologies represent the standard way to model the knowledge about specific domains. This holds also for the legal domain where several ontologies have been put forward to model specific kinds of legal knowledge. Both for standard users and for law scholars, it is often difficult to have an overall view on the existing alternatives, their main features and their interlinking with the other ontologies. To answer this need, in this work, we address an analysis of the state-of-the-art in legal ontologies and we characterise them along with some distinctive features. This work aims to guide generic users and law experts in selecting the legal ontology that better fits their needs and in understanding its specificity so that proper extensions to the selected model could be investigated 13.

7.4.1 Uncertainty Evaluation for Linked Data

Participant: Ahmed Elamine Djebri, Fabien Gandon, Andrea Tettamanzi.

7.4.2 Leveraging Data with Uncertain Labels for Machine Learning

Participants: Andrea Tettamanzi.

Prompted by an application in the area of human geography using machine learning to study housing market valuation based on the urban form, we proposed a method based on possibility theory to deal with sparse data, which can be combined with any machine learning method to approach weakly supervised learning problems 54. More specifically, the solution we propose constructs a possibilistic loss function to account for an uncertain supervisory signal. Although the proposal was motivated by a specific application, its basic principles are general. The proposed method has then been empirically validated on real-world data.

7.4.3 SPARQL Function: LDScript

Participant: Olivier Corby.

We have continued the implementation and validation of LDScript, Linked Data Script, a programming language compatible with SPARQL that enables users to write extension functions that are directly executable in SPARQL queries. LDScript is an extension of SPARQL Filter language with function definition, variable declaration, iteration, second order and anonymous function, pattern matching. It provides users with extension datatypes that enables them to manage Semantic Web objects such as RDF triple and graph as well as SPARQL Query result. In addition, extension datatypes provide implementations for list, hashmap, XML document and JSON object. A SHACL interpreter has been entirely written using LDScript.

Linked Data Script documentation: https://files.inria.fr/corese/doc/ldscript.html

7.4.4 Linked Data Access and Event Driven programming

Participant: Olivier Corby.

We started a preliminary work on an access control model for RDF graphs. It is possible to specify access rights at the scale of nodes and predicates URIs or namespaces. Preliminary report: https://files.inria.fr/corese/doc/access.html.

We started a preliminary work on a safety model to protect a SPARQL endpoint where functions can be protected (forbidden) e.g. Linked Functions. Preliminary report: https://files.inria.fr/corese/doc/level.html

We generalized the Event Driven programming model for the HTTP server (SPARQL endpoint), SPARQL Update, SHACL interpreter, Rule engine and Transformation engine.

Linked Data Event Driven Programming documentation: https://files.inria.fr/corese/doc/event.html.

7.4.5 Linked Data Crawling

Participant: Fabien Gandon, Hai Huang.

A Linked Data crawler performs a selection to focus on collecting linked RDF (including RDFa) data on the Web. From the perspectives of throughput and coverage, given a newly discovered and targeted URI, the key issue of Linked Data crawlers is to decide whether this URI is likely to dereference into an RDF data source and therefore if it is worthy downloading the representation it points to. Current solutions adopt heuristic rules to filter irrelevant URIs. But when the heuristics are too restrictive this hampers the coverage of crawling. We proposed and compared approaches to learn strategies for crawling Linked Data on the Web by predicting whether a newly discovered URI will lead to an RDF data source or not. We detailed the features used in predicting the relevance and the methods we evaluated including a promising adaptation of FTRL-proximal online learning algorithm. We compared several options through extensive experiments including existing crawlers as baseline methods to evaluate their efficiency 26.

7.4.6 Semantic Overlay Network for Linked Data Access

Participant: Fabien Gandon, Mahamadou Toure.

We proposed and evaluated MoRAI (Mobile Read Access in Intermittent internet connectivity), a distributed peer-to-peer architecture organized in three levels dedicated to RDF data exchanges by mobile contributors. We presented the conceptual and technical aspects of this architecture as well as a theoretical analysis of the different characteristics. We then evaluated it experimentally and results show the relevance of considering geographical positions during data exchanges and of integrating RDF graph replication to ensure data availability in terms of requests completion rate and resistance to crash scenarios 37.

7.4.7 SHACL Extension

Participant: Olivier Corby, Iliana Petrova, Fabien Gandon.

In the context of a collaboration with Stanford University, we have been working on extensions of W3C SHACL Shape Constraint Language 1.

We have proposed extensions of SHACL path language with a xsh:predicatePath operator that enables the interpreter to navigate from a node in the RDF graph to the set of predicates the node is subject, object or both. In addition, we propose to extend the path language to navigate from nodes to triples and back with two operators: xsh:triplePath and xsh:nodePath. The path language is also extended with xsh:exist and xsh:filter that enable the interpreter to check conditions.

SHACL shape constraints are extended with a xsh:function statement that enable the user to specify constraints using LDScript functions. Additional detailed validation results can be obtained for node and boolean constraints.

Linked Data SHACL Extension documentation: https://files.inria.fr/corese/doc/shacl.html

7.4.8 Injection of Knowledge in a Sourcing Recommender System

Participants: Molka Dhouib, Catherine Faron, Andrea Tettamanzi.

7.4.9 Identifying argumentative structures in clinical trials

Participants: Elena Cabrio, Serena Villata, Tobias Mayer, Santiago Marro.

We annotated first a dataset of 159 abstracts of Randomized Controlled Trials (RCTs) from the MEDLINE database, comprising 4 different diseases (i.e., glaucoma, hypertension, hepatitis b, diabetes), then a larger dataset of 500 abstracts on the neoplasm disease, leading to a dataset of 4113 argument components and 2601 argument relations. We then proposed a complete argument mining pipeline for RCTs, classifying argument components as evidence and claims, and predicting the relation, i.e., attack or support, holding between those argument components 45. We experimented with deep bidirectional transformers in combination with different neural architectures (i.e., LSTM, GRU and CRF) and outperformed current state-of-the-art end-to-end argument mining systems. In addition, we also included the identification of PICO elements in the abstracts (PICO is a framework to answer health-care related questions in evidence-based practice. Elements comprise patients/population (P), intervention (I), control/comparison (C) and outcome (O) information). We finally investigated the robustness of language models like BERT for the argument classification task 47.

7.4.10 Relation Prediction in Argument Mining

Participants: Elena Cabrio, Serena Villata.

Argument(ation) Mining (AM) is the research area which aims at extracting argument components and predicting argumentative relations (i.e., support and attack) from text. In particular, numerous approaches have been proposed in the literature to predict the relations holding between arguments and application-specific annotated resources were built for this purpose. Despite the fact that these resources were created to experiment on the same task, the definition of a single relation prediction method to be successfully applied to a significant portion of these datasets is an open research problem in AM. This means that none of the methods proposed in the literature can be easily ported from one resource to another. Together with Oana Cocarascu and Francesca Toni from Imperial College London (UK), we addressed this problem by proposing a set of dataset independent strong neural baselines which obtain homogeneous results on all the datasets proposed in the literature for the argumentative relation prediction task in AM 22. Thus, our baselines can be employed by the AM community to compare more effectively how well a method performs on the argumentative relation prediction task.

7.4.11 Injection of Automatically Selected DBpedia Subjects in Electronic Medical Records to boost Hospitalization Prediction

Participants: Catherine Faron, Fabien Gandon, Raphaël Gazzotti.

Although there are many medical standard vocabularies available, it remains challenging to properly identify domain concepts in electronic medical records. Variations in the annotations of these texts in terms of coverage and abstraction may be due to the chosen annotation methods and the knowledge graphs, and may lead to very different performances in the automated processing of these annotations. We proposed a semi-supervised approach based on DBpedia to extract medical subjects from EMRs and evaluate the impact of augmenting the features used to represent EMRs with these subjects in the task of predicting hospitalization. We compared the impact of subjects selected by experts vs. by machine learning methods through feature selection. Our approach was experimented on data from the database PRIMEGE PACA that contains more than 600,000 consultations carried out by 17 general practitioners (GPs) 25.

7.4.12 A Knowledge Graph Enhanced Learner Model to Predict Outcomes to Questions

Participants: Antonia Ettorre, Catherine Faron, Fabien Gandon, Mathis Le Quiniou, Franck Michel, Oscar Rocha Rodriguez, Yuting Sun.

7.4.13 Machine Learning for Operations Research

Participants: Andrea Tettamanzi.

Together with Alberto Ceselli and Saverio Basso of the University of Milan we used machine learning techniques to understand good decompositions of linear programming problems 6.

7.4.14 RDF Mining

Participants: Thu Huong Nguyen, Andrea Tettamanzi.

In the framework of Nguyen Thu Huong's thesis, we have continued to explore the use of grammar-based evolutionary method to mine RDF datasets for OWL class disjointness axioms. In particular, we addressed the problem of discovering disjointness axioms involving complex class expressions 32, 33. As it turns out, this problem involves at least two conflicting criteria an axiom should meet, namely possibility (i.e., truth, acceptability, likelihood), and generality. This prompted us to adapt our evolutionary approach to suit multi-objective optimization 31.

On the other hand, our evolutionary approach critically rely on (candidate) axiom scoring. In practice, testing an axiom boils down to computing an acceptability score, measuring the extent to which the axiom is compatible with the recorded facts. Methods to approximate the semantics of given types of axioms have been thoroughly investigated in the last decade, but a promising alternative to their direct computation is to train a surrogate model on a sample of candidate axioms for which the score is already available, to learn to predict the score of a novel, unseen candidate axiom. Together with Dario Malchiodi of the University of Milan and Célia da Costa Pereira of the I3S Laboratory, we assess the role of similarity measures and learning methods in classifying candidate axioms for automated schema induction through kernel-based learning algorithms. The evaluation was based on three different similarity measures between axioms and two alternative dimensionality reduction techniques to check the extent to which the considered similarities allow to separate true axioms from false axioms. The result of the dimensionality reduction process is subsequently fed to several learning algorithms, comparing the accuracy of all combinations of similarity, dimensionality reduction technique, and classification method. As a result, it is observed that it is not necessary to use sophisticated semantics-based similarity measures to obtain accurate predictions, and furthermore that classification performance only marginally depends on the choice of the learning method. Our results open the way to implementing efficient surrogate models for axiom scoring to speed up ontology learning and schema induction methods 29.

PREMISSE Collaborative Project

Participants: Molka Dhouib, Catherine Faron, Andrea Tettamanzi.

This collaborative project with the SILEX France company started in March 2017, funded by the ANRT (CIFRE PhD). SILEX France is developing a B2B platform where service providers and consumers upload their service offers or requests in free natural language; the platform is intended to recommend service providers to the applicant, which are likely to fit his/her service request. The aim of this project is to propose a decision support system by exploiting the semantic knowledge that are extracted from the textual descriptions of requests for services and providers, in order to recommend relevant providers for a service request.

HealthPredict Collaborative Project

Participants: Raphaël Gazzotti, Catherine Faron, Fabien Gandon.

This collaborative project with the Synchronext company started in April 2017, funded by the ANRT (CIFRE PhD). Synchronext is a startup aiming at developing Semantic Web business solutions. The aim of this project is to design a digital health solution for the early management of patients through consultations with their general practitioner and health care circuit. The goal is to develop a predictive Artificial Intelligence interface that allows to cross the data of symptoms, diagnosis and medical treatments of the population in real time to predict the hospitalization of a patient. We presented at SAC 2020 25 a semi-supervised approach based on DBpedia to select from electronic medical records subjects designating medical aspects relevant to the prediction of hospitalization.

Curiosity Collaborative Project

Participants: Catherine Faron, Oscar Rodríguez Rocha.

This collaborative project with the TeachOnMars company started in October 2019. TeachOnMars is developping a platform for mobile learning. The aim of this project is to develop an approach for automatically indexing and semantically annotating heterogeneous pedagogical resources from different sources to build up a knowledge graph enabling to compute training paths, that correspond to the learner's needs and learning objectives.

CIFRE Contract with Doriane

Participants: Andrea Tettamanzi, Rony Dupuy Charles.

This collaborative contract for the supervision of a CIFRE doctoral scholarship, relevant to the PhD of Rony Duput Charles, is part of Doriane's Fluidity Project (Generalized Experiment Management), the feasibility phase of which has been approved by the Terralia cluster and financed by the Région Sud-Provence Alpes Côte d'Azur and BPI France in March 2019. The objective of the thesis is to develop machine learning methods for the field of agro-vegetation-environment. To do so, this research work will take into account and address the specificities of the problem, i.e. data with mainly numerical characteristics, scalability of the study object, small data, availability of codified background knowledge, need to take into account the economic stakes of decisions, etc., as explained in the section on the context of the project. To enable the exploitation of ontological resources, the combination of symbolic and connective approaches will be studied, among others. Such resources can be used, on the one hand, to enrich the available datasets and, on the other hand, to restrict the search space of predictive models and better target learning methods.

The PhD student will develop original methods for the integration of background knowledge in the process of building predictive models and for the explicit consideration of uncertainty in the field of agro-plant environment.

CIFRE Contract with Kinaxia

Participants: Andrea Tettamanzi, Lucie Cadorel.

This thesis project is part of a collaboration with Kinaxia that began in 2017 with the Incertimmo project. The main theme of this project was the consideration of uncertainty for a spatial modeling of real estate values in the city. It involved the computer scientists of the Laboratory and the geographers of the ESPACE Laboratory. It allowed the development of an innovative methodological protocol to create a mapping of real estate values in the city, integrating fine-grained spatiality (the street section), a rigorous treatment of the uncertainty of knowledge, and the fusion of multi-source (with varying degrees of reliability) and multi-scale (parcel, street, neighbourhood) data.

This protocol was applied to the Nice-Côte d'Azur metropolitan area case study, serving as a test bed for application to other metropolitan areas.

The objective of this thesis, which will be carried out by Lucie Cadorel with the advice of Andrea Tettamanzi, is, on the one hand, to study and adapt the application of methods for extracting knowledge from texts (or text mining) to the specific case of real estate ads written in French, before extending them to other languages, and, on the other hand, to develop a methodological framework that makes it possible to detect, explicitly qualify, quantify and, if possible, reduce the uncertainty of the extracted information, in order to make it possible to use it in a processing chain that is finalized for recommendation or decision making, while guaranteeing the reliability of the results.

9.1.1 Inria associate team not involved in an IIL

9.1.2 Participation in other international programs

9.2.1 Visits of international scientists

Andrei Ciortea, researcher at University of St. Gallen, visited Wimmics in September to work on RDF PubSub, security in CORESE and multi-agent systems on the Web.

9.3.1 FP7 & H2020 Projects

9.3.2 Collaborations in European programs, except FP7 and H2020

PIA GDN ANSWER

Participants: Fabien Gandon, Hai Huang, Vorakit Vorakitphan, Serena Villata, Elena Cabrio.

ANSWER stands for Advanced aNd Secured Web Experience and seaRch 2. It is a GDN project (Grands Défis du Numérique) from the PIA program (Programme d’Investissements d'Avenir) on Big Data. The project is between four Inria research teams and the Qwant company.

The aim of the ANSWER project is to develop the new version of the Qwant 3 search engine by introducing radical innovations in terms of search criteria as well as indexed content and users’ privacy.

The purpose is to strengthen everyone’s confidence in the search engine and increase the effectiveness of Web search. Building trust in the search engine is based on innovations in (1) Security: computer security, privacy; (2) Completeness: completeness and heterogeneity of (re)sources; and (3) Neutrality: analysis, extraction, indexing, and classification of data.

Increasing the effectiveness of Web-based research relies on innovations related to (1) Relevance: variety and value of content taken into account, measurement of emotions carried by query results; (2) Interaction with the user: adaptation of the interfaces to the types of research; and (3) Performance: perceived relevance of results and response time.

The proposed innovations include:

• Design and develop models and tools for the detection of emotions in query results:
  • Ontology, thesaurus, linguistic resources
  • Metrics, indicators, classification of emotions
• Design and develop new crawling algorithms:
  • Dynamic crawling strategies
  • Crawlers and indexes for linked open data
• Ensure respect for privacy:
  • Detection of Internet tracking
  • Preventive display of tracing techniques
  • Certified security of automatic adaptation of ads to keywords entered by the user

DGA CONFIRMA

Participants: Elena Cabrio, Serena Villata.

This DGA project aims at automatically detecting fake news and limit their diffusion. In addition to identifying the communities propagating these fake news, we used methods from Natural Language Processing and Argumentation Theory to propose automatically extracted counter-arguments (adapted to target audience) from the existing reference press articles. These arguments allow to attack the false information detected in the fake news. Argument Mining techniques make it possible to (1) analyse the argumentation in natural language, for example by looking for the argumentative structures, identifying the relations of support or attack between the arguments; (2) locate the data related to specific information (related to fake news) on the Web. In the context of this project, Elena Cabrio and Serena Villata supervised the post-doc of Jerome Delobelle, now McF at University of Paris (LIPADE). Partners of the project: Storyzy, INRIA, Institut Jean Nicod. Duration of the project: 2018-2020.

Ministry of Culture: MonaLIA 3.0

Participants: Anna Bobasheva, Fabien Gandon, Frédéric Precioso.

The objective of the MonaLIA project is to exploit the crossover of the automatic learning methods particularly applied to image analysis and knowledge-based representation and reasoning, in particular for the semantic indexing of annotated works and images in JocondeLab. The goal is to identify automated or semi-automatable tasks to improve the annotation. This project follows the preliminary project “MonaLIA 1” which established the state of the art in order to evaluate the potential and the combination of learning (notably deep learning) and the semantization of annotations on the case of JocondeLab. In the project MonaLIA we now want to go beyond the preliminary study and to design and build a prototype and the methods assisting the creation, the improvement and the maintenance of the metadata of the image database in order to assist the actors of the cultural world in their daily tasks. The preliminary study identified several possible coupling points between deep learning from non-necessarily structured data and reasoning from structured data. This project proposes to select the most promising of them to carry out a proof of concept combining these methods by focusing on the assistance to the annotation and curation tasks of the metadata of a real base to improve the contents, the course and exploitation thereafter.

ANR WASABI

Participants: Michel Buffa, Elena Cabrio, Catherine Faron, Alain Giboin.

The ANR project WASABI started in January 2017 with IRCAM, Deezer, Radio France and the SME Parisson, consists in building a 2 million songs knowledge base of commercial popular music (rock, pop, etc.) Its originality is the joint use of audio-based music information extraction algorithms, song lyrics analysis algorithms (natural language processing), and the use of the Semantic Web. Web Audio technologies will then explore these bases of musical knowledge by providing innovative applications for composers, musicologists, music schools and sound engineers, music broadcasters and journalists. This project is in its mid-execution and gave birth to many publications in international conferences as well as some mainstream coverage (i.e for “la fête de la Science”). Participation in the ANR OpenMiage project aimed at offering online Bachelor and Master degrees.

Industrial transfer of some of the results of the WASABI project (partnership with AmpedStudio.com/Amp Track company) for integration of our software into theirs), SATT PACA.

Web site: http://wasabihome.i3s.unice.fr

ANR SIDES 3.0

Participants: Catherine Faron, Olivier Corby, Antonia Ettore, Fabien Gandon, Alain Giboin, Mathis Le Quiniou, Franck Michel.

Partners: Université Grenoble Alpes, Inria, Ecole Normale Supérieure de Lyon, Viseo, Theia.

SIDES 3.0 is an ANR project which started in fall 2017. It is led by Université Grenoble Alpes (UGA) and its general objective is to introduce semantics within the existing SIDES educational platform 4 for medicine students, in order to provide them with added value educational services. Within this project Catherine Faron supervised the post-doctoral work of Oscar Rodriguez now research engineer in the TeachOnMars company, the master internship of Mathis Le Quiniou and is now supervising the PhD work of Antonia Ettorre with Franck Michel. We are developping an approach to predict the success of students on training quizzes based on the knowledge graph representing their interactions with the pedagogical ressources within the SIDES platform.

Web site: https://www.uness.fr/anr/projets/dune/sides3.0

ANR D2KAB

Participants: Olivier Corby, Catherine Faron, Franck Michel.

Partners: LIRMM, INRA, IRD, ACTA

D2KAB is an ANR project which started in June 2019, led by the LIRMM laboratory (UMR 5506). Its general objective is to create a framework to turn agronomy and biodiversity data into knowledge –semantically described, interoperable, actionable, open– and investigate scientific methods and tools to exploit this knowledge for applications in science and agriculture. Within this project the Wimmics team is contributing to the lifting of heterogeneous dataset related to agronomy coming from the different partners of the project and is responsible to develop a unique entry point with semantic querying and navigation services providing a unified view on the lifted data.

Web site: http://www.d2kab.org

ANR DeKaloG

Participants: Olivier Corby, Catherine Faron, Fabien Gandon, Pierre Maillot, Franck Michel.

Partners: Université Nantes, INSA Lyon, INRIA Sophia Antipolis-Méditerranée

DeKaloG (Decentralized Knowledge Graphs) aims to: (1) propose a model to provide fair access policies to KGs without quota while ensuring complete answers to any query. Such property is crucial for enabling web automation, i.e. to allow agents or bots to interact with KGs. Preliminary results on web preemption open such perspective, but scalability issues remain; (2) propose models for capturing different levels of transparency, a method to query them efficiently, and especially, techniques to enable web automation of transparency. (3) propose a sustainable index for achieving the findability principle.

Web site: https://dekalog.univ-nantes.fr/

Smart Enseigno

Participants: Catherine Faron, Yuting Sun.

The Smart Enseigno project started in September 2019, led by Educlever. It is funded by the Ministry of National Education (MEN), within the Programme des Investissements d'Avenir (PIA2), action Partenariat d'innovation Intelligence artificielle(PI-IA) 56. This project aims at developing resources and intelligent services within the Educlever platform for secondary school mathematics education. Within this project Catherine Faron supervised the work of Yuting Sun aiming to adapt the approach developed in the framework of the SIDES project to the Enseigno platform. This platform now relies on a knowledge graph to capture the interactions for students with pedagogical resources.

DBpedia.fr

Participants: Fabien Gandon, Franck Michel.

The DBpedia.fr project proposes the creation of a French chapter of the DBpedia database. This project was the first project of the Semanticpedia convention signed by the Ministry of Culture, the Wikimedia foundation and Inria.

Web site: http://dbpedia.fr

Convention between Inria and the Ministry of Culture

Participants: Fabien Gandon.

We supervise the research convention with the Ministry of Culture to foster research and development at the crossroad of culture and digital sciences. This convention signed between Inria and the Ministry of Culture provides a framework to support projects at the cross-road of the cultural domain and the digital sciences.

Qwant-Inria Joint Laboratory

Participants: Fabien Gandon.

We supervise the Qwant-Inria Joint Laboratory where joint teams are created and funded to contribute to the search engine research and development. The motto of the joint lab is Smart Search and Privacy with five research directions:

• Crawling, Indexing, Searching
• Execution platform, privacy by design, security, ethics
• Maps and navigation
• Augmented interaction, connected objects, chatbots, personnal assistants
• Education technologies (EdTech)

We released the final, but confidential, report of the Qwant-Culture short-term project. This project aimed at identifying possibilities of exploiting the Qwant search engine to improve the search for information in the digital cultural resources of the French Ministry of Culture. Some possibilities have been selected to be the subject of research actions in the context a long-term project.

CovidOnTheWeb - Covid Inria program

Participants: Valentin Ah-Kane, Anna Bobasheva, Lucie Cadorel, Olivier Corby, Elena Cabrio, Jean-Marie Dormoy, Fabien Gandon, Raphaël Gazzotti, Alain Giboin, Abdelhadi Lebbar, Santiago Marro, Tobias Mayer, Aline Menin, Franck Michel, Andrea Tettamanzi, Serena Villata, Marco Winckler.

The project CovidOnTheWeb 48 aims to allow biomedical researchers to access, query and make sense of COVID-19 scholarly literature. To do so, we designed and implemented a pipeline that extends and combines tools meant to process, analyze and enrich corpora such as the COVID-19 Open Research Dataset (CORD-19) that gathers 100,000+ full-text scientific articles related to the coronaviruses. The methods employed leverage knowledge representation, text mining, argument mining, as well as data visualization and exploration techniques.

The generated RDF dataset comprises the Linked Data description of (1) named entities (NE) mentioned in the CORD-19 corpus and linked to DBpedia, Wikidata and other BioPortal vocabularies, and (2) arguments extracted using ACTA, a tool automating the extraction and visualization of argumentative graphs, meant to help clinicians analyze clinical trials and make decisions.

Among other tools, we rely on DBpedia Spotlight to identify and disambiguate NEs, and we used a local DBpedia instance to generate richer linksets linking NEs to other DBpedia chapters and Wikidata.

On top of this dataset, we have adapted Semantic Web tools (Corese, MGExplorer) to provide Linked Data visualizations that meet the expectations of the biomedical community. We are currently working on the implementation of data curation techniques that could be used to detect errors in the extraction and disambiguation of named entities. We plan for our future release of the dataset to use the latest English model of DBpedia Spotlight and then, in a next step, to detect entities in other languages with the same tool.

Web site: https://github.com/Wimmics/CovidOnTheWeb

3IA Côte d'Azur

Participants: Catherine Faron, Fabien Gandon, Freddy Limpens, Andrea Tettamanzi, Serena Villata.

We have three 3IA chairs for tenured researchers of Wimmics and several grants for PhD and postdocs.

We also have an industrial 3IA Affiliate Chair with the company Mnemotix focused on the industrialisation and scalability of the CORESE software.

10.1.1 Scientific events: organisation

10.1.2 Scientific events: selection

10.1.3 Journal

10.1.4 Invited talks

• Serena Villata was invited speaker of the 3rd International Conference on Intelligent Technologies and Applications (INTAP-2020): "Artificial Machines Arguing For And With People". September 28-30, 2020, Gjovik, Norvege14, and invited speaker of the Workshop on Dialogue, Explanation and Argumentation for Human-Agent Interaction, co-located with ECAI2020, September 7th, 202015.
• Elena Cabrio and Serena Villata were invited to present the Master Class organized by Telecom Valley: "Monitoring Cyberbullying through Message Classification and Social Network Analysis", November 2020, online16.
• Fabien Gandon was panelist of the ACM Web Science Conference 2020 Spotlight Panel 3 “Research Roadmap” https://www.southampton.ac.uk/wsi/websci20-panels.page
• Fabien Gandon gave an invited talk for ISWC 2020 Vision track https://www.youtube.com/watch?v=b9GPOOu2PTM.

10.1.5 Leadership within the scientific community

Fabien Gandon is a member of Semantic Web Science Association (SWSA) a non-profit organisation for promotion and exchange of the scholarly work in Semantic Web and related fields throughout the world ans steering committee of the ISWC conference.

Marco Winckler is Secretary of the IFIP TC13 on Human-Computer Interaction.

10.1.6 Scientific expertise

• Fabien Gandon : ERC-StG 2020 reviewer.
• Catherine faron : member of the ANR scientific evaluation committe ”Artificial Intelligence” (CE23) ; member of the scientific evaluation committe ”National Research Data Infrastructure” (NFDI) of the German research agency (DFG) ; reviewer of project proposals for the ANR regional call for projects Résilience Grand Est ; reviewer for the National Research Programme "Covid-19" of the Swiss National Science Foundation (SNSF) ; reviewer for the SESAME call for projects of Région Ile de France ; scientific referent of the Inria Learning Lab.
• Andrea Tettamanzi: reviewer of a CIFRE thesis proposal for ANRT; reviewer for the Swiss National Science Foundation.
• Marco Winckler: CHIST-ERA & ERC-AdG reviewer.

10.1.7 Research administration

• Andrea Tettamanzi and Marco Winckler are responsible for the SPARKS team of I3S.
• Fabien Gandon : evaluation committee for 3IA Côte d'Azur chairs ; Vice-director of Research Inria Sophia Antipolis ; jury DR2 Inria ; jury PEDR Inria ; Evaluation Committee of Inria.
• Catherine Faron : member of the HCERES comittee in charge of the evaluation of the LIRIS laboratory ; General Treasurer of the French Society for Artificial Intelligence (AFIA) ; member of the steering committee of the AFIA college on Knowledge Engineering ; member of the 2020 evaluation committee of Inria ; member of the CPRH 27 commission at Université Côte d'Azur.

10.2.1 Teaching

• Licence: Andrea Tettamanzi, Introduction à l'Intelligence Artificielle, 27 h ETD, L2, UCA, France.
• Licence: Michel Buffa, JavaScript, 40h, L3 MIAGE - Univ Côte d'Azur, France.
• Licence: Elena Cabrio, Web Technologies, 80 hours, (Portail Sciences de la Vie), UCA, France.
• Licence: Elena Cabrio, Internship supervision, 27 hours, (L3MIAGE), UCA, France.
• Master: Michel Buffa, Web technologies front and back end, 40h, M1, UCA, France.
• Master: Michel Buffa, Server Side JavaScript and modern front-end frameworks, 60h, M2 MIAGE - Univ Côte d'Azur NTDP and MBDS, UCA, France.
• Master: Michel Buffa, Introduction to AI, MIAGE - Univ Côte d'Azur Master 1 and Master 2 IA2.
• Master: Michel Buffa, Multiplayer game programming, IA for games.
• Master: Elena Cabrio, Computational Linguistics, 30 hours, (Lettres), UCA, France.
• Master: Elena Cabrio, Natural Language Processing for AI, 30 hours, (M1 INFO), UCA, France.
• Master and Licence: Elena Cabrio, Responsible of the intership programme, 40 hours, (L3 and M2 MIAGE), UCA, France.
• Master:Olivier Corby, Semantic Web, 20h, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master: Catherine Faron, Web languages, 48h, M1, Polytech Nice Sophia - Univ Côte d'Azur, UCA.
• Master: Catherine Faron, Semantic Web technologies (EN), 48h, M2 Informatique, Polytech Nice Sophia - Univ Côte d'Azur, UCA.
• Master: Catherine Faron, Knowledge Engineering 28h, M2 Informatique, Polytech Nice Sophia - Univ Côte d'Azur, UCA.
• Master: Catherine Faron, Semantic Web technologies (EN), 30h, M1 Data Science, UCA.
• Master: Catherine Faron, XML technologies, 16h, M2 IMAFA, Polytech Nice Sophia - Univ Côte d'Azur, UCA.
• Master: Catherine Faron, Projects and Internship tutoring, 32h, M2, Polytech Nice Sophia - Univ Côte d'Azur, UCA.
• Master: Fabien Gandon, Integrating Semantic Web technologies in Data Science developments, 78 h, M2, DSTI, France.
• Master: Oscar Rodríguez Rocha, Web of Data, 15h, M2, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master: Oscar Rodríguez Rocha, Knowledge Engineering, 10h, M2, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master: Andrea Tettamanzi, Logic for AI, 30 h ETD, M1, UCA, France.
• Master: Andrea Tettamanzi, Web, 30 h ETD, M1, UCA, France.
• Master: Andrea Tettamanzi, Algorithmes Évolutionnaires, 24.5 h ETD, M2, UCA, France.
• Master: Andrea Tettamanzi, Modélisation del l'Incertitude, 24.5 h ETD, M2, UCA, France.
• Licence (L3/SI3): Marco Winckler, Introduction to Human-Compute Interaction. 40 h ETD, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master (M2/SI5): Marco Winckler, Design adn Evaluation of Interactive Systems. 40 h ETD, Polytech Nice, France.
• Master (M2/SI5): Marco Winckler, Interaciton Techniques. 10 h ETD, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master (M2 DS4H): Accessibilité et Design Universel. 15 h ETD, UCA, France.
• Master (M2/SI5): Introduction to Scientific Research. 6 h EDT, Polytech Nice Sophia - Univ Côte d'Azur, France.
• Master (M2): Introduction to Scientific Research. 6 h EDT, UCA, France.
• Master (M2 MBDS): Visualization de données. 15 h EDT, UCA, France.
• Master (M1 SDAI): Visualization de données. 15 h EDT, UCA, France.
• Master 2: coordinator of the 5th year UE TER (Travaux de Recherche et Etude). 15h EDT, Polytech Nice Sophia - Univ Côte d'Azur, France.

E-learning

• Mooc: Michel Buffa, ”JavaScript Intro” published first in Juin 2017 on the EDx platform (MIT/Harvard), still active and updated regularly.
• Mooc: Michel Buffa, ”HTML5 Coding Essentials and Best Practices”
• Mooc: Michel Buffa, ”HTML5 Apps and Games”, also on EDx, are still active and updated regularly. More than 700.000 registered users since 2015 for these MOOCS.
• Mooc: Fabien Gandon, Olivier Corby & Catherine Faron, Web of Data and Semantic Web (FR), 7 weeks, http://www.france-universite-numerique.fr/, Inria, France Université Numérique, self-paced course 41002, Education for Adults, 10324 learners registered for 2020, https://www.fun-mooc.fr/courses/course-v1:inria+41002+self-paced/about
• Mooc: Fabien Gandon, Olivier Corby & Catherine Faron, Introduction to a Web of Linked Data (EN), 4 weeks, http://www.france-universite-numerique.fr/, Inria, France Université Numérique, self-paced course 41013, Education for Adults, 3827 learners registered for 2020, https://www.fun-mooc.fr/courses/course-v1:inria+41013+self-paced/about
• Mooc: Fabien Gandon, Olivier Corby & Catherine Faron, Web of Data (EN), 4 weeks, https://www.coursera.org/, Coursera, self-paced course Education for Adults, 3228 learners registered, https://coursera.org/learn/web-data

10.2.2 Supervision

• PhD in progress: Maroua Tikat, Interactive multimedia visualization for the exploration of multidimensional metadata database of popular music, UCA, Michel Buffa, Marco Winckler.
• PhD in progress: Shihong Ren, Which tools for music composition and real-time signal processing on the web, in a collaborative approach? UCA, Michel Buffa, Université de St Etienne, Laurent Pottier.
• PhD in progress: Molka Dhouib, Knowledge engineering in the sourcing domain for the recommendation of providers, UCA, Catherine Faron, Andrea Tettamanzi.
• PhD in progress: Ahmed El Amine Djebri, Uncertainty in Linked Data, UCA, Andrea Tettamanzi, Fabien Gandon.
• PhD in progress: Antonia Ettore, Artificial Intelligence for Education and Training: Knowledge Representation and Reasoning for the development of intelligent services in pedagogical environments, UCA, Catherine Faron, Franck Michel.
• PhD: Michael Fell, Natural Language Processing of Song Lyrics, UCA, Co-supervision Elena Cabrio and Fabien Gandon, July 2020. 3
• PhD: Raphaël Gazzotti, Knowledge graphs based extension of patients’ files to predict hospitalization, UCA, Catherine Faron, Fabien Gandon, April 59.
• PhD in progress: Santiago Marro, Argument-based Explanatory Dialogues for Medicine , UCA 3IA, Elena Cabrio and Serena Villata.
• PhD in progress: Nicholas Halliwell, Explainable and Interpretable Prediction, UCA, Fabien Gandon.
• PhD: Tobias Mayer, Argument Mining for Clinical Trials, UCA, Serena Villata, Elena Cabrio and Céline Poudat (UCA), December 2020 5.
• PhD in progress: Thu Huong Nguyen, Mining the Semantic Web for OWL Axioms, Andrea Tettamanzi, UCA.
• PhD in progress: Mahamadou Toure, Models and architectures for restricted and local mobile access to the Data Web , UCA, Fabien Gandon, Moussa Lo (UGB, Senegal).
• PhD in progress: Vorakit Vorakitphan, Argumentation and Emotions Emotion Detection with Adaptive Sentiment Analysis, Elena Cabrio, Serena Villata, UCA.
• PhD in progress: Ali Ballout, Active Learning for Axiom Discovery, Andrea Tettamanzi, UCA.
• PhD in progress: Rony Dupuy Charles, Combinaison d'approches symboliques et connexionnistes d'apprentissage automatique pour les nouvelles méthodes de recherche et développement en agro-végétale-environnement, Andrea Tettamanzi, UCA.
• PhD in progress: Lucie Cadorel, Localisation sur le territoire et prise en compte de l'incertitude lors de l’extraction des caractéristiques de biens immobiliers à partir d'annonces, Andrea Tettamanzi, UCA.

Internship

• Master internship: ElMahdi Ammari, GUI builder for WebAudio plugins (WebComponents) developed as part of the WASABI project. Integration into the FAUST IDE.
• Master Internship: Valeria Bellusci, Evolutionary Axiom Discovery from Populated Knowledge Bases, UCA, Andrea Tettamanzi.
• Master internship: Matthis Lequiniou, Prediction of student's success on the TeachOnMars Knowledge Graph, UCA, Catherine Faron & Oscar Rodríguez Rocha.
• Master internship: Zineb Rahhali, Machine learning to associate songs with presets of instruments and audio effects encoded in WebAudio.
• Master internship: Yuting Sun, Prediction of student's success on the Educlever Knowledge Graph, UCA, Catherine Faron & Franck Michel.
• Master internship: Abdelhadi Lebbar, Exploitation de données géospatiales à l’intersection entre graphes de connaissance et données d'imagerie satellitaire, Franck Michel & Marco Winckler.
• Master apprenticeship: Benjamin Molinet, Enriching the WASABI semantic dataset with NLP and audio processing.
• Master 2 internship: Valentin Ah-Kane. LinkedDataVis-bis - Vers un modèle de transformation générique pour la visualisation interactive de données linked-data, UCA.
• Master 1 intership: Jean-Marie Dormoy. Adaptation de l’outil de visualisation LinkedDataViz au domaine du COVID-19 : Interrogation et visualisation de données liées. UCA, Alain Giboin & Olivier Corby

10.2.3 Juries

Michel Buffa: Reviewer of Pasquale LISENA PhD : “Recommandation musicale basée sur la connaissance : modèles, algorithmes et recherche exploratoire”, defended October 11th, 2019, EURECOM – Sophia Antipolis

Elena Cabrio:

• Reviewer of the PhD committee of Giovanni Siragusa, University of Turin (Italy), 2020.
• Member of the PhD committee of Gabriel Meseguer Brocal, Ircam, 2020.

Catherine Faron:

• reviewer of Pierre Larmande's HDR, entitled Intégration de Données Multi-Echelles et Extraction de Connaissances en Agronomie: Exemples et Perspectives, defended on September 11 at Université de Montpellier;
• member of Konstantin Todorov's HDR, entitled Towards a Web of Structured Knowledge: Methods, Applications and Perspectives, defended on June 29 at Université de Montpellier;
• member of Patricia Serrano Alvarado's HDR, entitled Protecting user data in distributed systems, defended on June 16 at Université de Nantes;
• reviewer of Yves Mercadier's PhD thesis, entitled Classification automatique de textes par réseaux de neurones profonds : application au domine de la santé, defended on November 17 at Université de Montpellier;
• member of Pierre-Henri Paris' PhD thesis jury, entitled Identity in RDF Knowledge Graphs, defended on June 17 at Sorbonne Université;
• external member of the monitoring committee of Stella Zevio's PhD thesis at Université Paris Nord;
• external member of the monitoring committee of Francesco Bariatti's PhD thesis at Université de Rennes;
• external member of the monitoring committee of Charbel Obeid's PhD thesis at Université de Lyon.
• member of the monitoring committee of Thu Huong Nguyen's PhD thesis at Université Côte d'Azur.

Fabien Gandon:

• external member of the monitoring committee of Hicham Hossayni PhD thesis at Telecom SudParis, Institut Polytechnique de Paris;
• reviewer of Thomas Minier PhD thesis, entitled Web Preemption for Querying the Linked Open Data, defended on November 10th, 2020 at Université de Nantes, France;
• reviewer of Pierre Monnin PhD thesis, entitled Matching and mining in knowledge graphs of the Web of data Applications in pharmacogenomics, defended on December 16th, 2020 at Université de Lorraine, Loria, France;
• president of Elena Cabrio HDR thesis, entitled Artificial Intelligence to Extract, Analyze and Generate Knowledge and Arguments from Texts to Support Informed Interaction and Decision Making defended 22/10/2020, Université Côte d'Azur.
• reviewer for the Fondazione Bruno Kessler (FBK) Tenure Track program.

Alain Giboin :

• Invited Member of the PhD thesis jury of Marie Destandau (thesis title: "Path-Based Interactive Visual Exploration of Knowledge Graphs"), December 18, Paris-Saclay University.

Serena Villata:

• Gia-Lac Tran, EURECOM. Title of the thesis: "Advances of Deep Gaussian Processes: Calibration and Sparsification". Role: member of the jury. PhD defense: 2020.
• Benjamin Moreau, University of Nantes. Title of the thesis: “Facilitating Reuse on the Web of Data”, Role: reviewer. PhD defense: 2020.

Andrea Tettamanzi:

• Reviewer of Victor Eduardo Fuentes, PhD, Université du Québec à Montréal, Méta alignement méta heuristique, 6 octobre 2020;
• PhD Committee Chair for Edson Florez, Adverse drug reactions detection in clinical notes, Université Côte d'Azur, 01/07/2020;
• PhD Committee Chair for Raphaël Gazzotti, Prédiction d'hospitalisation par la génération de caractéristiques extraites de graphes de connaissances, Université Côte d'Azur, 30/04/2020;
• PhD Committee Chair for Gérald Rocher, Évaluation de l'Effectivité des Systèmes Ambiants, Université Côte d'Azur, 10/02/2020;

Marco Winckler:

• Reviewer of Jérôme Dupire HDR. “Vers une Accessibilité Accessible”. Presented on December 4th 2020, Université Paris 8 Vincennes Saint Denis, Paris, France.
• Reviewer of Tanguy Giuffrida PhD. “Fuzzy4U : un système d'adaptation des IHM en logique floue pour l'accessibilité”. Presented on December 12th 2020, Université Grenoble Alpes, Grenoble, France.
• Jury member of Aline Menin PhD. “eSTIMe: a visualization framework for assisting a multi-perspective analysis of daily mobility data”. Presented on November 26th 2020, Université Grenoble Alpes, Grenoble, France.

10.2.4 Teaching Administration

• Michel Buffa: director of MIAGE - Univ Côte d'Azur.
• Elena Cabrio: vice director of MIAGE - Univ Côte d'Azur.
• Catherine Faron: coordinator of the Web and AI option of the 5th year of Polytech Nice Sophia - Univ Côte d'Azur engineering school; pedagogical responsible of continuous training for the computer science department of Polytech Nice Sophia - Univ Côte d'Azur.
• Marco Winckler: coordinator of the Human-Computer Interaction track of the 5th year of Polytech Nice Sophia - Univ Côte d'Azur engineering school.

10.3.1 Articles and contents

Fabien Gandon:

• Article in “Annales des Mines Enjeux Numériques” – about “Une toile de fond pour le Web : lier les données et lier leurs vocabulaires sur la toile, pour un Web plus accessible aux machines” 12.
• Contributor to book / whitepaper “Éducation et numérique, Défis et enjeux” 69.
• Contributor again of the second version of the book / whitepaper “Artificial Intelligence: Current challenges and Inria's engagement” 65.
• Animation of reading sessions at the Knowledge Graph Conference book club on chapters of the textbook “Semantic Web for the Working Ontologist” 5117.

Elena Cabrio and Fabien Gandon are two characters in the comic book “Les défis de l'intelligence artificielle – Un reporter dans les labos de recherche” 66.

Publication of the third edition of the textbook “Semantic Web for the Working Ontologist” 51 with Fabien Gandon as new co-author.

International journals

• 6 articleSaverioS. Basso, AlbertoA. Ceselli and Andrea G. B.A. Tettamanzi. Random sampling and machine learning to understand good decompositionsAnnals of Operations Research28422020, 501-526
  HALDOIback to text
• 7 article BettinaB. Berendt, FabienF. Gandon, SusanS. Halford, WendyW. Hall, JimJ. Hendler, Katharina EK. Kinder-Kurlanda, EiriniE. Ntoutsi and SteffenS. Staab. Web Futures: Inclusive, Intelligent, Sustainable The 2020 Manifesto for Web Science Dagstuhl Manifestos 2021
  HAL DOI
• 8 articleJudyJ. Bowen, JeanJ. Vanderdonckt and MarcoM. Winckler. A Glimpse into the Past, Present, and Future of Engineering Interactive Computing SystemsProceedings of the ACM on Human-Computer Interaction 4 : EICSarticle 71June 2020, 1-32
  HALDOI
• 9 article AndreiA. Ciortea, SimonS. Mayer, SimonS. Bienz, FabienF. Gandon and OlivierO. Corby. Autonomous search in a social and ubiquitous Web Personal and Ubiquitous Computing June 2020
  HAL DOI back to text
• 10 articleMicheleM. Corazza, StefanoS. Menini, ElenaE. Cabrio, SaraS. Tonelli and SerenaS. Villata. A Multilingual Evaluation for Online Hate Speech DetectionACM Transactions on Internet Technology202May 2020, 1-22
  HALDOIback to text
• 11 articleRomainR. David, LaurenceL. Mabile, AlisonA. Specht, SarahS. Stryeck, MogensM. Thomsen, MohamedM. Yahia, ClementC. Jonquet, LaurentL. Dollé, DanielD. Jacob, DanieleD. Bailo, ElenaE. Bravo, SophieS. Gachet, HannahH. Gunderman, Jean-EudesJ.-E. Hollebecq, VassiliosV. Ioannidis, YvanY. Le Bras, EmilieE. Lerigoleur and AnneA. Cambon-Thomsen. FAIRness Literacy: The Achilles’ Heel of Applying FAIR PrinciplesCODATA Data Science Journal1932August 2020, 1-11
  HALDOI
• 12 article FabienF. Gandon. Une toile de fond pour le Web : lier les données et lier leurs vocabulaires sur la toile, pour un Web plus accessible aux machines Annales des Mines - Enjeux Numériques June 2020
  HAL back to text
• 13 articleValentinaV. Leone, LuigiL. Di Caro and SerenaS. Villata. Taking stock of legal ontologies: a feature-based comparative analysisArtificial Intelligence and Law2822020, 207-235
  HALDOIback to text
• 14 article ShihongS. Ren, StephaneS. Letz, YannY. Orlarey, RomainR. Michon, DominiqueD. Fober, MichelM. Buffa and JeromeJ. Lebrun. Using Faust DSL to Develop Custom, Sample Accurate DSP Code and Audio Plugins for the Web Browser Journal of the Audio Engineering Society 68 10 November 2020
  HAL back to text
• 15 articleThiagoT. Rocha Silva, MarcoM. Winckler and HallvardH. Trætteberg. Ensuring the Consistency between User Requirements and Task Models: A Behavior-Based Automated ApproachProceedings of the ACM on Human-Computer Interaction 4 : EICSarticle 77June 2020, 1-32
  HALDOI
• 16 articleGéraldG. Rocher, Jean-YvesJ.-Y. Tigli, StéphaneS. Lavirotte and NhanN. Le Thanh. Effectiveness assessment of Cyber-Physical SystemsInternational Journal of Approximate Reasoning1182020, 112-132
  HALDOI
• 17 articleOscarO. Rodríguez Rocha, CatherineC. Faron Zucker, AlainA. Giboin and AurélieA. Lagarrigue. Automatic Generation of Questions from DBpediaInternational Journal of Continuing Engineering Education and Life-Long Learningx12020, 1
  HALDOI

International peer-reviewed conferences

• 18 inproceedingsNicolasN. Broders, CéliaC. Martinie, PhilippeP. Palanque, MarcoM. Winckler and KimmoK. Halunen. A Generic Multimodels-Based Approach for the Analysis of Usability and Security of Authentication MechanismsHCSE 2020: Human-Centered Software EngineeringHCSE 2020 - 8th International Conference on Human-Centered Software Engineering - IFIP WG 13.2 International Working Conference12481Lecture Notes in Computer Science book series (LNCS)Eindhoven/ Online, NetherlandsSpringerNovember 2020, 61-83
  HALDOI
• 19 inproceedings MichelM. Buffa and JeromeJ. Lebrun. A FAUST-based re-creation of the power amp stage for WebAudio-based simulations of guitar tube amplifiers IFC 2020 - Second International Functional Audio Stream (Faust) Conference Saint-Denis / Virtual, France December 2020
  HAL back to text
• 20 inproceedings MichelM. Buffa, JeromeJ. Lebrun, ShihongS. Ren, StéphaneS. Letz, YannY. Orlarey, RomainR. Michon and DominiqueD. Fober. Emerging W3C APIs opened up commercial opportunities for computer music applications The Web Conference 2020 - DevTrack Taipei, Taiwan April 2020
  HAL back to text
• 21 inproceedings LucieL. Cadorel and Andrea G. B.A. Tettamanzi. Mining RDF Data of COVID-19 Scientific Literature for Interesting Association Rules WI-IAT'20 - IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology The 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT'20), 14-17 December 2020, a fully virtual conference Melbourne, Australia 2020
  HAL back to text back to text
• 22 inproceedingsOanaO. Cocarascu, ElenaE. Cabrio, SerenaS. Villata and FrancescaF. Toni. Dataset Independent Baselines for Relation Prediction in Argument MiningCOMMA 2020 - 8th International Conference on Computational Models of Argument326Frontiers in Artificial Intelligence and ApplicationsPerugia, ItalySeptember 2020, 45-52
  HALDOIback to text
• 23 inproceedingsAnne-MarieA.-M. Déry-Pinna, Philippe RenevierP. Gonin, MarcoM. Winckler, ChristopheC. Kolski, SophieS. Lepreux and JeanJ. Vanderdonckt. Entrain, exploring new territorial user interfacesEICS '20 Companion: Companion Proceedings of the 12th ACM SIGCHI Symposium on Engineering Interactive Computing SystemsEICS 2020 - 12th ACM SIGCHI Symposium on Engineering Interactive Computing SystemsArticle 7Sophia Antipolis, FranceACM: Association for Computing MachineryJune 2020, 1-4
  HALDOI
• 24 inproceedings AntoniaA. Ettorre, Oscar RodríguezO. Rocha, CatherineC. Faron Zucker, FranckF. Michel and FabienF. Gandon. A Knowledge Graph Enhanced Learner Model to Predict Outcomes to Questions in the Medical Field EKAW 2020 - 22nd International Conference on Knowledge Engineering and Knowledge Management Bolzano, Italy September 2020
  HAL back to text
• 25 inproceedings RaphaëlR. Gazzotti, CatherineC. Faron Zucker, FabienF. Gandon, VirginieV. Lacroix-Hugues and DavidD. Darmon. Injection of Automatically Selected DBpedia Subjects in Electronic Medical Records to boost Hospitalization Prediction SAC 2020 - 35th ACM/SIGAPP Symposium On Applied Computing Brno, Czech Republic March 2020
  HAL DOI back to text back to text
• 26 inproceedings HaiH. Huang and FabienF. Gandon. Learning URI Selection Criteria to Improve the Crawling of Linked Open Data (Extended Abstract) The 29th International Joint Conference on Artificial Intelligence (IJCAI 2020) Yokohama, Japan January 2021
  HAL back to text
• 27 inproceedings AmirA. Laadhar, ElcioE. Abrahão and ClementC. Jonquet. Analysis of Term Reuse, Term Overlap and Extracted Mappings across AgroPortal Semantic Resources EKAW 2020 - 22nd International Conference on Knowledge Engineering and Knowledge Management Bozen-Bolzano, Italy September 2020
  HAL
• 28 inproceedings AmirA. Laadhar, ElcioE. Abrahão and ClementC. Jonquet. Investigating One Million XRefs in Thirthy Ontologies from the OBO World ICBO'20 - 11th International Conference on Biomedical Ontologies 2807 G.1-12 Bozen-Bolzano, Italy CEUR Workshop Proceedings September 2020
  HAL
• 29 inproceedingsDarioD. Malchiodi, CéliaC. Da Costa Pereira and Andrea G. B.A. Tettamanzi. Classifying Candidate Axioms via Dimensionality Reduction TechniquesModeling Decisions for Artificial Intelligence - 17th International Conference, MDAI 2020, Sant Cugat, Spain, September 2-4, 2020, Proceedings; Modeling Decisions for Artificial Intelligence - 17th International Conference, MDAI 2020, Sant Cugat, Spain, September 2-4, 2020, ProceedingsMDAI 2020 - 17th International Conference on Modeling Decisions for Artificial IntelligenceSant Cugat, SpainSpringerAugust 2020, 179-191
  HALDOIback to text
• 30 inproceedings FranckF. Michel, GargominyG. Olivier, BenjaminB. Ledentec and BioschemasB. Community. Unleash the Potential of your Website! 180,000 webpages from the French Natural History Museum marked up with Bioschemas/Schema.org biodiversity types TDWG 2020 annual conference 4 Biodiversity Information Science and Standards (BISS) Virtual, France September 2020
  HAL DOI back to text
• 31 inproceedings Thu HuongT. Nguyen and Andrea G. B.A. Tettamanzi. A Multi-Objective Evolutionary Approach to Class Disjointness Axiom Discovery WI-IAT 2020 - IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology Melbourne/ Virtual, Australia December 2020
  HAL back to text
• 32 inproceedingsThu HuongT. Nguyen and Andrea G. B.A. Tettamanzi. Grammatical Evolution to Mine OWL Disjointness Axioms Involving Complex Concept ExpressionsCEC 2020 - IEEE Congress on Evolutionary ComputationGlasgow, United KingdomIEEEJuly 2020, 1-8
  HALDOIback to text
• 33 inproceedingsThu HuongT. Nguyen and Andrea G. B.A. Tettamanzi. Using Grammar-Based Genetic Programming for Mining Disjointness Axioms Involving Complex Class ExpressionsProceedings of the 25th International Conference on Conceptual Structures (ICCS 2020)ICCS 2020 - 25th International Conference on Conceptual Structures12277Lecture Notes in Computer ScienceBozen-Bolzano, ItalySpringerSeptember 2020, 18-32
  HALDOIback to text
• 34 inproceedingsClaudeC. Pasquier, CéliaC. Da Costa Pereira and Andrea G. B.A. Tettamanzi. Extending a Fuzzy Polarity Propagation Method for Multi-Domain Sentiment Analysis with Word Embedding and POS TaggingFrontiers in Artificial Intelligence and ApplicationsECAI 2020 - 24th European Conference on Artificial Intelligence325Santiago de Compostela, SpainIOS PressAugust 2020, 2140-2147
  HALDOIback to text
• 35 inproceedings ShihongS. Ren, LaurentL. Pottier and MichelM. Buffa. From Diagram to Code: a Web-based Interactive Graph Editor for Faust DSP Design and Code Generation IFC 2020 - Second International Functional Audio Stream (Faust) Conference Saint-Denis / Virtual, France December 2020
  HAL back to text
• 36 inproceedings MolkaM. Tounsi Dhouib, CatherineC. Faron and AndreaA. Tettamanzi. Injection of Knowledge in a Sourcing Recommender System WI-IAT'20 - IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology Melbourne / Virtual, Australia December 2020
  HAL back to text
• 37 inproceedings MahamadouM. Toure, KaladzaviK. Guidedi, FabienF. Gandon, MoussaM. Lo and ChristopheC. Guéret. MoRAI: Geographic and Semantic Overlay Network for Linked Data Access with Intermittent Internet Connectivity WI-IAT 2020 - IEEE/WIC/ACM International Joint Conference On Web Intelligence And Intelligent Agent Technology (WI-IAT '20) Melbourne, Australia 2021
  HAL back to text
• 38 inproceedings VorakitV. Vorakitphan, MarcoM. Guerini, ElenaE. Cabrio and SerenaS. Villata. Regrexit or not Regrexit: Aspect-based Sentiment Analysis in Polarized Contexts COLING'2020 - 28th International Conference on Computational Linguistics (COLING'2020) Barcelona / Online, Spain December 2020
  HAL back to text

National peer-reviewed Conferences

• 39 inproceedings HasnaaH. Daoud, MolkaM. Tounsi Dhouib, JerômeJ. Rancati, CatherineC. Faron and AndreaA. Tettamanzi. A Hybrid Bi-LSTM-CRF Model for Sequence Labeling Applied to the Sourcing Domain PFIA-APIA 2020 - 5ème Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle Angers, France June 2020
  HAL back to text

Conferences without proceedings

• 40 inproceedings MicheleM. Corazza, StefanoS. Menini, ElenaE. Cabrio, SaraS. Tonelli and SerenaS. Villata. Hybrid Emoji-Based Masked Language Models for Zero-Shot Abusive Language Detection EMNLP 2020 - Conference on Empirical Methods in Natural Language Processing Virtual, France November 2020
  HAL
• 41 inproceedings JérômeJ. Delobelle, AmauryA. Delamaire, ElenaE. Cabrio, RamónR. Ruti and SerenaS. Villata. Sifting the Arguments in Fake News to Boost a Disinformation Analysis Tool NL4AI 2020 - 4th Workshop on Natural Language for Artificial Intelligence Online, Italy November 2020
  HAL back to text
• 42 inproceedings Ahmed El AmineA. Djebri, AntoniaA. Ettorre and JohannJ. Mortara. Towards a Linked Open Code ESWC 2021 - The 18th Extended Semantic Web Conference Heraklion / Virtual, Greece June 2021
  HAL
• 43 inproceedings Ahmed El AmineA. Djebri, Andrea G. B.A. Tettamanzi and FabienF. Gandon. Task-Oriented Uncertainty Evaluation for Linked Data Based on Graph Interlinks EKAW 2020 - 22nd International Conference on Knowledge Engineering and Knowledge Management Bozen-Bolzano, Italy September 2020
  HAL DOI back to text
• 44 inproceedings MichaelM. Fell, ElenaE. Cabrio, ElmahdiE. Korfed, MichelM. Buffa and FabienF. Gandon. Love Me, Love Me, Say (and Write!) that You Love Me: Enriching the WASABI Song Corpus with Lyrics Annotations LREC 2020 - 12th edition of the Language Resources and Evaluation Conference Marseille, France May 2020
  HAL back to text
• 45 inproceedings TobiasT. Mayer, ElenaE. Cabrio and SerenaS. Villata. Transformer-based Argument Mining for Healthcare Applications Proceedings of ECAI 2020 - 24th European Conference on Artificial Intelligence ECAI 2020 - 24th European Conference on Artificial Intelligence Santiago de Compostela / Online, Spain August 2020
  HAL back to text
• 46 inproceedings TobiasT. Mayer. Enriching Language Models with Semantics ECAI 2020 - 24th European Conference on Artificial Intelligence Santiago de Compostela / Online, Spain August 2020
  HAL
• 47 inproceedings TobiasT. Mayer, SantiagoS. Marro, ElenaE. Cabrio and SerenaS. Villata. Generating Adversarial Examples for Topic-dependent Argument Classification COMMA 2020 - 8th International Conference on Computational Models of Argument Perugia, Italy September 2020
  HAL back to text
• 48 inproceedings FranckF. Michel, FabienF. Gandon, ValentinV. Ah-Kane, AnnaA. Bobasheva, ElenaE. Cabrio, OlivierO. Corby, RaphaëlR. Gazzotti, AlainA. Giboin, SantiagoS. Marro, TobiasT. Mayer, MathieuM. Simon, SerenaS. Villata and MarcoM. Winckler. Covid-on-the-Web: Knowledge Graph and Services to Advance COVID-19 Research ISWC 2020 - 19th International Semantic Web Conference Athens / Virtual, Greece November 2020
  HAL DOI back to text back to text back to text
• 49 inproceedings MargaridaM. Romero, NataliaN. Timus, SophieS. Raisin, IsabelleI. Mirbel and IannisI. Aliferis. Accompagner la transformation des pratiques enseignantes à l’Université Côte d’Azur par le biais d’une approche centrée sur la professionnalisation des enseignant.e.s AIPU 2020 - 31e Congrès de l’Association Internationale de Pédagogie Universitaire Québec, Canada 2020
  HAL DOI

Scientific books

• 50 book JoséJ. Abdelnour Nocera, AntigoniA. Parmaxi, MarcoM. Winckler, FernandoF. Loizides, CarmeloC. Ardito, GaneshG. Bhutkar and PeterP. Dannenmann. Beyond Interactions: INTERACT 2019 IFIP TC 13 Workshops 2019, Revised Selected Papers 11930 Lecture Notes in Computer Science book series (LNCS) Paphos, Cyprus Springer January 2020
  HAL DOI
• 51 book DeanD. Allemang, JimJ. Hendler and FabienF. Gandon. Semantic Web for the Working Ontologist 3 ACM June 2020
  HAL DOI back to text back to text back to text

Scientific book chapters

• 52 inbook FrédériqueF. Bertoncello, Marie-JeanneM.-J. Ouriachi, CéliaC. Da Costa Pereira, Andrea G. B.A. Tettamanzi, LouiseL. Purdue and RamiR. Ajroud. Using ABM to explore the role of socio-environmental interactions on Ancient Settlement Dynamics. Human history and digital future, Proceeding of the 46th Computer Applications and Quantitative Methods in Archaeology international conference (CAA 2018), Tübingen, mars 2018. 2021
  HAL back to text
• 53 inbookAdrianA. Reuter, KarimaK. Boudaoud, MarcoM. Winckler, AhmedA. Abdelmaksoud and WadieW. Lemrazzeq. Secure Email - A Usability Study12063FC 2020: Workshops, International Conference on Financial Cryptography and Data SecurityLecture Notes in Computer Science book series (LNCS, volume 12063)August 2020, 36-46
  HALDOI
• 54 inbookAndrea G. B.A. Tettamanzi, DavidD. Emsellem, CéliaC. Da Costa Pereira, AlessandroA. Venerandi and GiovanniG. Fusco. Possibilistic Estimation of Distributions to Leverage Sparse Data in Machine LearningInformation Processing and Management of Uncertainty in Knowledge-Based Systems. IPMU 2020.Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Lisbon, Portugal, June 15-19, 2020, Proceedings, Part ILisbon, PortugalSpringerJune 2020, 431-444
  HALDOIback to textback to text

Edition (books, proceedings, special issue of a journal)

• 55 proceedings J. Bowen J. Vanderdonckt M. Winckler EICS '20 Companion: Companion Proceedings of the 12th ACM SIGCHI Symposium on Engineering Interactive Computing Systems 12th ACM SIGCHI Symposium on Engineering Interactive Computing Systems (EICS 2020) Sophia Antipolis (virtual), France ACM : Association for Computing Machinery, New York, NY, United States June 2020
  HAL DOI
• 56 proceedings G. Tortora G. Vitiello M. Winckler AVI '20: Proceedings of the International Conference on Advanced Visual Interfaces International Conference on Advanced Visual Interfaces (AVI 2020) Ischia Island, Italy ACM : Association for Computing Machinery, New York, NY, United States September 2020
  HAL DOI

Doctoral dissertations and habilitation theses

• 57 thesis ElenaE. Cabrio. Artificial Intelligence to Extract, Analyze and Generate Knowledge and Arguments from Texts to Support Informed Interaction and Decision Making Université Côte d’Azur October 2020
  HAL back to text
• 58 thesis MichaelM. Fell. Natural language processing for music information retrieval : deep analysis of lyrics structure and content Université Côte d'Azur May 2020
  HAL
• 59 thesis RaphaëlR. Gazzotti. Knowledge graphs based extension of patients’ files to predict hospitalization Université Côte d'Azur April 2020
  HAL back to text
• 60 thesis TobiasT. Mayer. Argument Mining on Clinical Trials Université Côte d'Azur December 2020
  HAL
• 61 thesis AlineA. Menin. eSTIMe: a visualization framework for assisting a multi-perspective analysis of daily mobility data Grenoble UGA - Université Grenoble Alpes November 2020
  HAL

Reports & preprints

• 62 misc OanaO. Cocarascu, ElenaE. Cabrio, SerenaS. Villata and FrancescaF. Toni. A Dataset Independent Set of Baselines for Relation Prediction in Argument Mining December 2020
  HAL
• 63 misc MarieM. Destandau, OlivierO. Corby, Jean-DanielJ.-D. Fekete and AlainA. Giboin. Path Outlines: Browsing Path-Based Summaries of Linked Datasets February 2020
  HAL

Scientific popularization

• 64 misc F. Gandon. Les IA comprennent-elles ce qu’elles font ? November 2020
  HAL