Section: New Results

The Mining of Complex Data

Participants : Mehwish Alam, Isiru Bayissa, Thomas Bourquard, Aleksey Buzmakov, Victor Codocedo, Adrien Coulet, Elias Egho, Nicolas Jay, Mehdi Kaytoue, Florence Le Ber, Ioanna Lykourentzou, Luis Felipe Melo, Amedeo Napoli, Chedy Raïssi, Lian Shi, Yannick Toussaint.

Formal concept analysis, together with itemset search and association rule extraction, are suitable symbolic methods for KDDK, that may be used for real-sized applications. Global improvements may be carried on the scope of applicability, the ease of use, the efficiency of the methods, and on the ability to fit evolving situations. Accordingly, the team is working on extensions of such symbolic data mining methods to be applied on complex data such as biological or chemical data or textual documents, involving objects with multi-valued attributes (e.g. domains or intervals), n-ary relations, sequences, trees and graphs.

FCA, RCA, and Pattern Structures

Recent advances in data and knowledge engineering have emphasized the need for Formal Concept Analysis (FCA) tools taking into account structured data. There are a few extensions of FCA for handling contexts involving complex data formats, e.g. graphs or relational data. Among them, Relational Concept Analysis (RCA) is a process for analyzing objects described both by binary and relational attributes [116] . The RCA process takes as input a collection of contexts and of inter-context relations, and yields a set of lattices, one per context, whose concepts are linked by relations. RCA has an important role in KDDK, especially in text mining [85] , [84] .

Another extension of FCA is based on Pattern Structures (PS) [90] , which allows to build a concept lattice from complex data, e.g. nominal, numerical, and interval data. In (major [5] ), pattern structures are used for building a concept lattice from intervals, in full compliance with FCA, thus benefiting of the efficiency of FCA algorithms. Actually, the notion of similarity between objects is closely related to these extensions of FCA: two objects are similar as soon as they share the same attributes (binary case) or attributes with similar values or the same description (at least in part). Various results were obtained in the study of the relations existing between FCA with an embedded explicit similarity measure and FCA with pattern structures [48] . Moreover, similarity is not a transitive relation and this lead us to the study of tolerance relations. In addition, a new research perspective is aimed at using frequent itemset search methods for mining interval-based data being guided by pattern structures and biclustering as well [50] , [49] .

Pattern structures in association with a similarity measure were applied in the field of decision support in agronomy. In this domain, a set of agro-ecological indicators is aimed at helping farmers to improve their agricultural practices by estimating the impact of cultivation practices on the “agrosystem”. The modeling and the assessment of environmental risk require a large number of parameters whose measure is imprecise. The propagation of the imprecision and the different types of imprecision have to be taken into account in the computation of the value of indicators for decision support. Actually, based on pattern structures with a associated similarity measure, this problem has been approached as an information fusion problems with substantial results [34] , [35] .

Miscellaneous in FCA and Pattern Mining

In the field of medicine, an approach based on a combination of FCA with sequential pattern mining was developed to explore patients care trajectories (PCT) [46] . When PCT are modeled as multidimensional and multilevel sequences [108] , the results of a frequent sequential itemsets search feed an FCA step in order to compute interests measures such as concept stability. These measures help the experts to find the most interesting sequential patterns.

In the context of environmental sciences, research work is in concern with the mining of complex hydroecological data with concept lattices. FCA was compared and combined with statistical approaches to deal with multi-valued contexts in hydroecology [31] , [27] , [39] . Regarding the preparation of agronomical data, we have developed an episode-based analysis about the design of information systems (actually, this work was carried out during the ANR-ADD COPT project between 2005 and 2008). We focused on the experience of persons in charge of building observatoires, i.e. information systems, for the monitoring and the management of rural territories [32] . Moreover, Florence Le Ber –as a member of UMR 7517 Lhyges, Strasbourg– is the scientific head of an ANR project named “FRESQUEAU” (2011–2014) dealing with FCA and data mining and hydroecological data (see http://fresqueau.engees.eu/ ).

For completing the work on itemset search, there is still on-going work on frequent and rare itemset search, for being able to build lattices from very large data and completing the algorithm collection of the Coron platform. This year, results were obtained on the design of an integrated and modular algorithm for searching for closed and generators itemsets, and equivalence classes of itemsets, thus enabling the construction of the associated lattice [56] . This research aspect is also linked to the research carried on within a the PICS CaDoE research project (see Section  8.1.3 ).

Skylines, sequences and privacy

Pattern discovery is at the core of numerous data mining tasks. Although many methods focus on efficiency in pattern mining, they still suffer from the problem of choosing a threshold that influences the final extraction result. The goal of a study done this current year (2011) is to make the results of pattern mining useful from a user-preference point of view. That is, take into account some domain knowledge to guide the pattern mining process. To this end, we integrate into the pattern discovery process the idea of skyline queries in order to mine skyline patterns in a threshold-free manner. This forms the basis for a novel approach to mining skyline patterns. The efficiency of our approach was illustrated over a use case from chemoinformatics and we showed that small sets of dominant patterns are produced under various measures that are interesting for chemical engineers and researchers [55] .

Sequence data is widely used in many applications. Consequently, mining sequential patterns and other types of knowledge from sequence data has become an important data mining task. The main emphasis has been on developing efficient mining algorithms and effective pattern representation.

However, important fundamental problems still remained open: $\left(i\right)$ given a sequence database, can we have an upper bound on the number of sequential patterns in the database? $\left(ii\right)$ Is the efficiency of the sequence classifier only based on accuracy? $\left(iii\right)$ Do the classifiers need the entire set of extracted patterns or a smaller set with the same expressiveness power?

In three different works on sequences, we study the problem of bounding sequential patterns with the combinatorial complexity of sequences and the problem of sequence classifiers with the constraints of optimizing both accuracy and earliness [53] , [46] .

Orpailleur is one of the few project-teams working on privacy challenges which are becoming a core issue with different scientific problems in computer science. Privacy-preserving data publication has been studied intensely in the past years. In our recent works, we introduce two different data anonymization methodologies based on different usability scenarios [57] , [58] .

KDDK in Text Mining

Ontologies help software and human agents to communicate by providing shared and common domain knowledge, and by supporting various tasks, e.g. problem-solving and information retrieval. In practice, building an ontology depends on a number of “ontological resources” having different types: thesaurus, dictionaries, texts, databases, and ontologies themselves. We are currently working on the design of a methodology and the implementation of a system for ontology engineering from heterogeneous ontological resources. This methodology is based on both FCA and RCA, and was previously successfully applied in contexts such as astronomy and biology. At present, an engineer is in charge of implementing a robust system being guided by the previous research results and preparing the way for some new research directions involving trees and graphs.

In another work in text mining [19] , we propose a method based on a syntactic parsing for extracting rich semantic relationships between pairs of entities co-occurring in a single sentence. The method was applied in pharmacogenomics (study of the impact of individual genomic variation on drug responses) and we obtained a resource encoded in RDF that summarizes pharmacogenomics relationships mentioned into roughly 17 million Medline abstracts. This resource appears to be of major interest since it is used to guide human curation of biomedical databases, and to derive new knowledge about drug-drug interactions [92] .