Section: New Results
In this section, we organize our contributions this year along two of our research axes, namely Pattern Mining and Decision Support. These correspond to the contributions that has been accepted for publication this year.
In the domain of pattern mining we can categorize our contributions along the following lines:
Mining of novel types of patterns. This includes temporal pattern mining, signature mining, opinion mining in uncertain databases, interval rules, and top-k item-centric mining. All these contributions have been proposed as solutions to problems in the domains of pharmaco-epidemiology, retail databases, biomedical databases, and analysis of speech corpora. We provide more details about these results in Sections 7.2 to 7.9.
Data Mining with ASP. Answer Set Programming is a powerful search tool in combinatorial spaces, which can be naturally ported to pattern mining, as the latter is a specific type of search problem. Our contributions include the application of ASP in the discovery of frequent, constrained, condensed, and rare sequential patterns. Sections 7.11 and 7.12 elaborate on our new research insights.
Data Mining for the masses. In , we propose a communication model that bridges knowledge delivery between data miners and domain users in the field of library science. Our model proposes a five-steps process in order to achieve effective knowledge synthesis and delivery of insights to the domain users.
In regards to the axis of decision support, our contributions can be organized in two categories: exploration and diagnosis.
Exploration. We propose two exploration methods in the context of analysis of trajectories and agro-environmental systems. We propose customized data models and resort to data-warehousing and multidimensional data representations to facilitate the querying, and thus the exploration and understanding of the data, for the sake of decision making. Our results in this line are further detailed in Sections 7.13 to 7.15.
Diagnosis. In Section 7.16 we propose a novel method for anomaly detection in time series by resorting to Extreme Value Theory. In addition,  offers a formalization of diagnosis based on automata with focus on discrete event systems.