Section: Research Program

Hybrid and Exploratory Knowledge Discovery

Keywords: knowledge discovery in databases, knowledge discovery in databases guided by domain knowledge, data mining, data exploration, formal concept analysis, classification, pattern mining, numerical methods in data mining.

Knowledge discovery in databases (KDD) aims at discovering intelligible and reusable patterns in possibly large databases. These patterns can then be interpreted as knowledge units to be reused in knowledge-based systems. From an operational point of view, the KDD process is based on three main steps: (i) selection and preparation of the data, (ii) data mining, (iii) interpretation of the discovered patterns. Moreover, the KDD process is iterative, interactive, and generally controlled by an expert of the data domain, called the analyst. The analyst selects and interprets a subset of the extracted units for obtaining knowledge units having a certain plausibility. In this view, KDD is an exploratory process similar to “exploratory data analysis”.

The KDD process –as implemented in the Orpailleur team– is based on data mining methods which are either symbolic or numerical. Symbolic methods are based on pattern mining (e.g. mining frequent itemsets, association rules, sequences...), Formal Concept Analysis (FCA) and extensions such as Pattern Structures and Relational Concept Analysis (RCA), and redescription mining. Numerical methods are based on Random Forests, Support Vector Machines (SVM), Neural Networks, and probabilistic approaches such as second-order Hidden Markov Models (HMM). Moreover, for being able to deal with complex data, numerical data mining methods can be associated with symbolic methods, for improving applicability and efficiency of knowledge discovery. This is particularly true in classification, where supervised and unsupervised approaches may be combined with benefits.

A main operation in the research work of Orpailleur is “classification”, which is a polymorphic process involved in modeling, mining, representing, and reasoning tasks. In this way, domain knowledge, when available, can improve and guide the KDD process, materializing the idea of Knowledge Discovery guided by Domain Knowledge or KDDK. In KDDK, domain knowledge plays a role at each step of KDD: the discovered patterns can be interpreted as knowledge units and reused for problem-solving activities in knowledge systems, implementing the exploratory process “mining, interpreting, modeling, representing, and reasoning”. Then knowledge discovery can be considered as a key task in knowledge engineering (KE), having an impact in various semantic activities, e.g. information retrieval, recommendation, and ontology engineering. In addition, if knowledge discovery can feed knowledge-based systems, in turn, domain knowledge can be used to support the knowledge discovery process.

Finally, life sciences, i.e. agronomy, biology, chemistry, and medicine, are application domains where the Orpailleur team has a very rich experience. The team intends to keep and to extend this experience, paying also more attention to the impact of knowledge discovery in the real world. This should lead to the design of green (sustainable), explainable, and fair data mining systems.