Expressive and interpretable data mining models for life sciences applications.

The goal of this project is to develop data mining techniques that produce expressive and interpretable models. Such models are of crucial importance to life sciences, where they help the domain expert gain insight in the processes that take place in this field, which are typically complex and difficult to observe. The project focuses on inductive databases and statistical relational mining; these form a solid basis for developing the necessary techniques. Inductive databases combine data mining algorithms and a database in one system. Statistical relational data mining algorithms implement relational mining techniques that explicitly represent their uncertainty about the data in the models by means of probabilities. The planned research consists of the following three parts: (1) improving the expressivity of inductive databases; (2) developing new statistical relational mining techniques; and (3) applying the results of part (1) and (2) in life sciences applications.

Keywords:Life Sciences, Probabilistic Models, Logic, Data Mining, Machine Learning

Disciplines:Applied mathematics in specific fields, Computer architecture and networks, Distributed computing, Information sciences, Information systems, Programming languages, Scientific computing, Theoretical computer science, Visual computing, Other information and computing sciences, Artificial intelligence, Cognitive science and intelligent systems