Natural Language Learning: Learning for syntactic and semantic parsing, lexicon acquisition, information extraction, machine translation, and word sense disambiguation.
Text Mining: Discovering knowledge from text using information extraction and rule induction.
Information Extraction: Identifying specific pieces of structured data in web pages or natural-language documents.
Learning for Recommender Systems and Text Categorization: Content-based recommending and automated web-page classification.
Semi-Supervised Learning : Learning from a mixture of labeled and unlabeled data.
Inductive Logic Programming: Learning Prolog programs (rules in first-order predicate logic) from examples.
Knowledge-Base and Theory Refinement: Automatically modifying rule bases and Bayesian networks to fit empirical data.