max planck institut
mpii logo Minerva of the Max Planck Society

Seminar "Probabilistic Models for Information Extraction", Summer '11

Dr. Martin Theobald, M.Sc. Maximilian Dylla


Contents of the Seminar

The seminar focuses on probabilistic models which are commonly used in the context of information extraction tasks, such as named-entity recognition, part-of-speech tagging, co-reference resolution, and statistical parsing. The probabilistic models we cover in the seminar include Hidden Markov Models (HMMs), Conditional Random Fields (CRFs), Constrained Conditional Models (CCMs), Markov Logic Networks (MLNs) and related techniques. Talks will be based on recent research papers in this field.

Background Literature

(Books can be borrowed from the campus library or directly from us.)

Requirements for the Certificate

Papers & Talks