Information Extraction - Sequential Probability Models

Table of Contents / To Do

• Introduction
• Information Extraction (IE)
  • Definition
  • Distinction between Information Extraction, Information Retrieval, Data Mining, Text Mining
• Methodology
  • Present and compare different IE approaches
• Sequential Probability Models
  • Theory, Application for IE in combination with trigger phrases
  • Extensions of Trigger Phrases (Meronymic Relations, …)
• Implementation
  • identify weaknesses in the current trigger phrase implementation
  • apply and implement a sequential probability model
  • evaluation

Practical points

• identify weaknesses in the current trigger phrase implementation
• make use of sequential probability models
• evaluation of the new approach

Literature

• Text Mining - Predictive Methods for Analyzing Unstructured Information (Sholom Weiss)
• Hearst, M. A. : “Automatic Acquisition of Hyponyms from Large Text Corpora”
• Berland, M. and Charniak, E. :”Finding Parts in Very Large Corpora”
• Grefenstette, G. and Hearst, M. A.: “A Method for Refining Automatically-Discovered Lexical Relations: Combining Weak Techniques for Stronger Results”
• Litz, B. and Langer, H. and Malaka, R.: “Trigrams’n’Tags for Lexical Knowledge Acquisition”, Proceedings of the First International Conference on Knowledge Mining and Information Retrieval (KDIR 2009), Madeira, Portugal
• Hamish Cunningham: “Software Architecture for Language Engineering”
• Literature covering Hearst Patterns

Student profile

• good programming skills

Source

The sources can be downloaded from the svn repository:

https://svn.semanticlab.net/svn/oss/thesis/information-extraction/infoex-workspace