Inhalt

[ 921INSYIREK13 ] KV Information Retrieval and Extraction

Versionsauswahl
Es ist eine neuere Version 2021W dieser LV im Curriculum Master's programme Business Informatics 2021W vorhanden.
(*) Unfortunately this information is not available in english.
Workload Education level Study areas Responsible person Hours per week Coordinating university
3 ECTS M1 - Master's programme 1. year Computer Science Birgit Pröll 2 hpw Johannes Kepler University Linz
Detailed information
Original study plan Master's programme Computer Science 2021S
Objectives Students have competence in fundamentals and technologies of (1) Information Retrieval, comprising representation, storage and retrieval of textual unstructured information, (2) Information Extraction, comprising Named Entity Recognition (NER) and Natural Language Processing/Understanding (NLP/NLU), and (3) Dialogue Systems. They are able to implement and evaluate applications in these fields and have knowledge about related fields and current research topics.
Subject 1) Fundamentals and concepts of traditional information retrieval (IR)

  • Document representation: indexing, weighting (tf*idf)
  • IR models: boolsch, vector space etc.
  • Architectures and (natural language) user interfaces
  • Evaluation in IR: recall, precision
  • Related concepts: string similarity, thesaurus, classification, relevance feedback, query expansion, context-based IR
  • IR tools and applications

2) Fundamentals and concepts of information extraction (IE)

  • IE types: NER, relation extraction etc.
  • IE approaches and architectures: focusing on knowledge/rule-based approaches
  • natural language processing/understanding (NLP/NLU)
  • evaluation in IE
  • IE tools and applications

3)Fundamentals and concepts of dialogue systems (DS)

  • properties of human conversation, dialogue structure and state
  • DS approaches, architectures, end evaluation
  • search and extraction in dialogue systems
  • DS tools and applications

4) Selected topics and current research

  • information filtering & recommender systems
  • text recognition, optical character recognition (OCR)
  • multilingual/crosslingual IR
  • text summarization
  • natural language generation (NLG) etc
Criteria for evaluation exercises, exam, in-class contribution
Methods slide presentation (slides on Moodle), exercises (group work)
Language English
Study material
  • Ricardo Baeza-Yates, Berthier Ribeiro-Neto: Modern Information Retrieval, Addison Wesley 2010
  • Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze: Introduction to Information Retrieval, Cambridge University Press 2008
  • W. Bruce Croft, Donald Metzler, Trevor Strohman: Search Engines – Information Retrieval in Practice, Pearson 2009
Changing subject? No
Corresponding lecture (*)in collaboration with 921INSYASWK13: KV Accessible Software and Web Design (1,5 ECTS) equivalent to
INMIPKVKCSY: KV Knowledge-centered Systems (4,5 ECTS)
On-site course
Maximum number of participants -
Assignment procedure Direct assignment