What's in subtitles?

From Master Projects
Revision as of 15:17, 14 December 2015 by Vmo600 (talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


About What's in subtitles?

Description

TV programs subtitles can be a rich source of information which main titles usually don't. Especially when talking about series, only the subtitle give some indication of the actual content of a program. For instance, take into consideration documentaries.

This Master Project has the aim of showing how subtitles can improve the recall in recommender systems. Would subtitles be more effective than the description of the program? Can we extract keywords that can classify the program topic better? Shall we combine subtitles with other sources to reach a better result?

Tasks

  • extracting descriptive words from subtitles
  • performing teletext-based TV recommendation experiments

Tools and Data

  • Teletext + vocabularies + wikipedia
  • data analysis tools, e.g. Excel, R, etc.
  • Natural Language Tools

Recommended prior knowledge

  • Knowledge & Media course
  • Social Web course
  • Research methods course
  • Semantic Web

Extra Information

Contact Valentina_Maccatrozzo, Lora_Aroyo or Guus_Schreiber for more information about this project.