Project information

Inteligentní software pro sémantické hledání dokumentů (ISSHD)

Project Identification
TD03000295
Project Period
1/2016 - 12/2017
Investor / Pogramme / Project type
Technology Agency of the Czech Republic
MU Faculty or unit
Faculty of Informatics
Project Website
https://scaletext.com
Keywords
scalable semantic search systems; semantic search; document topic modeling; machine learning; search; deep learning
Cooperating Organization
RaRe Technologies s.r.o.
Responsible person RNDr. Radim Řehůřek, Ph.D.
Responsible person RNDr. Radim Řehůřek, Ph.D.
Responsible person RNDr. Jan Pomikálek, Ph.D.
Responsible person RNDr. Jan Rygl
Investor logo

Our society, research and culture is defined by words, which in today's information society
constitute _documents_.
Project goal is to develop a database system (software),
which will allow searching based on related documents based on their _meaning_ (semantics).
System Scaletext consists from three parts:

  • semantic analysis: arbitrary unstructured document in natural language (English, Czech) is analyzed
  • indexing: document topics and structure are represented and stored internally using _semantic_


representation in such a way, that system is then capable of semantic similarity search given a document query.

  • search: given input query document, system finds semanticaly closed documents, that are closest to [latent] meaning of the query, even though they do not share same keywords

Results

https://www.rvvi.cz/cep?s=jednoduche-vyhledavani&ss=detail&n=0&h=TD03000295

Publications

2017

2016

You are running an old browser version. We recommend updating your browser to its latest version.

More info