Main Sector of relevance\IRC classification 5. Electronics, IT and Telecommunication

5.4 System of automatic indexing and annotation of scientific and technical information

Developers’ contact information

State Scientific Institution “The United Institute of Informatics Problems of the National Academy of Sciences of Belarus”
6 Surganova Str., 220012 Minsk

Summary

The system is designed for automated detection of key word and informative sentences in scientific and technical texts. It can be used in scientific-and-technical library for automated indexation and annotation of scientific and technical articles and books as well as by Internet users for preliminary studying of text-based documents content.

Description

The system uses new approach for information content estimation of words based on comparison of their statistical characteristics in annotated text and complete text frame. The system allows:
- excluding non-informative common words without use of preliminary list of such words;
- descending ranking of informative words under degree of their information content (percents);
- presetting of required report volume (number of sentences).

Technology type

• Software

Technical advantages and economic benefits

The system provides for:

• a possibility to make general report and a list of key words reflecting main text content without considering information requirements of a user.
• a possibility to make topically oriented text of a report and a list of key words taking into account user specialization.

Technology differentiation and uniqueness

The system uses new approach of estimation of information content of words based on comparison of their statistical characteristics in annotated text and complete text frame.

Context in which technology was identified

The system put in experimental operation in the Presidential Library of the Republic of Belarus.

Technological keywords

Software, automatic indexing, automated abstracting, text-based document, informative word, informative sentence, informativeness.

Development Stage

• Applied research
• Experimental development stage

Intellectual property rights

Range of applications

Scientific-and-technical libraries. Information Analysis Centers of enterprises and organizations.

Classifier Used at the EU Innovation Relay Centres

• Electronics, IT and Telecommunication

Preferable Regions

• Europe
• Asia

Practical experience

Experimental operation of the system in the Presidential Library of the Republic of Belarus.

Environmental impact

None

Type of collaboration sought

• R&D contract

Terms and restrictions

— None

Available technical assistance

• Documentation