SUMMA is a text summarization toolkit which follows the architectural precepts of the GATE framework therefore providing much needed functionality in the form of language and processing resources for composition of practical summarization applications.

BACKGROUND

Our lives are saturated with information from multiple channels such as news agencies, mails, tweets, commentaries, social networks, etc. Automatic Text Summarization can help extract, merge, and condense what is important from a stream of textual information reducing information overload. There is no implemented summarizer which can summarize all types of document, therefore the need for a toolkit which allows composition to customize specific summatization applications.

THE TECHNOLOGY

SUMMA main features:

• Resources for statistical text analysis
• Resources for features computation
• Resources for customization of summaries
• Resources for exporting results
• Single-document summarization
• Multi-document summarization
• Multilingual processing
• Ready-made summarizers and baselines for research comparison
• Easy to install and use
• Easy to extend
• Easy to customize

ADVANTAGES

• Implemented algorithms for creating summarization systems
• Multi-platform
• Multi-lingual

STATE OF DEVELOPMENT

Fully developed.

INTELLECTUAL PROPERTY

©UPF 2014, software registered.

MARKET OPPORTUNITY

Customizable summarization applications for government and private institutions. Software distribution and consultancy.

COMMERCIAL OPPORTUNITY

Software available for licensing with technical cooperation.

CONTACT

Marc Santandreu

Technology Transfer Unit

(+34) 93 542 2896

[email protected]

KEYWORDS

Text Processing, Summarization, Information Extraction.