ATA (system): Difference between revisions

From HLT@INESC-ID

No edit summary
 
 
(One intermediate revision by the same user not shown)
Line 5: Line 5:


== Features ==
== Features ==
Team / Authors
:Team / Authors
* Joana L. Paulo
:* [[Joana Paulo Pardal|Joana L. Paulo]]
 
:Platforms
Platforms
:* Windows / Linux / MacOSX (as long as all the tools needed are still available in both platforms)
* Windows / Linux / MacOSX (as long as all the tools needed are still available in both platforms)
:Developing status
 
:* Stable
Developing status
:* Project ended in 2004/07
* Stable
* Project ended in 2004/07


[[category:Tools]]
[[category:Tools]]

Latest revision as of 18:34, 25 June 2008

The development of terminologies is a hard work, especially when it is done manually. ATA eases this task by extracting the terms contained in a specialized text. The main purpose of this tool is to suggest a list of words likely to be terms giving the linguistic and statistic information collected about them. The results can be used for example in topic detection.

Goals

The main goal of the ATA project is to automatically extract the terms in a specialized text. To do this, the text is feed into a chain of NLP tools: SMorph, PAsMo, MARv, SuSAna, GeTerms, Anota, PrintTerms. The first four modules will do the linguistic analysis. The two in the middle are responsible for the statistical enrichments of the text, counting the occurrences of each candidate. Also they add information about the word count on reference text. The last module is responsible for the decision on which words to suggest.

Features

Team / Authors
Platforms
  • Windows / Linux / MacOSX (as long as all the tools needed are still available in both platforms)
Developing status
  • Stable
  • Project ended in 2004/07