I'm evaluating the use of BigInsights for sentiment analysis of text in English and Italian. I've seen that with SystemT is possible to do the Named Entity Recognition using the file serverSideNEFinancial_BI.ao) on a text in these and other languages, and inside the JARs of BigInisghts there's some code for POS tagging.
What are the other tools available in BigInsights (version 1.4 Enterprise) for text analytics, in English and Italian ? In the documentation at http://publib.boulder.ibm.com/infocenter/bigins/v1r2/topic/com.ibm.swg.im.infosphere.biginsights.doc/doc/textanalyticsoverview.html
I read, among initial bullet points
"A rich library of pre-compiled extractors for high-interest types of entities and relationships."
what is this referring to? It is the NER extractor or are there other tools ?
Re: Text Analitics tool2012-10-31T21:04:42ZThis is the accepted answer. This is the accepted answer.Hi Jacopo,
Below pasting the response I got from our dev:
"Correct, that text refers to the pre-compiled extractor libraries. The precompiled extractor for named entities in Western languages have support primarily for English, but also have more limited support for Italian. Have you given them a try ?
If the pre-compiled extractors are not suitable for your task, the BigInsights Eclipse Tools provide a full-fledged development environment for developing extractors for any extraction task, and a multitude of languages, using the state of the art AQL language. For a brief introduction to Text Analytics and BigInsights Eclipse Tools, have a look at these YouTube channel:
Multilingual support of the Text Analytics system and the AQL language is explained in Reference > Text Analytics > Multilingual Support for Text Analytics. For Italian, morphological support includes both tokenization and part-of-speech analysis."