Diskoveror-TA-engine

The DisKoveror Text analytics engine(DTAE) is a software product developed by Serendio for extracting information such as Sentiment, Text categories, named entities etc from unstructured text. DTAE is useful in extracting information from Twitter streams, product reviews, emails, webpages and any other source of unstructured information.

The workspace contains the core Text analytics engine which in turn optimally leverages multiple open source packages to give several results structed and presented in a clean user-friendly format.

Features

• Sentiment Extraction
• Topic detection
• Named Entity Recognition
• Java based API
• Coreference Resolution
• Sentiment Polarity

System Architecture

The intended architecture of the system is as given below.

The diskoveror-ta-engine leverages modules under the below independent categories :

Opensource softwares

• Stanford CoreNLP
• Apache OpenNLP
• DUKE

Serendio proprietary softwares

• Topic Modeling Algorithm
• Sentiment Analysis Algorithm

Serendio domain specific softwares

• Life Science Ontologies
• Legal Ontologies

Third party API/Software integrations

New modules could be supported under these categories without disturbing the existing system in any manner.

Compiling Diskoveror TA Engine

To package it in a single executable jar for distribution (.jar file), the following command has to be run from the command line.

  mvn clean compile assembly:single

Example usage from Tamanager

Software requirements

JDK

  Version 7 build 79 or above

Maven

  Apache Maven 3.0.5 or above

Thrift servers

Get Thrift Here