[TIKA-2263] Sentiment Analysis - Standardize the interface and integrate OpenNLP, CoreNLP with pluggable models

• Type: Improvement
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Labels: NLP classifier integration machine_learning
• Assignee: Unassigned
• Reporter: Thamme Gowda
• Created: 11/Feb/17 22:18
• Updated: 11/Feb/17 22:18

Description

Background

As per Wikipedia: Sentiment analysis (also known as opinion mining) refers to the use of natural language processing, text analysis and computational linguistics to identify and extract subjective information in source materials. Sentiment analysis is widely applied to reviews and social media for a variety of applications, ranging from marketing to customer service.

Goals

• Create an interface for SentimentAnalyser
• Create a parser for sentiment analysis with pluggable implementations of SentimentAnalyser and its model files
• Provide an implementation from Apache OpenNLP and its default models for english. Make this as default
  • Converge the efforts of chrismattmann and [e-mail] in TIKA-2016
• Provide an implementation from Stanford CoreNLP and make this as optional (due to its license incompatibility wit AL2.0). Make it attachable at runtime similar to already existing CoreNLP Named Entity Recogniser.
  • Leave room for plugging other/custom implementations at the runtime

References

• https://issues.apache.org/jira/browse/TIKA-2016
• https://issues.apache.org/jira/browse/TIKA-1787
• http://irds.usc.edu/SentimentAnalysisParser/index.html
• http://nlp.stanford.edu/sentiment/