SentiWordNet (Indian Languages: Hindi, Bengali, Telugu, and Tamil)
This license allows you to use SentiWordNet (ILs) only for research, non-profit purposes. Any institution, be it academic or non-academic, is allowed to use SentiWordNet (ILs) for any non-profit research project.
Using SentiWordNet (ILs): Please cite the following references:
1. A. Das and B. Gambäck. Sentimantics: The Conceptual Spaces for Lexical Sentiment Polarity Representation with Contextuality, In the 3rd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis (WASSA), ACL 2012, Pages 38–46, Jeju, South Korea. pdf
2. A. Das and S. Bandyopadhyay. Dr Sentiment Knows Everything! ACL/HLT 2011 Demo Session, Pages 50-55, June, Portland, Oregon, USA. pdf
3. A. Das and S. Bandyopadhyay. SentiWordNet for Indian Languages, In the 8th Workshop on Asian Language Resources (ALR), COLING 2010, Pages 56-63, August, Beijing, China. pdf
File Details: <Lang Codes: HN: Hindi, BN: Bengali, TA: Tamil, TE: Telugu.
1. Lang_POS.txt: This file contains a list of POSITIVE sentiment words.
2. Lang_NEG.txt: This file contains a list of NEGATIVE sentiment words.
3. Lang_NEU.txt: This file contains a list of NEUTRAL sentiment words.
4. Lang_AMBI.txt: This file contains a list of AMBIGUOUS sentiment words.
Note: Positive, Negative, and Neutral are self-explanatory. There are many words, change polararity with context. For Example: Long could be positive or negative based on context
Long live <positive> Long road <negative> Long drive <positive> Long queue <negative>
These word are marked as ambiguous. For detailes please refer to the following paper:
A. Das and B. Gambäck. Sentimantics: The Conceptual Spaces for Lexical Sentiment Polarity Representation with Contextuality, In the 3rd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis (WASSA), ACL 2012, Pages 38–46, Jeju, South Korea. pdf
Lexicons are marked with POS categories:
a: Adjective n: Noun r: Adverb v: Verb u: unknown
N.B. Please fill out the form below with proper information to obtain SentiWordNet (ILs).