Aspect Based Sentiment Analysis task focuses on the recognition of aspect category and classification of emotions (positive, negative, neutral) in text. The aim of Aspect Based Sentiment Analysis is to determine the sentiment polarity expressed towards certain aspect. This system is based on supervised learning using Support Vector Machine (SVM). Multi label svc classifier is used to classify the laptop reviews into 20 aspect categories and also to classify the review as positive, negative or neutral. For each sentiment, a multi label svc classifier is used. Performance of the system is also compared with other classifiers.
Dataset is avaibale at SemEval website.
- Scikit Learn
- Nltk
- pandas
- numpy
P.S. This is a Mini-Project for a 2 month crash course (NLP) in the Final Year at the Department of Computer Science and Engineering, Kathmandu University.