Lexicon Generation Using Genetic Algorithm for Aspect-Based Sentiment Analysis
| Autor: | Hamidreza Keshavarz, Mohammad Erfan Mowlaei, Mohammad Saniee Abadeh |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Polarity (physics)
business.industry Computer science Sentiment analysis Feature extraction 020207 software engineering 02 engineering and technology Lexicon computer.software_genre Domain (software engineering) Task (project management) Genetic algorithm 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Social media Artificial intelligence business computer Natural language processing |
| Zdroj: | 2018 IEEE 22nd International Conference on Intelligent Engineering Systems (INES). |
| DOI: | 10.1109/ines.2018.8523902 |
| Popis: | Sentiment analysis is the task of extracting and analyzing opinions expressed in comments written in social media and websites and is performed to assist users and stakeholders to better understand the public opinion on a subject. In this paper, the main contribution is a lexicon generation method using genetic algorithm for aspect level sentiment analysis. The proposed method produces a lexicon which can be used in polarity detection of aspects in written comments and reviews. The proposed method is an extension over the ALGA algorithm developed to work on tweet level and in this paper, it is modified to fit in aspect level problems. The results of experiment on the SemEval-2014 dataset proves that the proposed lexicon can improve F-measure of polarity detection on subjective comments by at least 1.9 percentage points in Laptop domain compared to prominent lexicons such as AFINN and NRC. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|