Sentiment Analysis of Social Networking Websites using Gravitational Search Optimization Algorithm

Sentiment Analysis of Social Networking Websites using Gravitational Search Optimization Algorithm

Lavika Goel, Anubhav Garg
Copyright: © 2018 |Pages: 10
DOI: 10.4018/IJAEC.2018010105
(Individual Articles)
No Current Special Offers


Analysing sentiments of various online communities have become now an interesting topic of research and industry. The behaviour of online communities resembles that of a swarm. This article presents a Gravitational Search algorithmic approach for sentiment analysis of online communities, and an optimization algorithm which is based on the mass interactions and law of gravity. In the end, the authors present comparisons with other techniques, particularly ant colony optimization and Naive Bayes classification for sentiment analysis.
Article Preview

2. Literature Survey

The study of content and analyzing it has a long history, dating back to 1966. Analysing reviews is different from that of analyzing posts as negative or positive. Tweets from Twitter has been a source of data for many researchers. In the past, sentiment analysis and opinion mining were used as synonyms. Das and Chen (2007) used the term first in the paper on market sentiment analysis. In 2002, ACL published a paper in which the term appeared.

Nasukawa et al. (2003) paper, “Sentiment analysis: Capturing favorability using natural language processing”, also a paper named “Sentiment Analyzer: Extracting sentiments about a given topic using natural language processing techniques.” Some researchers use the term most commonly to predict the polarity of a given piece of text, thereby referring the phrase for this task only. However, the term evolved more broadly to mean the subjectivity in the text and computational treatment of opinion, sentiment.

Go et al. (2009) used distant supervision in their paper on Twitter sentiment analysis. They have shown using Naive Bayes, Maximum entropy and SVM and achieved accuracy up to 80%.

Kun-Lin Liu et al. (2013) in their paper on Twitter sentiment analysis integrated noisy labelled data for training using emoticon smoothed language model (ESLAM) and used this for the unsupervised machine learning problem of classification.

Pak and Paroubek (2010) in their paper suggested a method for collecting corpus automatically from microblogs and build a sentiment classifier from it. In this instance, Twitter is used to gather corpus. The authors have used only English language, but their claim is that it can be used for multiple languages.

Stylios et al. (2014) first used the swarm intelligence algorithms in their work and compared the results with that of traditional methods like decision trees for evaluating the polarity sentences. Their result showed drastic improvements as compared to traditional methods.

Tumasjan et al. (2010) examined 100,000 tweets for election results. The highlight of their paper was that forty percent of the messages were posted by only four percent of the subscribers.

Goel (2011) used Ant Colony Optimization as a swarm intelligence algorithm for sentiment analysis, modelling it as a problem of social swarming resembling with that of ants.

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing