Sentiment Analysis on Viral Issues in Social Media

HASMAH, ZANUDDIN and NURUL, HIDAYAH (2016) Sentiment Analysis on Viral Issues in Social Media. In: Fifth International Conference On Advances In Economics, Social Science and Human Behaviour Study - ESSHBS 2016, 25-26 September 2016, Bangkok, Thailand.

[img]
Preview
Text
20161007_093109.pdf - Published Version

Download (219kB) | Preview
Official URL: https://www.seekdl.org/conferences/paper/details/8...

Abstract

Social media is the most effective medium of communication bilaterally over online. Facebook is a platform of choice for a variety of purposes such as communication with family and friends, developing or establishing business and spreading the influence of ideology to attract more followers. Two-way communication in social media is able to effectively promote information sharing, but if the sharing involves non authentic or misinformation disseminated widely, it may create unhealthy situation and potentially threaten the public order. This quantitative content analysis study investigated the influence and relevance of information sources on the distribution of public opinion through the comments on each Facebook’s post studied using sentiment analysis API webtools. Through sentiment analysis software, sentiment for each comment on the selected Facebook sentiment polarity of the postings will be categorized into positive, negative and neutral sentiment. There were 4 parameters measured as independent variables, namely independent sources, media sources, public opinion source and source from monitoring organization to the dependent variable, public opinion. The case studies were selected from posting related to public order issue. The sample was selected using multistage sampling techniques involves purposive sampling and systematic sampling. The data obtained were analyzed using non-parametric statistical analysis. The results obtained through the five stages of data collection starting from Data Extraction, Language Detection, Data Cleaning and Pre-processing, Translating API and last but not least Sentiment Analysis. The polarity of the sentiment were then categorized into a few categories to investigate further the critical values on the sentiment reveal by the social media users in their comments on a specific facebook posting of the case study. The critical values such as sentiment emotion, comment direction and sensitivity/negativity level could help monitoring bodies to enhance their monitoring effectiveness toward controversial issues and strategizing the action to overcome any unexpected situation before it occurred.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Public Opinion, Viral Issues, Social Media, Media Surveillance, Viral Postings, Sentiment Analysis, API Webtools
Depositing User: Mr. John Steve
Date Deposited: 20 Mar 2019 11:32
Last Modified: 20 Mar 2019 11:32
URI: http://publications.theired.org/id/eprint/697

Actions (login required)

View Item View Item