Open AccessImplementation of Machine Learning for Sentiment Analysis of Social and Political Orientation in Pekanbaru City
Author(s) -
Zul Indra,
Azhari Setiawan,
Yessi Jusman
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1803/1/012032
Subject(s) - sentiment analysis , biology and political orientation , social media , politics , public opinion , computer science , social network (sociolinguistics) , preprocessor , data pre processing , orientation (vector space) , big data , data science , artificial intelligence , world wide web , political science , data mining , geometry , law , mathematics
Nowadays, people are free to express their opinions regarding a problem in online social networks. One of most popular social network used to express opinions is Twitter. Public opinion in online social network has become a new source of big data that is interesting to be investigated. Opinion expressed by the public through social media is valuable data that can be further processed by using natural language processing (NLP). This research is expected to explain the social, economic, and political orientation of the people of Pekanbaru city by utilizing NLP algorithm. In addition, in terms of data sources, similar research is dominated by national studies, a little local. This research used Sentiment Analysis of Natural Language Processing (NLP) algorithm to analyze Pekanbaru citizen’s views and perceptions about social and political issues. The methods consist of: (i) data collection, (ii) data preprocessing, and (iii) sentiment classification. Thousands of tweets were extracted from Twitter API platform as research samples. As a result, the research has obtained 833 tweets about social orientation and 156 tweets about political trends. Overall, our tweets mined data were dominated with positive sentiments (53%). Education was the topic with most positive sentiments (42%) while political figure was the topic with most neutral sentiments (65%) and environment was the topic with most negative sentiments (56%). Regarding the discourses, “sampah” (garbage, waste, trash, etc) was the most posted and discussed in Twitter along with floods and air pollution topics.