TY - GEN
T1 - Performance of DKI Jakarta governor and vice governor on 2017-2018 based on sentiment analysis using twitter and instagram data
AU - Julian, Bintang Glenn
AU - Budi, Indra
AU - Tanaya, Dipta
N1 - Funding Information:
We gratefully thank the Universitas Indonesia for the International Publication Grants (Hibah PIT-9) No. NKB-0010/UN2.R3.1/HKP.05.00/2019
Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/7/19
Y1 - 2019/7/19
N2 - Sentiment analysis is one of the topics that recently getting more popular on political field or government-related things. Analyzing citizens' view of the government, including Governor and Vice Governor of DKI Jakarta for 2017-2022 period, is one of tasks that can be done using sentiment analysis. Data related to that topic ar e gathered from Twitter and Instagram for further analysis. N-gram, emoji, and all-caps is used as features to classify sentiment of each item. Based on the experiment, those features can help to increase classification performance. Naïve Bayes, Random Forest, and SVM algorithm are compared to select the best algorithm out of those three algorithms. Based on the experiment, SVM get the best result with highest accuracy and F1-score on both domains. The result of the classification shows that citizens tend to have neutral view on Governor and Vice Governor of DKI Jakarta for 2017-2022 period on their first year of governance. In addition, there are more positives than negatives on citizens' view.
AB - Sentiment analysis is one of the topics that recently getting more popular on political field or government-related things. Analyzing citizens' view of the government, including Governor and Vice Governor of DKI Jakarta for 2017-2022 period, is one of tasks that can be done using sentiment analysis. Data related to that topic ar e gathered from Twitter and Instagram for further analysis. N-gram, emoji, and all-caps is used as features to classify sentiment of each item. Based on the experiment, those features can help to increase classification performance. Naïve Bayes, Random Forest, and SVM algorithm are compared to select the best algorithm out of those three algorithms. Based on the experiment, SVM get the best result with highest accuracy and F1-score on both domains. The result of the classification shows that citizens tend to have neutral view on Governor and Vice Governor of DKI Jakarta for 2017-2022 period on their first year of governance. In addition, there are more positives than negatives on citizens' view.
KW - All-caps
KW - Emoji
KW - Governor
KW - N-gram
KW - Sentiment analysis
UR - http://www.scopus.com/inward/record.url?scp=85072808849&partnerID=8YFLogxK
U2 - 10.1145/3352411.3352431
DO - 10.1145/3352411.3352431
M3 - Conference contribution
AN - SCOPUS:85072808849
T3 - ACM International Conference Proceeding Series
SP - 122
EP - 127
BT - Proceedings of the 2019 2nd International Conference on Data Science and Information Technology, DSIT 2019
PB - Association for Computing Machinery
T2 - 2nd International Conference on Data Science and Information Technology, DSIT 2019
Y2 - 19 July 2019 through 21 July 2019
ER -