TY - GEN
T1 - Sentiment Analysis of 5G Network Implementation in Indonesia Using Twitter Data
AU - Komang Ananta Aryadinata, I.
AU - Pangesti, Dyah
AU - Anugerah, Gilbert Badia
AU - Aditya, Ivan Eka
AU - Ruldeviyani, Yova
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - 5G network is the evolution of 4G LTE network. One of the telecommunication providers in Indonesia has won the bid to implement the 5G network and successfully done the operational feasibility test. Currently it is the pioneer commercial use phase in Indonesia using the N40 band at 2300 MHz frequency. This research discusses how the social media users' sentiment toward the implementation of 5G network, the discussion will be divided into four main topics that are 5G network bid, 5G implementation, telecommunication provider's network quality, and customer service interaction. Prediction model is made based on the comparison between three methods that are naive bayes, support vector machine, and fast large margin. Predictive model built with fast large margin method resulting the highest accuracy (93.29%) and chosen as the reference model for sentiment analysis. Sentiment analysis is done to a data set containing 1620 tweets that are categorized to the four main topics. Topics that have significant positive sentiments are 5G network bid (98.76%), 5G network implementation (98.35%), and customer service interaction (96.73%). Topic with quite many negative sentiments is about telecommunication provider's network quality (41.31%). The results also expected to discover possible corrective actions that should be done by telecommunication provider.
AB - 5G network is the evolution of 4G LTE network. One of the telecommunication providers in Indonesia has won the bid to implement the 5G network and successfully done the operational feasibility test. Currently it is the pioneer commercial use phase in Indonesia using the N40 band at 2300 MHz frequency. This research discusses how the social media users' sentiment toward the implementation of 5G network, the discussion will be divided into four main topics that are 5G network bid, 5G implementation, telecommunication provider's network quality, and customer service interaction. Prediction model is made based on the comparison between three methods that are naive bayes, support vector machine, and fast large margin. Predictive model built with fast large margin method resulting the highest accuracy (93.29%) and chosen as the reference model for sentiment analysis. Sentiment analysis is done to a data set containing 1620 tweets that are categorized to the four main topics. Topics that have significant positive sentiments are 5G network bid (98.76%), 5G network implementation (98.35%), and customer service interaction (96.73%). Topic with quite many negative sentiments is about telecommunication provider's network quality (41.31%). The results also expected to discover possible corrective actions that should be done by telecommunication provider.
KW - 5G
KW - fast large margin
KW - sentiment analysis
KW - telecommunication
KW - twitter
UR - http://www.scopus.com/inward/record.url?scp=85124366371&partnerID=8YFLogxK
U2 - 10.1109/IWBIS53353.2021.9631863
DO - 10.1109/IWBIS53353.2021.9631863
M3 - Conference contribution
AN - SCOPUS:85124366371
T3 - Proceedings - IWBIS 2021: 6th International Workshop on Big Data and Information Security
SP - 23
EP - 28
BT - Proceedings - IWBIS 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 6th International Workshop on Big Data and Information Security, IWBIS 2021
Y2 - 23 October 2021 through 26 October 2021
ER -