TY - GEN
T1 - Sentiment Analysis of Banking Risk Profile Determination
T2 - 3rd International Conference on Information Technology and Advanced Mechanical and Electrical Engineering, ICITAMEE 2022
AU - Yuniarti, Rina
AU - Ruldeviyani, Yova
N1 - Publisher Copyright:
© 2023 American Institute of Physics Inc.. All rights reserved.
PY - 2023/12/28
Y1 - 2023/12/28
N2 - Determination of the risk profile at Bank XYZ is carried out periodically. It can be weekly, monthly, and quarterly. This risk profile itself is an information that explains the risk conditions in banking activities, from which information becomes the basis for determining risk control. One of the parameters of the risk profile is negative sentiment from customers and the wider community. The large number of negative sentiments, the higher the risk that the bank has, in this case it is related to reputation risk. In addition, monitoring negative sentiment on social media is very necessary because it affects the implementation of tasks in IT work units and service work units, especially those related to complaint handling activities originating from social media. Based on the above conditions, this study discusses how to analyze the sentiments of customers on social media which are busy with pros and cons opinions so that they can be monitored, and problems resolved immediately. This sentiment data analysis uses the Lexicon-Based method. Sentiment analysis results are used to determine the score of one of the parameters in the reputation risk profile. This is adjusted to the provisions of the Financial Services Authority (OJK) in determining the soundness of the Bank. From the analysis results, the distribution of sentiment results is 11.95% positive sentiment, 45.97% negative sentiment, and 42.08% neutral sentiment. From these results, a comparison is made with the risk profile matrix and the frequency of negative sentiments is included in the High category.
AB - Determination of the risk profile at Bank XYZ is carried out periodically. It can be weekly, monthly, and quarterly. This risk profile itself is an information that explains the risk conditions in banking activities, from which information becomes the basis for determining risk control. One of the parameters of the risk profile is negative sentiment from customers and the wider community. The large number of negative sentiments, the higher the risk that the bank has, in this case it is related to reputation risk. In addition, monitoring negative sentiment on social media is very necessary because it affects the implementation of tasks in IT work units and service work units, especially those related to complaint handling activities originating from social media. Based on the above conditions, this study discusses how to analyze the sentiments of customers on social media which are busy with pros and cons opinions so that they can be monitored, and problems resolved immediately. This sentiment data analysis uses the Lexicon-Based method. Sentiment analysis results are used to determine the score of one of the parameters in the reputation risk profile. This is adjusted to the provisions of the Financial Services Authority (OJK) in determining the soundness of the Bank. From the analysis results, the distribution of sentiment results is 11.95% positive sentiment, 45.97% negative sentiment, and 42.08% neutral sentiment. From these results, a comparison is made with the risk profile matrix and the frequency of negative sentiments is included in the High category.
UR - http://www.scopus.com/inward/record.url?scp=85182356093&partnerID=8YFLogxK
U2 - 10.1063/5.0181966
DO - 10.1063/5.0181966
M3 - Conference contribution
AN - SCOPUS:85182356093
T3 - AIP Conference Proceedings
BT - AIP Conference Proceedings
A2 - Jusman, Yessi
A2 - Zaki, Ahmad
A2 - Damarjati, Cahya
PB - American Institute of Physics Inc.
Y2 - 20 July 2022 through 21 July 2022
ER -