SENTIMENT ANALYSIS PADA MIKROBLOG MENGGUNAKAN NAIVE BAYES MENGENAI PELAYANAN JASA KEMKOMINFO
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
KYANA, NI LUH GEDE SARI MARTA
Arianto, Rakhmat
Siregar, Riki Ruli A
Arianto, Rakhmat
Siregar, Riki Ruli A
Subject
Teknik Informatika
Datestamp
2022-09-23 02:42:48
Abstract :
To improve service to the community, a government agency, namely the Ministry of Communication and Information (Kemkominfo), needs feedback from the community. One way to get feedback from the public is to use social media. Kemkominfo has provided the media, one of them Mikroblog Twitter to accommodate the public participation in the form of complaints, information, and suggestions on service process (policy, regulation, work program). Therefore, a classification system was developed using the Naïve Bayes Clasifier method to help determine the sentiments on the ¬tweet that the community sent on twitter @Kemkominfo account. Naive Bayes Clasifier is a simple probabilistic classification that calculates a set of probabilities by adding up the frequency and combination of values from a given dataset. Classification is divided into three groups, namely, positive, negative and neutral. In this study the development of the system uses CRISP-DM so that the work becomes more ordered and the test is done by the Confusion Matrix method to test the level of classification accuracy. From this research, the result of testing with Confusion Matrix shows 81.21% accuracy where precision and recall value is 0.80 and 0.89. These results, proving that the Naïve Bayes Classifier method can be used to assist in the classification of tweets based on Kemkominfo services, because it produces a high level of accuracy, so it can help in improving the performance of Kemkominfo.
To improve service to the community, a government agency, namely the Ministry of Communication and Information (Kemkominfo), needs feedback from the community. One way to get feedback from the public is to use social media. Kemkominfo has provided the media, one of them Mikroblog Twitter to accommodate the public participation in the form of complaints, information, and suggestions on service process (policy, regulation, work program). Therefore, a classification system was developed using the Naïve Bayes Clasifier method to help determine the sentiments on the ¬tweet that the community sent on twitter @Kemkominfo account. Naive Bayes Clasifier is a simple probabilistic classification that calculates a set of probabilities by adding up the frequency and combination of values from a given dataset. Classification is divided into three groups, namely, positive, negative and neutral. In this study the development of the system uses CRISP-DM so that the work becomes more ordered and the test is done by the Confusion Matrix method to test the level of classification accuracy. From this research, the result of testing with Confusion Matrix shows 81.21% accuracy where precision and recall value is 0.80 and 0.89. These results, proving that the Naïve Bayes Classifier method can be used to assist in the classification of tweets based on Kemkominfo services, because it produces a high level of accuracy, so it can help in improving the performance of Kemkominfo.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
