Applying Machine Learning Algorithms for Opinion Mining on a Digital Internet Service Provider

Amalia Utamima*, Farrel Istihsan Aditya, Firdaus W.G. Ashaari, Alfito K. Nugraha

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

By.U is the first digital internet service provider in Indonesia that provides complete digital services for all telecommunication requirements. The By.U application on the Google Play Store has garnered many ratings and comments from users. However, the unstructured nature of these reviews makes it challenging to extract and classify valuable information, which is why opinion mining is preferred. This research compares three machine learning algorithms, namely K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Naive Bayes Classifier (NBC), to conduct opinion mining. The research findings suggest that the KNN outperforms the NBC and SVM methods in terms of accuracy, precision, and recall.

Original languageEnglish
Title of host publicationISDFS 2023 - 11th International Symposium on Digital Forensics and Security
EditorsAsaf Varol, Murat Karabatak, Cihan Varol, Ahad Nasab
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350336986
DOIs
Publication statusPublished - 2023
Event11th International Symposium on Digital Forensics and Security, ISDFS 2023 - TN, United States
Duration: 11 May 202312 May 2023

Publication series

NameISDFS 2023 - 11th International Symposium on Digital Forensics and Security

Conference

Conference11th International Symposium on Digital Forensics and Security, ISDFS 2023
Country/TerritoryUnited States
CityTN
Period11/05/2312/05/23

Keywords

  • K nearest neighbor
  • Naïve Bayes
  • Support Vector Machine
  • class
  • opinion mining

Fingerprint

Dive into the research topics of 'Applying Machine Learning Algorithms for Opinion Mining on a Digital Internet Service Provider'. Together they form a unique fingerprint.

Cite this