Perbandingan Algoritma Naïve Bayes Classifier Dan Support Vector Machine Untuk Analisis Kinerja Pelayanan PT. Telkom Indonesia Di Social Media Twitter (Studi Kasus : Indihome)

Ulung, Priyo Bintoro (2021) Perbandingan Algoritma Naïve Bayes Classifier Dan Support Vector Machine Untuk Analisis Kinerja Pelayanan PT. Telkom Indonesia Di Social Media Twitter (Studi Kasus : Indihome). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text 1. COVER.pdf Download (1MB)
	Text 2. Abstrak .pdf Download (113kB)
	Text 3. Abstract .pdf Download (91kB)
	Text 4. BAB I.pdf Download (170kB)
	Text 5. BAB II.pdf Download (1MB)
	Text 6. BAB III.pdf Download (246kB)
	Text 7. BAB IV.pdf Restricted to Registered users only Download (2MB) | Request a copy
	Text 8. BAB V.pdf Download (90kB)
	Text 9. DAFTAR PUSTAKA.pdf Download (125kB)
	Text 10. LAMPIRAN.pdf Restricted to Registered users only Download (132kB) | Request a copy

Abstract

PT Telkom Indonesia Tbk is a state-owned company engaged in information and communication as well as a complete telecommunications service and network provider. Indihome service is one of the products from PT. Telkom is widely used by the community. Of course, with the various products provided, there are often several problems for its users, and customer service plays an important role in the development of service results from Indihome. Indihome itself uses Social Media Twitter as a means to receive customer complaints. Various complaints on Social Media Twitter then carry out data processing or what is known as Data Mining. Support Vector Machine and Naive Bayes Classifier are popular algorithm methods used in Data Mining. Support Vector Machine is a set of guided learning methods that analyze data and recognize patterns, used for classification and regression analysis. The Support Vector Machine (SVM) tries to find the best hyperplane in the input space. The Naïve Bayes Classifier includes a "probabilistic classifier" model which is based on the Bayesian theorem. Parameter estimation for the Naïve Bayes model using the maximum likelihood method. By applying the SVM and NBC methods. Comparison of the performance of the classification results that have been carried out for analysis of service performance of PT. Telkom Indonesia uses 340 sentiment data consisting of 170 having a positive sentiment label of 170 and a negative sentiment label, namely the SVM method gets a higher level of accuracy than the NBC method with the final result of an accuracy value of 0.85 or 85% while NBC gets the final result. an accuracy value of 0.80 or 80%. Keywords: NBC, SVM, Indihome, Classification

Actions (login required)

View Item