Implementasi Metode Naïve Bayes Classifier Untuk Klasifikasi Sentimen Masyarakat Terhadap Layanan Indihome di Twitter

Rahmi, Fasya Byan (2023) Implementasi Metode Naïve Bayes Classifier Untuk Klasifikasi Sentimen Masyarakat Terhadap Layanan Indihome di Twitter. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10611910000053-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (4MB) | Request a copy

Abstract

PT. Telekomunikasi Indonesia meluncurkan sebuah produk teknologi internet yang canggih bernama Indihome. Indihome (Indonesia Digital Home) merupakan salah satu layanan triple play dari produk Telkom berupa paket layanan telekomunikasi data yaitu telepon rumah (voice), internet (internet on fiber), dan layanan TV (useetv cable). Produk Indihome memberikan kemudahan dalam mengakses internet dengan kecepatan yang cukup untuk digunakan oleh masyarakat Indonesia. Pandangan masyarakat merupakan sumber daya paling penting dan berharga bagi perusahaan dalam menilai produk mereka agar senantiasa tetap bertahan dalam menghadapi persaingan dengan kompetitor lain. Oleh karena itu, PT. Telekomunikasi Indonesia perlu mengetahui bagaimana opini masyarakat terkait layanan Indihome untuk mengembangkan dan meningkatkan kualitas layanan serta mengetahui kebutuhan pelanggan terhadap apa saja yang menjadi kekurangan dalam layanan Indihome. Penelitian ini bertujuan untuk mengetahui hasil analisis sentimen dan ketepatan klasifikasi sentimen masyarakat terhadap layanan Indihome di twitter. Hasil penelitian ini diharapkan dapat memberikan informasi kepada PT. Telkom Indonesia terkait layanan Indihome. Sehingga kebutuhan dari pelanggan dapat terpenuhi, dan tingkat kepercayaan pelanggan tetap terjaga. Hasil penelitian menunjukkan bahwa dari 17697 tweet didapatkan 36,3% tweet masuk ke dalam kategori sentimen positif dan 63,7% masuk ke dalam kategori sentimen negatif. Hasil ketepatan klasifikasi pada data training didapatkan nilai accuracy sebesar 51,49%, sensitivity sebesar 96,45%, specificity sebesar 25,91%, dan AUC sebesar 0,6118. Sedangkan pada data testing diperoleh nilai accuracy sebesar 51,09%, sensitivity sebesar 96,39%, specificity sebesar 23,51%, dan AUC sebesar 0,6085.
================================================================================================================================
PT. Telekomunikasi Indonesia launched a sophisticated internet technology product called Indihome. Indihome (Indonesia Digital Home) is a triple play service from Telkom products in the form of data telecommunication service packages, namely home telephone (voice), internet (internet on fiber), and TV services (useetv cable). Indihome products provide convenience in accessing the internet with sufficient speed for use by the Indonesian people. Public opinion is the most important and valuable resource for companies in assessing their products so that they can always survive in the face of competition with other competitors. Therefore, PT. Telekomunikasi Indonesia needs to know what public opinion is regarding Indihome services in order to develop and improve service quality and know customer needs for what are the deficiencies in Indihome services. This study aims to determine the results of sentiment analysis and the accuracy of the classification of public sentiment towards Indihome services on Twitter. The results of this study are expected to provide information to PT. Telkom Indonesia regarding Indihome services. So that the needs of customers can be met, and
the level of customer trust is maintained. The results showed that of the 17697 tweets, 36,3% of the tweets were in the positive sentiment category and 63,7% were in the negative sentiment category. The results of the classification accuracy on the training data obtained an accuracy value of 51,49%, a sensitivity of 96,45%, a specificity of 25,91%, and an AUC of 0,6118. While the testing data obtained an accuracy value of 51,09%, a sensitivity of 96,39%, a specificity of 23,51%, and an AUC of 0,6085.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Analisis Sentimen, Indihome, Naïve Bayes Classifier, Twitter
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD108 Classification (Theory. Method. Relation to other subjects ) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.5956 Quality of service. Reliability Including network performance
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Fasya Byan Rahmi
Date Deposited:	22 Feb 2023 06:31
Last Modified:	22 Feb 2023 06:42
URI:	http://repository.its.ac.id/id/eprint/97663

Actions (login required)

View Item