Pakpahan, Swandy Raja Manaek (2019) Analisis Sentimen Tentang Opini Performa Klub Sepak Bola Pada Dokumen Twitter Menggunakan Support Vector Machine Dengan Perbaikan Kata Tidak Baku. Sarjana thesis, Universitas Brawijaya.
Abstract
Sepak Bola merupakan salah satu olahraga paling diminati masyarakat dunia, termasuk Indonesia. Sebuah klub sepak bola sangat bergantung pada pendukungnya sehingga kepuasan pendukung sebuah klub sepak bola harus dapat dijaga. Pendukung klub sepak bola sendiri cukup sering memberikan argumen terhadap sebuah klub sepak bola melalui media Twitter. Oleh karena itu, penulis mengusulkan penelitian untuk membangun sistem analisis sentimen terhadap opini performa klub sepak bola pada dokumen Twitter. Penelitian ini menggunakan metode Support Vector Machine dan perbaikan kata Levenshtein Distance. Tahapan yang dilakukan dimulai dengan melakukan preprocessing pada data, kemudian melakukan perbaikan kata dengan Levenshtein Distance, pembobotan Term Frequency-Inverse Document Frequency, dilanjutkan dengan melakukan klasifikasi menggunakan Support Vector Machine. Hasil pengujian dengan nilai tertinggi didapatkan sebesar 83,25% dengan parameter learning rate = 0,0001, complexity = 0,001, lambda = 0,1, epsilon = 0,0001 dan nilai iterasi maksimal = 50
English Abstract
Football is one of the most popular sports in the world, including in Indonesia. A football club is very dependent on its supporters so that the satisfaction of supporters of a football club must be maintained. Supporters of football clubs themselves often provide arguments to a football club via Twitter media. Therefore, the authors propose research to build a sentiment analysis system for football club performance opinions on Twitter documents. This research uses the Support Vector Machine method and Levenshtein Distance for non-standard word correction.The process starts with preprocessing the data, then do word correction with Levenshtein Distance, weighting using Term Frequency-Inverse Document Frequency, followed by classification using Support Vector Machine. The test results with the highest accuracy were obtained at 83.25% with learning rate = 0,0001, complexity = 0,001, lambda = 0,1, epsilon = 0,0001 and maximum iteration = 50
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FILKOM/2019/281/051905599 |
| Uncontrolled Keywords: | sepak bola, twitter, analisis sentimen, support vector machine, perbaikan kata, levenshtein distance/football, twitter, sentiment analysis, support vector machine, word correction, levenshtein distance |
| Subjects: | 000 Computer science, information and general works > 004 Computer science > 004.2 System analysis and design, computer architecture, performance evaluation |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Endang Susworini |
| Date Deposited: | 30 Jul 2020 06:04 |
| Last Modified: | 30 Jul 2020 06:04 |
| URI: | http://repository.ub.ac.id/id/eprint/171809 |
Actions (login required)
 |
View Item |