Nuansa, Eza Putra (2017) Analsis Sentimen Pengguna Twitter Terhadap Pemilihan Gubernur Dki Jakarta Dengan Metode Naïve Bayesian Classification Dan Support Vector Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1313100114-Undergraduate-Theses.pdf - Published Version Download (2MB) | Preview |
Abstract
Pada tahun 2017, DKI Jakarta akan melakukan pesta demokrasi yang tentunya berpengaruh terhadap dunia social media. Sayangnya, sebagian besar perbincangan di Twitter itu adalah bentuk serangan verbal yang sering menggunakan kata-kata kasar dan menghembuskan isu sensitif seperti agama serta etnis untuk menyerang kandidat lain. Sehingga, twitter sangat cocok untuk dijadikan sumber analisa sentimen dan opinion mining karena penggunaan Twitter untuk mengekspresikan opini mereka terhadap berbagai topik.
Penelitian ini bertujuan untuk mencari kata kunci dan hubungan pola antar tiap calon Gubernur DKI Jakarta. Metode Naive Bayes Classifier pada masing-masing pengukuran performa akurasi, precision, recall, dan F-Measure sebesar 85.77%; 85.90%; 85.77%; 85.67%. Metode Support Vector Machine kernel RBF tiap pengukuran performa akurasi, precision, recall, dan F-Measure adalah 87.80%; 98.48%; 87.80%; 92.64%. Untuk hasil SNA didapatkan hasil yang tinggi untuk degree centrality sebeasar 0.865 yang menunjukan pengaruh antar node kata kunci dengan kata kunci yang lain.
=================================================================
In 2017, DKI Jakarta will celebrate a democracy party which certainly affects the netizen in social media. Unfortunately, most of the conversation on Twitter is a form of verbal attack that often uses harsh words and exhaling sensitive issues such as religion and ethnicity to attack other candidates. Thus, twitter is perfect for being a source of sentimental and opinion mining analysis because of the use of Twitter to express their opinions on various topics.
This study to find the keyword and pattern relationship between each candidate of Governor of DKI Jakarta. Method of Naive Bayes Classifier in each measurement of accuracy, precision, recall, and F-Measure of 85.77%; 85.90%; 85.77%; 85.67%. Support Method Vector Machine RBF kernel for each measurement of accuracy, precision, recall, and F-Measure performance is 87.80%; 98.48%; 87.80%; 92.64%. For Social Network Analysis results obtained high results for degree centrality 0.865 which shows the influence of keyword nodes with other keywords.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, Naïve Bayes Classi-fication, Pemilihan Gubernur DKI Jakarta, Social Network Analysis, Support Vector Machine |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Eza Putra Nuansa |
| Date Deposited: | 20 Oct 2017 07:41 |
| Last Modified: | 05 Mar 2019 08:41 |
| URI: | http://repository.its.ac.id/id/eprint/47741 |
Actions (login required)
 |
View Item |