Analisis sentimen pindah Ibu Kota Negara berbasis text mining menggunakan algoritma K-Nearest Neighbors dan Naive Bayes

Gama, Fatwa Penata (2022) Analisis sentimen pindah Ibu Kota Negara berbasis text mining menggunakan algoritma K-Nearest Neighbors dan Naive Bayes. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

[img]
Preview
Text (COVER)
1_Cover.pdf

Download (55kB) | Preview
[img]
Preview
Text (ABSTRAK)
2_Abstrak.pdf

Download (30kB) | Preview
[img]
Preview
Text (DAFTAR ISI)
3_Daftar Isi.pdf

Download (58kB) | Preview
[img]
Preview
Text (BAB I)
4_BAB1.pdf

Download (205kB) | Preview
[img] Text (BAB II)
5_BAB2.pdf
Restricted to Registered users only

Download (225kB) | Request a copy
[img] Text (BAB III)
6_BAB3.pdf
Restricted to Registered users only

Download (58kB) | Request a copy
[img] Text (BAB IV)
7_BAB4.pdf
Restricted to Registered users only

Download (714kB) | Request a copy
[img] Text (BAB V)
8_BAB5.pdf
Restricted to Registered users only

Download (370kB) | Request a copy
[img] Text (BAB VI)
9_BAB6.pdf
Restricted to Registered users only

Download (28kB) | Request a copy
[img] Text (DAFTAR PUSTAKA)
10_Daftar Pustaka.pdf
Restricted to Registered users only

Download (160kB) | Request a copy

Abstract

INDONESIA : Kebijakan pemerintah terhadap pemindahan ibu kota negara menimbulkan perdebatan pada kalangan masyarakat yang pro maupun kontra. Dengan melakukan analisis sentimen kita dapat mengetahui kecenderungan opini masyarakat terhadap topik pindah ibu kota Negara agar dapat dilakukan langkah strategis selanjutnya untuk meningkatkan suatu pelayanan. Penelitian ini menggunakan 3966 dataset yang diambil dari media sosial twitter dengan membagi kelas sentimen menjadi 2, yaitu positif dan negatif. Dataset akan diolah melalui beberapa tahapan text mining, diantaranya adalah pra-pemrosesan data, pembobotan kata dengan TF-IDF, dan proses pembelajaran mesin menggunakan algoritma K-Nearest Neighbors dan Naive Bayes. Pengujian sistem analisis sentimen dilakukan menggunakan perhitungan confusion matrix dengan akurasi terbaik yang diperoleh pada pembagian data train dan data test 80:20. Algoritma Naive Bayes menghasilkan akurasi sebesar 85,4%, precision 78%, recall 86%, dan F1-Score 81%. Sedangkan Algoritma K-Nearest Neighbors menghasilkan akurasi sebesar 83,3%. precision 75%, recall 83% dan F1-score 78%. Hasil analisis sentimen opini masarakat pada topik pindah ibu kota negara menggunakan model terbaik adalah 75,14% positif dan 24,86% negatif pada algoritma Naive Bayes, sedangkan untuk hasil analisis sentimen pada algoritma K-Nearest Neighbors adalah 69% positif dan 31% negatif. ENGLISH : The government's policy towards relocating the nation's capital has created debate among the people who are both pro and contra. By doing a sentiment analysis, we can find out the tendency of public opinion on the topic of moving the state capital so that further strategic steps can be taken to improve a service. This study uses 3966 datasets taken from social media twitter by dividing the sentiment class into 2, namely positive and negative. The dataset will be processed through several stages of text mining, including data pre-processing, word weighting with TF-IDF, and machine learning using the K-Nearest Neighbors and Naive Bayes algorithms. Sentiment analysis system testing is carried out using the confusion matrix calculation with the best accuracy obtained from the distribution of train data and test data 80:20. The Naive Bayes algorithm produces an accuracy of 85.4%, precision 78%, recall 86%, and F1-Score 81%. while the K-Nearest Neighbors Algorithm produces an accuracy of 83.3%. precision 75%, recall 83% and F1-score 78%. The results of the analysis of public opinion sentiment on the topic of moving the country's capital using the best model are 75.14% positive and 24.86% negative on the Naive Bayes algorithm, while the results of sentiment analysis on the K-Nearest Neighbors algorithm are 69% positive and 31% negative.

Item Type: Thesis (Sarjana)
Uncontrolled Keywords: K-Nearest Neighbors; Naive Bayes; Machine learning; Data mining; Klasifikasi; Analisis sentimen;
Subjects: Data Processing, Computer Science
Data Processing, Computer Science > Computers Mathematical Principles
Special Computer Methods > Artificial Intelligence
Divisions: Fakultas Sains dan Teknologi > Program Studi Teknik Elektro
Depositing User: Fatwa Penata Gama
Date Deposited: 11 Nov 2022 00:42
Last Modified: 11 Nov 2022 00:42
URI: https://etheses.uinsgd.ac.id/id/eprint/60657

Actions (login required)

View Item View Item