Analisis Sentimen Opini Masyarakat Terhadap Bakal Calon Walikota Surabaya 2020 Berdasarkan Social Media Mining Menggunakan Algoritma N-Gram-Multichannel CNN

Prestasi, Ferisa Tri Putri (2020) Analisis Sentimen Opini Masyarakat Terhadap Bakal Calon Walikota Surabaya 2020 Berdasarkan Social Media Mining Menggunakan Algoritma N-Gram-Multichannel CNN. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06111640000031-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (5MB) | Request a copy

Abstract

Pilwali Surabaya mendatang menjadi perhatian publik, dimana para bakal calon mulai mengajukan diri dan ramai diperbincangkan di media sosial, diantaranya Facebook dan Twitter. Penting bagi bakal calon untuk mengetahui sentimen opini yang berkembang di media sosial dengan menggunakan implementasi analisa sentimen pada social media mining untuk dapat di klasifikasi sentimen positif atau negatif dari opini tersebut. Oleh karena itu, pada penelitian analisa sentimen ini digunakan algoritma N-Gram Multichannel CNN untuk mendapatkan analisa sentimen dengan memanfaatkan konsep Natural Language Processing yang memungkinkan komputer untuk memproses dan memahami bahasa alami manusia dan memperoleh akurasi model optimal terhadap dataset opini. Hasil diperoleh analisis sentimen pada teks berisikan opini masyarakat dari media sosial terhadap bakal calon Walikota Surabaya 2020 dapat diterapkan dengan baik menggunakan Algoritma N-Gram-Multichannel CNN dengan diperoleh akurasi model terhadap data latih sebesar 94.38% dan 96.12% pada data uji. ================================================================================================================== The next Surabaya elections are a matter of public concern, where prospective candidates begin to propose themselves and are widely discussed on social media, including Facebook and Twitter. It is important for prospective candidates to find out opinion sentiments that develop in social media by using the implementation of sentiment analysis in social media mining to be able to classify positive or negative sentiments from these opinions. Therefore, in this sentiment analysis research the N-Gram Multichannel CNN algorithm is used to get sentiment analysis by utilizing the concept of Natural Language Processing which enables computers to process and understand human natural language and obtain optimal model accuracy of opinion datasets. The results obtained sentiment analysis in the text containing public opinion from social media to the prospective Surabaya Mayor 2020 can be implemented well using the CNN N-Gram-Multichannel Algorithm with the accuracy of the model obtained for training data 94.38% and 96.12% in the test data.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Pilwali Surabaya, Facebook, Twitter, Social Media Mining, Analisis Sentimen, Natural Language Processing, N-Gram-Multichannel CNN, The mayor election of Surabaya, Facebook, Twitter, Social Media Mining, Sentiment Analysis, Natural Language Processing, N-Gram-Multichannel CNN
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA336 Artificial Intelligence
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA76.9.D343 Data mining
T Technology > T Technology (General) > T57.5 Data Processing
Divisions: Faculty of Mathematics, Computation, and Data Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Ferisa Tri Putri Prestasi
Date Deposited: 19 Aug 2020 07:18
Last Modified: 19 Aug 2020 07:18
URI: http://repository.its.ac.id/id/eprint/78826

Actions (login required)

View Item View Item