ANALISIS SENTIMEN DI TWITTER TERHADAP FENOMENA FAST FASHION MENGGUNAKAN ALGORITMA SVM DAN NAIVE BAYES

AZ-ZAHRA, SYAFIAH (2025) ANALISIS SENTIMEN DI TWITTER TERHADAP FENOMENA FAST FASHION MENGGUNAKAN ALGORITMA SVM DAN NAIVE BAYES. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) 01 Cover.pdf Download (672kB) | Preview
	Text (BAB I) 02 Bab 1.pdf Restricted to Registered users only Download (103kB)
	Text (BAB II) 03 Bab 2.pdf Restricted to Registered users only Download (251kB)
	Text (BAB II) 04 Bab 3.pdf Restricted to Registered users only Download (193kB)
	Text (BAB IV) 05 Bab 4.pdf Restricted to Registered users only Download (497kB)
	Text (BAB V) 06 Bab 5.pdf Restricted to Registered users only Download (93kB)
	Text (DAFTAR PUSTAKA) 07 Daftar Pustaka.pdf Restricted to Registered users only Download (174kB)
	Text (LAMPIRAN) 08 Lampiran.pdf Restricted to Registered users only Download (596kB)

Abstract

Fast fashion has become a global phenomenon in the fashion industry, this phenomenon drives the production and consumption of clothing quickly following changing trends. This trend has caused various responses among social media users, especially on the Twitter platform, where many users share their views on aspects of fast fashion. This study aims to analyze public sentiment toward fast fashion and test the SVM and Naïve Bayes algorithms in classifying sentiment with several test scenarios. The scenarios tested are algorithm modeling with and without the oversampling technique, algorithm modeling with and without ChiSquare feature selection, and algorithm modeling with and without the oversampling technique and Chi-Square techniques. The data used are tweets from Twitter social media users, which are then processed through sentiment labeling and data cleaning and preprocessing. The results show that the best model is obtained by the SVM algorithm with parameters C = 10, combined with random oversampling and Chi-Square, producing the highest accuracy of 79%. Data visualization with WordCloud shows that positive sentiment is mostly associated with low prices and attractive designs, while negative sentiment highlights low quality, fast production, and environmental issues. These findings are expected to provide input for industry players and policymakers to increase positive perceptions of fast fashion through improvements in quality, sustainability, and production ethics. Kata kunci: Fast fashion, Sentiment analysis, Support vector machine, Naive bayes, Twitter. Fast fashion telah menjadi fenomena global di industri mode, fenomena ini mendorong produksi dan konsumsi pakaian dengan cepat mengikuti tren yang berubah-ubah. Tren ini menimbulkan berbagai respons di kalangan pengguna media sosial, terutama di platform Twitter, dimana banyak pengguna berbagi pandangan mereka tentang aspek-aspek fast fashion. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat terhadap fast fashion dan pengujian algoritma SVM dan Naïve Bayes dalam melakukan klasifikasi sentimen dengan beberapa skenario pengujian. Skenario yang diujikan yaitu pemodelan algoritma dengan dan tanpa Teknik oversampling, pemodelan algoritma dengan dan tanpa seleksi fitur Chi-Square, dan pemodelan algoritma dengan dan tanpa teknik oversampling dan Chi-Square. Data yang digunakan berupa tweet pengguna media sosial Twitter, yang kemudian diproses melalui tahapan pelabelan sentimen dan pembersihan data atau preprocessing. Hasil menunjukkan bahwa model terbaik diperoleh oleh algoritma SVM dengan parameter C=10, dikombinasikan dengan teknik Random Oversampling dan Chi-Square, menghasilkan akurasi tertinggi sebesar 79%. Visualisasi data dengan WordCloud menunjukkan bahwa sentimen positif banyak dikaitkan dengan harga murah dan desain menarik, sedangkan sentimen negatif menyoroti kualitas rendah, produksi cepat, dan isu lingkungan. Temuan ini diharapkan dapat menjadi masukan bagi pelaku industri dan pembuat kebijakan untuk meningkatkan persepsi positif terhadap fast fashion melalui perbaikan kualitas, keberlanjutan, dan etika produksi. Kata kunci: Fast fashion, Analisis sentimen, Support vector machine, Naive bayes, Twitter

Item Type:	Thesis (S1)
Call Number CD:	FIK/INFO. 25 101
NIM/NIDN Creators:	41521010138
Uncontrolled Keywords:	Fast fashion, Analisis sentimen, Support vector machine, Naive bayes, Twitter
Subjects:	000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika 500 Natural Science and Mathematics/Ilmu-ilmu Alam dan Matematika > 510 Mathematics/Matematika > 518 Numerical Analysis/Analisis Numerik, Analisa Numerik > 518.1 Algorithms/Algoritma 600 Technology/Teknologi > 640 Home Economic and Family Living Management/Kesejahteraan Rumah Tangga dan Manajemen Kehidupan Keluarga > 646 Sewing Materials and Equipment, Clothing, Management of Personal and Family Life/Mesin Jahit dan Perlengkapan Menjahit, Pakaian, Management Pribadi dan Keluarga > 646.3 Clothing/Pakaian, Tata Busana, Mode Busana, Fashion
Divisions:	Fakultas Ilmu Komputer > Informatika
Depositing User:	khalimah
Date Deposited:	02 Aug 2025 03:53
Last Modified:	02 Aug 2025 03:53
URI:	http://repository.mercubuana.ac.id/id/eprint/96472

Actions (login required)

View Item