Perbandingan Metode Support Vector Machine Dan Naïve Bayes Untuk Analisis Sentimen Pada Opini Produk Kecantikan (Studi Kasus: Lacoco Watermelon Glow Mask)

Adela, Putri Handayani (2022) Perbandingan Metode Support Vector Machine Dan Naïve Bayes Untuk Analisis Sentimen Pada Opini Produk Kecantikan (Studi Kasus: Lacoco Watermelon Glow Mask). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover_18102254 Adela Putri Handayani (1).pdf Download (729kB)
	Text Abstraking_18102254_Adela putri handayani.pdf Download (41kB)
	Text Abstrakindo_18102254_Adela putri handayani.pdf Download (42kB)
	Text BAB1_18102254_Adela putri handayani.pdf Download (131kB)
	Text BAB2_18102254_Adela putri handayani.pdf Download (394kB)
	Text BAB3_18102254_Adela putri handayani.pdf Download (678kB)
	Text BAB4_18102254_Adela putri handayani.pdf Restricted to Registered users only Download (353kB) | Request a copy
	Text BAB5_18102254_Adela putri handayani.pdf Download (42kB)
	Text DaftarPustaka_18102254_Adela putri handayani.pdf Download (130kB)
	Text Lampiran_18102254_Adela putri handayani.pdf Restricted to Registered users only Download (124kB) | Request a copy

Abstract

Beauty products at this time have become a woman's need because to look beautiful is a pride and confidence for every woman. But to look beautiful must be accompanied by using beauty products that can treat and treat all the problems that exist on the face. However, not all beauty products are suitable for consumers' facial skin. To find out whether a product is suitable or not, consumers must look at reviews before buying a product. One of the most popular beauty products is from the Lacoco brand. Lacoco Watermelon Glow Mask is one of the best seller and recommended products from Lacoco. On the Female Daily website this product has 65% of users recommend this product. And 1023 reviews on the Sociolla site 903 users recommend this product. Therefore, it is necessary to conduct a sentiment analysis in order to classify consumer opinions into positive or negative and can be used as an evaluation of the company. In this study using the Support Vector Machine and Naïve Bayes methods. In this study, a comparison is made with 4 scenarios which will then be tested using 3 data divisions, namely 7:3, 8:2, and 9:1. The result of this research is that the Support Vector Machine method is superior to Naïve Bayes using TFIDF word weighting using stemming with 9:1 data distribution. The accuracy value obtained is 81.46% with 504 positive comments and 109 negative comments. Keywords: Sentiment Analysis, Lacoco Watermelon Glow Mask, Classification, Support Vector Machine, Naïve Bayes.

Actions (login required)

View Item