Claim Missing Document
Check
Articles

Found 1 Documents
Search

ANALISIS SENTIMEN DAN PAIN POINT KONSUMEN TERHADAP BRAND MASSTIGE PARAGON PADA TWITTER/X MENGGUNAKAN NAIVE BAYES Meisya Marta Rahayu
Jurnal Ilmiah Manajemen Informasi dan Komunikasi Vol 10 No 1 (2026): Juni 2026
Publisher : Sekolah Tinggi Multi Media "MMTC" Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jimik.v10i1.683

Abstract

Social media has become an important space for consumers to share opinions, experiences, and purchase considerations regarding beauty products. This study aims to analyze consumer sentiment and pain points toward Paragon’s masstige brands on Twitter/X. The initial scope of the study included five brands: Instaperfect, Crystallure, Labore, TAVI, and Make Over. However, Make Over was excluded from the final analysis because the keywords “make over” or “makeover” generated a high level of noise due to their common use as general English terms. The final dataset consisted of 488 comments from four brands, namely Instaperfect, Crystallure, Labore, and TAVI. This study employed a text mining approach, including data crawling, cleaning, preprocessing, sentiment labeling, Naive Bayes classification using Orange Data Mining, and pain point analysis. The results show that neutral sentiment dominated with 368 comments (75.41%), followed by positive sentiment with 103 comments (21.11%) and negative sentiment with 17 comments (3.48%). The main pain points identified were stock/access/transaction, shade/color, and price/promotion/value. These findings indicate that consumer conversations are largely positioned in the information-seeking and purchase-consideration stages, making neutral comments strategically valuable for brands in understanding consumer needs.