Penelitian ini bertujuan untuk mengetahui opini masyarakat yang dituangkan di media sosial tentang pelaksnaan pembelajaran daring selama masa pandemi COVID-19 menggunakan pendekatan Lexicon. Data ulasan yang digunakan bersumber dari media sosial Twitter antara Maret 2020 hingga Februari 2022. Kata kunci yang digunakan dalam pencarian data meliputi 1) belajar online, 2) pembelajaran daring, 3) sekolah online, dan 4) belajar dari rumah. Data tanggapan yang berhasil dikumpulkan 73.120 ulasan. Proses klasifikasi sentimen diawali dengan proses pembersihan opini untuk menghilangkan URL, tanda baca, hashtag, mention, tokenizing, normalisasi kata yang tidak baku, menghilangkan stopword, stemming, serta menghilangkan duplikasi ulasan. Selanjutnya klasifikasi opini menggunakan pendekatan lexicon. Sumber daya kamus lexicon yang digunakan yaitu InSet (Indonesian Sentiment) Lexicon. Hasil klasifikasi sentimen ulasan pembelajaran daring didominasi oleh klasifikasi opini positif mencapai 77,58 % atau 56.723 tanggapan. Sedangkan ulasan negatif hanya 17,97% atau 13.139 komentar. Lima kata yang paling sering muncul dalam kalimat opini positif pembelajaran daring adalah : 1) ”sekolah”, 2) ”rumah”, 3) ”guru”, 4) ”siswa”, dan 5) ”nilai”. Sedangkan lima kata yang muncul dalam sentimen negatif pembelajaran daring sesuai gambar 4 adalah : 1) ”ajar”, 2) ”anak”, 3) ”tugas”, 4) ”didik”, dan 5) ”kuota”.
The increased consumption of beauty products as a lifestyle has increased public opinion on the beauty products used. Generally, reviews are given through posts on social media. This study discusses the classification of sentiment analysis on the use of serum beauty products on Twitter using the Naïve Bayes Multinomial algorithm. Sentiment analysis of serum beauty products is carried out to provide information and preferences to the public regarding the quality of a product. The results of the information and preferences become a reference for consideration in choosing the appropriate serum beauty product. The data used in this study were 27,587 tweets using three keywords, namely "serum," "face serum", and "beauty serum". Tweet data is divided into training data and test data with the number of training data as much as 22,070 tweets and test data as much as 5,518 tweets. The data is categorized using the lexicon senticnet 7 dictionary based on polarity values. The results of the analysis of positive sentiment are 35%, negative sentiment is 63.8%, and neutral sentiment is 1.2%. The classification results using Naïve Bayes Multinomial obtain the highest accuracy value of 80%. The Confusion Matrix results get the highest precision value of 88%, the highest recall of 81%, and the highest f1-Score of 86%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.