NURROFIAH, NPM 2209100095 (2026) ANALISIS SENTIMEN TINGKAT KEPUASAN PENGGUNA APLIKASI SHOPEE MENGGUNAKAN METODE TF-IDF DAN NAIVE BAYES. Skripsi thesis, Universitas Labuhanbatu.
![[img]](http://repository.ulb.ac.id/style/images/fileicons/text.png) |
Text
COVER.pdf Download (1MB) |
|
Text
BAB I.pdf Download (179kB) |
|
Text
BAB II.pdf Download (288kB) |
|
Text
BAB III.pdf Download (389kB) |
|
Text
BAB IV.pdf Download (498kB) |
|
Text
BAB V.pdf Download (151kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (146kB) |
|
Text
LAMPIRAN.pdf Download (4MB) |
Abstract
Penelitian ini bertujuan untuk menganalisis sentimen tingkat kepuasan pengguna terhadap aplikasi Shopee berdasarkan ulasan yang diberikan pada platform digital. Analisis sentimen dilakukan untuk mengetahui persepsi pengguna terhadap kualitas layanan aplikasi sehingga dapat menjadi bahan evaluasi bagi pengembang aplikasi. Landasan teori dalam penelitian ini menggunakan konsep analisis sentimen, yaitu proses mengidentifikasi opini atau perasaan seseorang terhadap suatu produk atau layanan berdasarkan data teks. Metode yang digunakan dalam penelitian ini adalah pembobotan kata menggunakan TF-IDF dan proses klasifikasi menggunakan algoritma Naive Bayes yang dikenal efektif dalam pengolahan data teks. Pada tahap analisis dan perancangan, data ulasan pengguna terlebih dahulu melalui proses preprocessing yang meliputi Cleaning text, tokenisasi, Stopword removal, dan Stemming sebelum dilakukan pembobotan menggunakan TF-IDF. Data yang telah diproses kemudian dibagi menjadi data training dan data testing untuk membangun serta menguji model klasifikasi sentimen menggunakan algoritma Naive Bayes. Hasil klasifikasi menunjukkan bahwa model mampu mengidentifikasi sentimen pengguna dengan cukup baik dengan total 969 data ulasan, dimana 775 data digunakan sebagai data training dan 194 data sebagai data testing. Hasil evaluasi model menunjukkan nilai akurasi sebesar 84%, dengan performa klasifikasi yang lebih baik pada sentimen positif dibandingkan sentimen negatif, serta memiliki keunggulan dalam proses komputasi yang cepat dan kemampuan yang baik dalam mengolah data teks berukuran besar. Berdasarkan hasil penelitian yang telah dilakukan, dapat disimpulkan bahwa kombinasi metode TF-IDF dan Naive Bayes mampu memberikan performa yang cukup baik dalam melakukan analisis sentimen terhadap ulasan pengguna Shopee. Penelitian selanjutnya disarankan untuk menggunakan jumlah dataset yang lebih besar serta membandingkan metode klasifikasi lain agar dapat meningkatkan akurasi dan kualitas hasil analisis sentimen. Kata Kunci: Analisis Sentimen, TF-IDF, Naive Bayes, Klasifikasi Teks, Kepuasan Pengguna =============================================================== This study aims to analyze user satisfaction sentiment towards the Shopee application based on reviews given on the digital platform. Sentiment analysis is conducted to determine user perceptions of the application's service quality so that it can be used as evaluation material for application developers. The theoretical basis in this study uses the concept of sentiment analysis, namely the process of identifying a person's opinion or feelings towards a product or service based on text data. The method used in this study is word weighting using TF-IDF and a classification process using the Naive Bayes algorithm, which is known to be effective in processing text data. In the analysis and design stage, user review data first goes through a preprocessing process that includes text Cleaning, tokenization, Stopword removal, and Stemming before being weighted using TF-IDF. The processed data is then divided into training data and testing data to build and test a sentiment classification model using the Naive Bayes algorithm. The classification results show that the model is able to identify user sentiment quite well with a total of 969 review data, of which 775 data are used as training data and 194 data as testing data. The model evaluation results showed an accuracy value of 84%, with better classification performance for positive sentiment compared to negative sentiment, and has advantages in fast computational processes and good capabilities in processing large text data. Based on the results of the research that has been conducted, it can be concluded that the combination of TF-IDF and Naive Bayes methods is able to provide quite good performance in conducting sentiment analysis on Shopee user reviews. Further research is recommended to use a larger dataset and compare other classification methods to improve the accuracy and quality of sentiment analysis results. Keywords: Sentiment Analysis, TF-IDF, Naive Bayes, Text Classification, User Satisfaction
| Item Type: | Thesis (Skripsi) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, TF-IDF, Naive Bayes, Klasifikasi Teks, Kepuasan Pengguna===============Sentiment Analysis, TF-IDF, Naive Bayes, Text Classification, User Satisfaction |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases |
| Divisions: | Fakultas Sains Dan Teknologi > Sistem Informasi |
| Depositing User: | Unnamed user with email repository@ulb.ac.id |
| Date Deposited: | 21 Apr 2026 08:17 |
| Last Modified: | 21 Apr 2026 08:17 |
| URI: | http://repository.ulb.ac.id/id/eprint/2137 |
Actions (login required)
 |
View Item |
