Zahrani, Muthia and Aziza, Rosida Nur and Ely, Muhammad Jafar (2024) ANALISIS SENTIMEN PENGGUNA APLIKASI SHOPEE TERKAIT FITUR CASH ON DELIVERY (COD) MENGGUNAKAN METODE NAÏVE BAYES DAN SUPPORT VECTOR MACHINE (SVM). Diploma thesis, ITPLN.
![[thumbnail of 202031105_Muthia Zahhrani_Revisi_Skripsi_MUTHIA Zahrani.pdf]](https://repository.itpln.ac.id/style/images/fileicons/text.png)
202031105_Muthia Zahhrani_Revisi_Skripsi_MUTHIA Zahrani.pdf
Restricted to Repository staff only
Download (8MB)
Abstract
Terdapat masalah terkait social e-commerce di platform Shopee mengenai fitur pembayaran Cash on Delivery (COD). Untuk memahami pandangan masyarakat terhadap kebijakan ini, dilakukan analisis sentimen terhadap aplikasi Shopee mengenai fitur COD. Penelitian ini menggunakan algoritma Naïve Bayes dan Support Vector Machine (SVM) untuk menganalisis sentimen opini terhadap aplikasi Shopee pada ulasan Google Play Store. Proses pengumpulan data review menggunakan web scraping dengan jumlah 1840 ulasan kemudian diolah menjadi dataset dengan text pre-processing. Pelabelan sentimen menggunakan score yang sudah diberikan pengguna. Data tersebut terdiri dari 756 data positif, 890 data negatif, dan 194 data netral. Pembagian data dalam penelitian ini dibagi menjadi tiga bagian, yaitu: data pelatihan 80%, data pengujian 10%, dan validasi data 10%. Selanjutnya digunakan SMOTE dan Randomoversampler untuk menangani ketidakseimbangan kelas (class imbalance) dalam data. Hasil penelitian menunjukkan akurasi 74% dengan metode Naive Bayes, sedangkan dengan menggunakan metode Support Vector Machine mendapatkan akurasi 70%. Dengan demikian, metode Naïve Bayes lebih unggul dari SVM.
There is a problem related to social e-commerce on the Shopee platform regarding the Cash on Delivery (COD) payment feature. To understand the public's views on this policy, a sentiment analysis was conducted on the Shopee application regarding the COD feature. This study uses the Naïve Bayes and Support Vector Machine (SVM) algorithms to analyze opinion sentiments towards the Shopee application in Google Play Store reviews. The review data collection process uses web scraping with a total of 1840 reviews then processed into a dataset with text pre-processing. Sentiment labeling uses the scores that have been given by users. The data consists of 756 positive data, 890 negative data, and 194 neutral data. The data division in this study is divided into three parts, namely: 80% training data, 10% testing data, and 10% data validation. Furthermore, SMOTE and Randomoversampler are used to handle class imbalance in the data. The results of the study showed an accuracy of 74% with the Naive Bayes method, while using the Support Vector Machine method obtained an accuracy of 70%. Thus, the Naïve Bayes method is superior to SVM.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | Analisis Sentimen, Naïve Bayes, Support Vector Machine, Shopee, Cash on Delivery Sentiment Analysis, Naïve Bayes, Support Vector Machine, Shopee, Cash on Delivery |
Subjects: | Skripsi Bidang Keilmuan > Teknik Informatika |
Divisions: | Fakultas Telematika Energi > S1 Teknik Informatika |
Depositing User: | Sudarman |
Date Deposited: | 15 Sep 2025 03:23 |
Last Modified: | 15 Sep 2025 03:23 |
URI: | https://repository.itpln.ac.id/id/eprint/1054 |