Kasturi, Mulya Hara and Siregar, Riki Ruli Affandi and Munir, Buyung Sofiarto (2024) ANALISIS SENTIMEN OPINI TERHADAP SOCIAL COMMERCE MENGGUNAKAN METODE NAÏVE BAIYES & SUPPORT VECTOR MACHINE. Diploma thesis, ITPLN.
![[thumbnail of 202031165_Mulya Hara Kasturi_Revisi_Skripsi_MULYA HARA Kasturi.pdf]](https://repository.itpln.ac.id/style/images/fileicons/text.png)
202031165_Mulya Hara Kasturi_Revisi_Skripsi_MULYA HARA Kasturi.pdf
Restricted to Registered users only
Download (11MB)
Abstract
Terdapat adanya masalah terkait social commerce di platfrom TikTok, dengan masalah tersebut maka dilakukan analisis sentimen terhadap social commerce di platfrom TikTok untuk memahami pandangan masyarakat terhadap kebijakan ini. Penelitian ini menggunakan algoritma Naive Baiyes dan Support Vector Machine untuk menganalisis sentimen opini terhadap social commerce pada media sosial Twitter. Data di dapatkan melalui crawling Twitter, dengan jumlah data awal 1372. Setelah dilakukan pembersihan data, tersisa 1333 data yang digunakan dalam penelitian. Data tersebut terdiri dari 577 data positif, 465 data negatif, dan 291 data netral. Pembagian data dalam penelitian ini, dibagi menjadi tiga bagian yaitu, data pelatihan 80%, data pengujian 10%, dan data validasi 10%. Hasil sentimen terhadap social commerce di platfrom TikTok dipandang positif pada media social Twitter. Hasil penelitian menunjukkan akurasi 64% dengan metode Naive Baiyes sedangkan dengan menggunakan metode Support Vector Machine mendapatkan akurasi 76% sehingga Support Vector Machine lebih unggul dari Naive Baiyes .
There are problems related to social commerce on the TikTok platform, with this problem, a sentiment analysis of social commerce on the TikTok platfrom is carried out to understand the public's view of this policy. This research uses the Naive Baiyes algorithm and Support Vector Machine to analyze opinion sentiment towards social commerce on Twitter social media. Data is obtained through crawling Twitter, with an initial data count of 1372. After data cleaning, the remaining 1333 data were used in the study. The data consists of 577 positive data, 465 negative data, and 291 neutral data. Data division in this study is divided into three parts, namely, 80% training data, 10% testing data, and 10% validation data. The results of sentiment towards social commerce on the TikTok platform are viewed positively on Twitter social media. The results showed an accuracy of 64% with the Naive Baiyes method while using the Support Vector Machine method to get 76% accuracy so that Support Vector Machine is superior to Naive Baiyes.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | Analisis Sentimen, Naive Baiyes, Support Vector Machine, Twitter, Social Commerce, TikTok Shop. Sentiment Analysis, Naive Baiyes, Support Vector Machine, Twitter, Social Commerce, TikTok Shop. |
Subjects: | Skripsi Bidang Keilmuan > Teknik Informatika |
Divisions: | Fakultas Telematika Energi > S1 Teknik Informatika |
Depositing User: | Sudarman |
Date Deposited: | 05 Oct 2025 10:59 |
Last Modified: | 05 Oct 2025 10:59 |
URI: | https://repository.itpln.ac.id/id/eprint/1758 |