Sentiment Analysis of Tokopedia Tweets using Multinomial Naive Bayes & TF-IDF

This repository contains a Jupyter Notebook program for classifying tweet data from Twitter using the Multinomial Naive Bayes algorithm and TF-IDF, utilizing the Scikit-learn library.

📂 Folder Structure

📁 Kode (Code)

This folder contains the scripts used for my thesis, with the following workflow:

1. Data Crawling – Collecting tweet data
2. Data Preprocessing & Visualization – Cleaning and analyzing data
3. NaN Detection (Optional) – Handling missing values caused by formatting issues (e.g., missing commas)
4. Multinomial Naive Bayes Classification – Training and evaluating the model

📁 Data (Dataset)

This folder includes three datasets:

• Dataset – The raw data, manually labeled
• Preprocessed Dataset – Cleaned data, without stemming
• Preprocessed + Stemmed Dataset – Cleaned data with stemming applied

🚀 Feel free to explore and contribute!