Overview
As part of a project in the Samsung Innovation Campus, I developed a comprehensive Twitter sentiment analysis tool. This project involved the application of various machine learning models to analyze and classify the sentiment of tweets.
Project Components
There are several key components to this project:
1. Compiling the Dataset
- File:
Compiling_Dataset.ipynb - Description: This notebook includes the process of collecting and compiling the dataset for sentiment analysis.
2. Exploratory Data Analysis and Preprocessing
- File:
EDA_and_Preprocessing.ipynb - Description: This notebook contains the steps taken to explore and preprocess the data to make it suitable for machine learning models.
3. Sentiment Analysis Models
- Support Vector Machine (SVM):
- File:
svm_model.ipynb - Description: This notebook demonstrates the implementation of the SVM model for sentiment classification.
- File:
- Long Short-Term Memory (LSTM):
- File:
LSTM.ipynb - Description: This notebook showcases the LSTM model used for analyzing the sentiment of tweets.
- File:
- Random Forest:
- File:
Random_Forest.ipynb - Description: This notebook details the application of the Random Forest algorithm for sentiment analysis.
- File:
- Naive Bayes:
- File:
Naive_Bayes.ipynb - Description: This notebook explains the use of the Naive Bayes algorithm for classifying tweet sentiments.
- File:
- Transfer Learning:
- File:
Transfer_Learning.ipynb - Description: This notebook illustrates the process of using transfer learning for sentiment analysis.
- File:
Technologies Used
- Python
- Jupyter Notebooks
- Scikit-learn
- TensorFlow/Keras
- Pandas
- NumPy
- Matplotlib/Seaborn
Data
- Dataset File:
tweet_and_emotion.csv - Description: This file contains the tweet data and their corresponding sentiment labels used for training and evaluating the models.
Repository
You can find the complete code and resources for this project in the GitHub repository.
Created as part of the Samsung Innovation Campus project in April 2023.