This project aims to develop a cyberbullying tweet recognition system using machine learning techniques. The project includes data preprocessing, model building, and a user-friendly web application built using Streamlit.
- Data preprocessing including text cleaning, tokenization, stemming, and lemmatization.
- Model training using Linear Support Vector Machine (LSVM) for cyberbullying tweet detection.
- Streamlit web application for user interface.
- Prediction of cyberbullying content based on user input.
-
Clone the repository:
git clone https://github.com/srishrachamalla7/cyberbullying-recognition.git cd cyberbullying-recognition
-
Install the required dependencies:
pip install -r requirements.txt
-
Run the Streamlit app:
streamlit run app.py
data/
: Contains the dataset used for training and testing.models/
: Includes saved model files after training.notebooks/
: Jupyter notebooks for data analysis and preprocessing.app.py
: Streamlit web application for user interaction.train_model.py
: Script for model training.preprocess.py
: Functions for data preprocessing.utils.py
: Utility functions used across the project.
- Run the Streamlit app using the command mentioned above.
- Input a tweet in the app.
- The app predicts whether the input tweet contains cyberbullying content or not.
- Enhance model performance by experimenting with different algorithms and hyperparameters.
- Include more advanced text processing techniques.
- Extend the web app with more interactive features and visualizations.
Remember to customize the content according to your project's details, including the project features, structure, usage instructions, and contributions. Also, include any relevant images, logos, and links to external resources.