Whosapp

Welcome to Whosapp! 🚀 This project was developed as part of the "Fundamentals of Artificial Intelligence" course at the University of Salerno. Our objective is to create a machine learning model capable of identifying the authors of WhatsApp chats. Read on to learn how to use the model! 📱

Project Structure 🏗️

The project is organized into the following directories:

• configs/: Contains configuration files, including:

  • configs/alias.json: Alias configuration
  • configs/config.json: Feature configuration

• data/: Holds project data with subdirectories:

  • data/rawdata/: Raw data storage
  • data/dataset/: Dataset used for training
  • data/wordlist/: Wordlist used for processing

• frontend/: Contains the project's frontend components

• logs/: Stores project logs, categorized into sections

• models/: Houses the machine learning models of the project

• src/: Hosts the source code of the project

Getting Started 🚀

Prerequisites

1. Create the following folders:

   • data/rawdata/
   • configs/

2. In the configs/ folder, create the following files:

   • configs/alias.json where you will put the alias configuration. The file must be in the following format:

   {
       "Username": ["Alias1", "Alias2", "Alias3"],
       "Username2": ["Alias1", "Alias2", "Alias3"]
   }

Demo Instructions 🛠️

1. Clone the repository and install the requirements:

   git clone https://github.com/danlig/WhosApp.git
   cd WhosApp
   pip install -r requirements.txt

2. Upload the chat you want to analyze in the data/rawdata/ folder

3. Run the py src/pipeline.py script. For more information, run py src/pipeline.py -h.

4. Run the py src/main.py script to load and use the model

5. Finally, run node frontend/index.js to start the frontend of the project

Other Useful Commands 🛠️

• py src/new_dataset.py: Creates a new dataset from the raw data in data/raw/ and saves it in data/dataset/ in .parquet format
• py src/new_model.py: Creates a new model from the dataset created with py src/new_dataset.py and saves it in models/ in .joblib format.
• py src/pipeline.py: Creates a new dataset and model from the raw data in data/raw/ and saves them in data/dataset/ and models/ respectively.
• py src/test_features.py: test the features that you have configured in the configs/config.json file For more information on how to use these scripts, run py <script_name>.py -h

---

Authors

• Daniele Liguori
• José Sgariglia
• Luigi Turco