MindCare: A Mental Health Intervention System Using Linguistic Intelligence

Overview

This repository contains the implementation of a Mood Classification and Recommendation System using a dataset and chatbot. The system provides mental health support and guidance through pre-programmed responses, emotional information extraction, activity recommendations, and music recommendations.

Tech Stack

Table of Contents

• System Architecture
• User Journey
• Implementation
  • Data Pre-processing
  • Rasa Framework
    • Natural Language Understanding (NLU)
• Features
• Demo Video
• Installation
• Usage
• Contributing
• License

System Architecture

The system architecture consists of the following components:

1. Chatbot Interface: The primary user interaction point providing support and guidance with pre-programmed responses. It collects emotional data from user inputs.
2. Data Pre-processing: Involves cleaning and transforming text or speech data, which is converted to text via speech recognition tools.
3. Dataset Handling: The dataset, containing labeled text or speech data, is split into training and testing sets.
4. Emotion Detection: Intents, entities, and responses are created for different emotions (Neutral, Happy, Sad, Love, Anger) using the Rasa framework.
5. Response Generation: Based on the predicted emotions, the system generates appropriate text, audio, and activity responses.
6. Sentiment Analysis: Weekly progress is analyzed using models like Naive Bayes to provide sentiment insights.
7. Location-Based Services: APIs or databases ascertain the user's location to provide tailored professional advice based on sentiment analysis.
8. Activity and Music Recommendation: Specific exercises or curated playlists are recommended based on the user's emotional state.

User Journey

1. Registration: New users create accounts and provide necessary information for personalization.
2. Login: Users log in using their credentials. Authentication ensures secure access.
3. Main Functionalities:
   • Progress Tracking: Users monitor their mental health progress over time.
   • Chatbot: Enables text-based conversations for support and guidance.
   • Activities: Curated exercises tailored to individual needs.
   • Professional Consultancies: Access to mental health experts for consultations.
4. Interaction: Users input text to initiate conversations, and the system classifies and responds to the input using natural language processing techniques.
5. End Session: Users can terminate their interaction at any time using the "Stop" option.

Implementation

Data Pre-processing

1. Loading the Dataset: The dataset containing tweets and their associated emotions is loaded.
2. Cleaning the Tweets: Special characters, URLs, and mentions are removed, and the text is converted to lowercase.
3. Label Encoding: Emotion labels are converted into numerical values.
4. Processing and Saving Data: Each tweet is processed, cleaned, and saved into a new CSV file with encoded labels.
5. Final Output: A cleaned dataset ready for further analysis or training machine learning models.

Rasa Framework

Natural Language Understanding (NLU)

Intent classification is performed to map input text to predefined intents. The goal is to assign a probability distribution over intents given an input text, represented as P(intent | x), where intent is one of the predefined intents.

Features

• Chatbot Interface: Engages users in text-based conversations and provides personalized recommendations.
• Activity Recommendation: Suggests exercises or activities based on emotional needs.
• Music Recommendation: Offers curated playlists to enhance mood or relaxation.
• Sentiment Analysis: Provides weekly progress analysis and insights.
• Location-Based Services: Delivers professional advice based on user's location.
• Secure Authentication: Ensures secure access through credential validation.
• Progress Visualization: Users can visualize their mental health progress over time.

Demo Video

Demo Video

Installation

To set up the project locally on Windows using Command Prompt, follow these steps:

Prerequisites

Note: You can download these from Prerequisites folder. This project works on Windows.

Python 3.8.10 Click here to download Python | Windows installer (64-bit)

Xamp Server | Click here to download XAMP and Database
After completing the installation and setup, create new database with the name 'mind_care' and import the database on phpMyAdmin.

Command Prompt 1

1. Clone the repository:

   git clone https://github.com/Roopesh519/Mindcare-Conversation-Chatbot-Rasa-Framework.git
   cd Mindcare-Conversation-Chatbot-Rasa-Framework

2. Create a virtual environment named venv:

   python -m venv venv

3. Activate the virtual environment:

   venv\Scripts\activate

4. Install the required dependencies:

   cd Mindcare
   pip install -r requirement.txt

5. Navigate to the Rasa model directory:

   cd MindCareModal

6. Run the Rasa server with API and CORS enabled:

   rasa run --enable-api --cors "*"

Command Prompt 2

1. Open a new Command Prompt and navigate to the cloned repository directory:

   cd path\to\Mindcare-Conversation-Chatbot-Rasa-Framework

2. Activate the virtual environment:

   venv\Scripts\activate

3. Navigate to the Rasa model directory:

   cd Mindcare\MindCareModal

4. Run the Rasa action server:

   rasa run actions

Command Prompt 3

1. Open another new Command Prompt and navigate to the cloned repository directory:

   cd path\to\Mindcare-Conversation-Chatbot-Rasa-Framework

2. Activate the virtual environment:

   venv\Scripts\activate

3. Navigate to the Django application directory:

   cd Mindcare\MindCare

4. Run the Django development server:

   python manage.py runserver

Usage

1. Start the Rasa server: Follow the steps in Command Prompt 1.
2. Start the Rasa action server: Follow the steps in Command Prompt 2.
3. Start the Django server: Follow the steps in Command Prompt 3.

You can now interact with the chatbot and use the mental health intervention system through the web interface provided by the Django server.

Contributing

Contributions are welcome! Please read the CONTRIBUTING.md file for more information on how to contribute to this project.

License

This project is licensed under the MIT License - see the LICENSE file for details.