Wellness Journal App

A full-stack wellness journaling application that combines personal journaling with AI-powered responses and sentiment analysis to support mental health and self-reflection.

Calendar Rating UI

Features

• Daily Journaling: Write entries with mood tracking (1-10 scale)
• AI Assistant: Get personalized responses based on your journal entries
• Sentiment Analysis: ML-powered mood analysis and insights
• Mood Visualization: Calendar view of your emotional patterns
• Responsive Design: Works on desktop and mobile devices

Tech Stack

• Frontend: React with TypeScript
• Backend: Node.js with Express and SQLite
• ML Service: Python with scikit-learn for sentiment analysis
• AI Responses: HuggingFace API integration
• Styling: CSS with custom design system

Machine Learning: Emotion-Based Binary Sentiment Classification

Overview:

This project features a custom-trained machine learning model for binary sentiment classification (positive/negative) specifically designed for journal wellness applications. The model leverages the GoEmotions dataset to provide accurate emotional tone analysis of user journal entries.

Methodology:

1. Data Acquisition: Used the GoEmotions dataset with 28 distinct emotion categories
2. Emotion-to-Sentiment Mapping: Converted granular emotions into binary sentiment labels
3. Data Preparation: Filtered neutral entries and split data for training/validation/testing
4. Text Preprocessing: Implemented cleaning pipeline (punctuation removal, lowercasing, stop word elimination)
5. Model Development: Built scikit-learn pipeline with TfidfVectorizer and LogisticRegression
6. Model Evaluation: Achieved strong performance on validation and test sets
7. Deployment Integration: Exported model via joblib for seamless API integration

Key Technologies:

• Python, pandas, scikit-learn, nltk, datasets library
• GoEmotions dataset for training data
• Custom preprocessing pipeline matching production environment

Training Notebook: See ml-training/sentiment_model_training.ipynb for complete model development process.

Quick Start

1. Install dependencies:

   npm run install:all

2. Start development servers:

   npm run dev

3. Access the application:

   • Frontend: http://localhost:3000
   • Backend API: http://localhost:3001
   • ML Service: http://localhost:3002

Project Structure

├── frontend/          # React TypeScript app
├── backend/           # Express.js API server  
├── ml-service/        # Python sentiment analysis service
├── ml-training/       # Machine learning model development
└── docs/             # Project documentation

Development

The app runs three services concurrently:

• Frontend: React development server
• Backend: Express API with SQLite database
• ML Service: FastAPI app for sentiment analysis

All services start automatically with npm run dev.