DreamLens AI — Dream Journal Interpreter with Emotion Detection | Flask Final Year Project

DreamLens AI is a machine learning powered dream journal application that turns a simple habit of writing down dreams into a structured, data driven experience. A user types out what they remember from a dream, and the system runs it through a trained classifier to identify the dominant emotion behind the entry, scans the text for recurring dream symbols using a keyword based detection layer, and produces a natural language interpretation of what the dream might mean. Every entry is saved to a personal journal, and a history page charts emotional trends over time so patterns become visible instead of forgotten.

Project Features

• Emotion classification across six categories — fear, joy, sadness, anxiety, peace, and confusion — using a TF-IDF vectorizer feeding an XGBoost classifier
• Keyword based symbol detector that flags recurring dream motifs directly from the entry text
• AI generated interpretation text powered by Groq's LLaMA models, with a local templated fallback so the app runs fully even without an API key
• SQLite backed journal history with a Chart.js emotion trend visualization
• Clean Flask backend exposing REST style endpoints for interpretation, entry storage, and statistics
• Seeded demo entries on first run so the history page and charts are never empty during a demo or evaluation
• Hand written CSS with a dark, starfield inspired theme built entirely with Jinja2 templates and vanilla JavaScript, no frontend framework dependency

Applications

• Personal dream journaling and self reflection tools
• Academic demonstration of applied NLP and text classification pipelines
• Mood and emotional pattern tracking applications
• Portfolio project showcasing scikit-learn, XGBoost, and LLM integration in a single Flask app

Who It's For

This project suits BCA, MCA, BTech CSE, and BSc IT students who want a final year project that combines classical machine learning with a modern LLM integration, without needing a deep learning background. It's also a good fit for anyone who wants a working example of how a TF-IDF plus XGBoost pipeline can sit alongside a Groq API call in the same Flask application, with a graceful fallback when no API key is present.

Why Choose This Project

DreamLens AI is fully functional out of the box. The model layer, UI, and backend API are all complete and tested, and the app seeds itself with demo data on first run, which makes live demonstrations and viva presentations straightforward. Students who want to extend it further can swap in a real trained model under the same artifact filenames without touching the interpretation logic, or connect a Groq API key to unlock live AI generated interpretations. If you are exploring other AI and machine learning final year projects, this one pairs well with conversational AI builds like the Travel ChatBot built with Flutter and Google Gemini API, or with other applied ML systems such as the AI powered plant disease detection system if you want to compare a text classification pipeline against a computer vision one.