https://github.com/amirhosseinhonardoust/ai-personal-study-tracker
An AI-driven productivity tracking app built with Python, Streamlit, SQLite, and Machine Learning. It logs and analyzes study sessions, predicts productivity using Random Forest models, and visualizes key insights to help learners improve focus, habits, and overall academic efficiency.
https://github.com/amirhosseinhonardoust/ai-personal-study-tracker
ai data-analytics data-visualization education learning-analytics machine-learning productivity python random-forest self-improvement sqlite streamlit student-success study-tracker time-management
Last synced: 3 months ago
JSON representation
An AI-driven productivity tracking app built with Python, Streamlit, SQLite, and Machine Learning. It logs and analyzes study sessions, predicts productivity using Random Forest models, and visualizes key insights to help learners improve focus, habits, and overall academic efficiency.
- Host: GitHub
- URL: https://github.com/amirhosseinhonardoust/ai-personal-study-tracker
- Owner: AmirhosseinHonardoust
- License: mit
- Created: 2025-10-30T06:34:55.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-10-30T07:25:04.000Z (9 months ago)
- Last Synced: 2025-10-30T08:44:32.467Z (9 months ago)
- Topics: ai, data-analytics, data-visualization, education, learning-analytics, machine-learning, productivity, python, random-forest, self-improvement, sqlite, streamlit, student-success, study-tracker, time-management
- Language: Python
- Homepage:
- Size: 87.9 KB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## AI-Powered Personal Study Tracker
An intelligent productivity analytics app built with **Python**, **SQLite**, **Machine Learning**, and **Streamlit**.
It empowers learners to track, visualize, and improve their study habits while using AI to estimate their expected productivity.
---
## Overview
The **AI-Powered Personal Study Tracker** transforms daily study logs into actionable insights.
It leverages a **Random Forest Regressor** to model productivity based on mood, distractions, study duration, caffeine intake, and more.
You can:
- Log study sessions (with mood, focus level, and caffeine)
- Analyze study performance trends and patterns
- Estimate expected productivity before studying
- View aggregated insights and KPIs
---
## Project Structure
```
ai-study-tracker/
├── app/
│ └── streamlit_app.py # Main Streamlit dashboard
├── src/
│ ├── generate_sample.py # Generate synthetic study sessions
│ ├── ingest.py # Load CSV into SQLite
│ ├── train_model.py # Train RandomForestRegressor
│ └── utils.py # Helper functions
├── sql/
│ └── schema.sql # Database schema
├── data/
│ ├── sessions.csv # Study session data
│ └── study.db # SQLite database
├── models/
│ └── pipeline.joblib # Saved ML model
└── README.md
```
---
## Dashboard Preview
### Productivity Overview
### Session Log and Estimator
---
## Quickstart
```bash
# Clone the repo
git clone https://github.com/yourusername/ai-study-tracker.git
cd ai-study-tracker
# Set up a virtual environment
python -m venv .venv
# Windows
.venv\Scripts\activate
# macOS/Linux
source .venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Generate sample data
python src/generate_sample.py --days 28 --sessions_per_day 2 --out data/sessions.csv
# Import to SQLite
python src/ingest.py --csv data/sessions.csv --db data/study.db --schema sql/schema.sql
# Train the ML model
python src/train_model.py --db data/study.db --out models/pipeline.joblib
# Launch the dashboard
streamlit run app/streamlit_app.py
```
---
## Tech Stack
| Layer | Technology |
|:--|:--|
| **Frontend** | Streamlit + Plotly |
| **Backend** | SQLite |
| **ML Model** | RandomForestRegressor (scikit-learn) |
| **Data Handling** | pandas, SQLAlchemy |
| **Language** | Python 3.10+ |
---
## Key Features
- Log daily study sessions with mood, distractions, caffeine, and techniques
- Predict productivity with ML model
- Visualize time trends and subject-wise averages
- Local SQLite database (portable and private)
- Add new sessions interactively via the Streamlit sidebar
- Estimate productivity in real-time
---
## Data Schema
| Column | Description |
|:--|:--|
| `date` | Session date (YYYY-MM-DD) |
| `start_time` / `end_time` | HH:MM 24h format |
| `duration_min` | Computed from start–end times |
| `subject` | Math, Physics, CS, Biology, History |
| `technique` | Pomodoro, Active Recall, etc. |
| `distractions` | Number of interruptions |
| `mood` | 1-5 scale |
| `caffeine_mg` | Caffeine intake (mg) |
| `productivity` | Self-rated (1-5) |
---
## Machine Learning
The app uses **RandomForestRegressor** from `scikit-learn` trained on historical study logs.
**Features used for prediction:**
- Duration (minutes)
- Subject
- Technique
- Distractions
- Mood
- Caffeine (mg)
**Target:**
- Productivity (1–5 scale)
The model is modular, you can upgrade it to **XGBoost** or **Neural Networks**.
---
## Example Insights
- Productivity tends to **increase with mood** and **decrease with distractions**
- “Active Recall” and “Spaced Repetition” correlate with higher productivity
- Optimal caffeine range: 100–200 mg
---
## UI Highlights
- Clean, interactive Streamlit interface
- Real-time productivity estimation
- KPI tiles for total minutes, average mood, and average productivity
- Line chart: *Study Minutes Over Time*
- Bar chart: *Average Productivity by Subject*