PhonePredict is a demo project that I built as my final assignment for my Data Science postgraduate studies, which I have just completed with a great result. I am very proud of how much I learned in such a short time. This tool helps you check the real market value of a smartphone in seconds and tells you if an offer is a “great deal” or “too expensive.”
Project Goal
The main goal was to create a system that clears up the confusion on the second-hand electronics market by combining data analysis with business utility. One of the biggest challenges was scraping a large number of ads from OLX. After collection, the data is automatically cleaned, and the Random Forest model is re-calculated so the system always uses the freshest data. This project was also a great way for me to learn Python and Django, which were completely new to me when I started the course.
Technologies
- Backend: Python and Django – manages the logic and user forms.
- ML: Scikit-Learn – a Random Forest regression model.
- Frontend: Tailwind CSS and Chart.js – a modern, mobile-friendly interface with dynamic charts.
- Analysis: A custom “Price Barometer” algorithm that compares the AI’s estimate with the price in an ad.
Key Features
- AI Valuation: Get an objective estimate of what your phone is worth.
- Price Barometer: Instant feedback – from “Super Deal” to “Too Expensive.”
- Dashboards: Interactive charts and maps that show current market trends.
The project is fully mobile-friendly, making it a useful tool for making buying decisions on the go.
https://predictor.gawrysiak.eu/
This is a demo version – feel free to test it and let me know what you think!