I recently developed a comprehensive price comparison and detection system leveraging advanced web scraping techniques and machine learning algorithms. The core of the project is built using Python, BeautifulSoup, and Selenium for web scraping, and a Roboflow model for object detection. The system is designed to detect specific objects such as the Apple Watch, iPhone, and PS5, and then compare their prices across popular e-commerce platforms like Flipkart, Amazon, and Reliance Digital. This project showcases my ability to integrate different technologies to solve real-world problems effectively.
The system starts by utilizing a Roboflow model to detect the object in question from a given URL. This model, accessed via a specified URL, identifies objects with a confidence threshold of 50%. Once the object is detected, the system then proceeds to scrape prices from the respective e-commerce websites. Using Selenium to automate browser actions and BeautifulSoup to parse the HTML, the system retrieves the price information. The price extraction logic is tailored to handle the unique structure of each website, ensuring accurate and reliable data collection. The scraped prices are then compared to find the lowest available price, and the user is redirected to the website offering the best deal.
This project demonstrates my proficiency in using machine learning for object detection and web scraping for real-time data retrieval. It also highlights my ability to write robust code that integrates multiple libraries and frameworks to achieve a cohesive solution. The application of this project can extend beyond price comparison to various other domains where real-time data extraction and analysis are crucial. This experience has significantly enhanced my skills in Python programming, web scraping, and working with machine learning models, preparing me for more complex challenges in the field of data science and automation.