This is a submission for the Bright Data Web Scraping Challenge: Most Creative Use of Web Data for AI Models
What I Built
The “Sentiment Analysis for Course Recommendation” project leverages web scraping and a robust machine learning model, featuring a Random Forest Classifier trained on the Coursera dataset, to assess courses based on user comments and sentiments, offering a more nuanced rating system beyond conventional star ratings. Users simply input the URL of a course they wish to evaluate, and the system extracts and classifies comments as positive, negative, or neutral. The resultant course rating is computed from these sentiments, facilitating informed decision-making for prospective learners. Additionally, the application offers course comparison functionality, allowing users to evaluate courses based on diverse criteria. This innovative approach, built with Python, Flask, and the YouTube API, empowers individuals to make education choices tailored to their specific needs and preferences, enhancing the online learning experience.
Demo
https://res.cloudinary.com/dpwustwce/image/upload/v1734845594/qjlclhnfikf3lja9qk6t.gif
How I Used Bright Data
Collect real-time user reviews and comments from course platforms like Coursera. Preprocess the data to remove noise, label sentiments (positive, negative, neutral), and structure it into a training-ready format. Fine-tune a sentiment analysis model like Random Forest or a transformer-based model, integrating it into a Flask-based application for personalized course recommendations and comparisons.
Source link
lol
