Recommendation System for eLearning contents

In this project eLearning contents are generated by scraping the websites and then keywords are extracted using Spacy library. Clustering techniques are used to cluster the contents as ‘Low’, ‘Medium’ and ‘High’ complexity. The contents are personalized by matching the contents with user’s profile. Cosine Similarity is used to recommend the websites and keywords based on search keywords as entered by the user. Top 10 search results along with the similarity scores will be displayed to the user when the user wants to search a specific eLearning material from the internet.

Salient Features:
– Natural Language Processing (NLP) technique used for vectorizing the text data
– Applying multiple clustering algorithms for segregating the technical contents
– Personalized experience using content based filtering
– Recommendation of reference materials based on user’s search key words
– Can be extended to other industry use case applications

Tech Stacks:
Numpy, Pandas, TFIDF Vectorizer, Cosine Similarity, Python

Industry:
EdTech, eCommerce