Nearly all, if not all, of us like that feeling of someone knowing what we want and placing it within our grasp, and that's the feeling a recommender system gives. This project was my way of contributing to the ecosystem of recommender engines. A lot of companies are looking for ways to personalize the user, as well as, search experience on their digital platforms and one of the ways around this is to ensure that products or offerings that resonate with the user are the ones always in their faces.

There are a lot of use-cases for recommender systems. E-commerce stores use it to suggest what a user should purchase next based on an earlier or current purchase. It is useful in real-estate, entertainment, name it!. On this occasion, my recommender system finds use with online news and article readers. Particularly, It suggests to a user what articles they should be reading after the one they are currently on and this is done based on relatedness of the articles, largely determined by presence of similar keywords.

Kmeans and Cosine similarity were not used as one approach but as two different approaches here, and the results show that both suggest quite similar articles based on relatedness to what the user is currently reading.

This project was my finale for mini-projects on the Women Techsters fellowship.

Happy exploring!