A recommendation system, or recommender system tries to make predictions on user preferences and make recommendations which should interest customers.
Recommendation systems typically appear on many e-commerce sites because of providing better conversion rates. According to some articles you can find on the Internet, 35% of Amazon's sales are result of its recommendation engine.
There are basically two approaches to make recommendations:
Collaborative filtering (or social filtering)
Content-based filtering
A Speaking Recommender System using and content based filtering
This type of filter does not involve other users if not ourselves. Based on what we like, the algorithm will simply pick items with similar content to recommend us.
Content-based filtering, also referred to as cognitive filtering, recommends items based on a comparison between the content of the items and a user profile. The content of each item is represented as a set of descriptors or terms, typically the words that occur in a document. The user profile is represented with the same terms and built up by analyzing the content of items which have been seen by the user.
Collaborative recommenders rely on data generated by users as they interact with items. Examples of this include:
Users rating movies on a scale of 1–5
Users purchasing or even viewing items on an online retail site
Users reacting with “thumbs up” or “thumbs down” to songs on an online music streaming service
Swiping left or right on a dating site
In the context of a movie recommender, collaborative filters find trends in how similar users rate movies based on rating profiles.
#Now Apart from building the recommeder system i have used the google's gtts i.e Google text-to-speech Api.
First the user tells the name of a particular movie then convert that audio into text and feed into the function to get
relevant recommendation
Then reading the recommendation for a particular movie
The dataset is freely available on the movie-lens website -https://movielens.org/