You are on page 1of 2

Machine Learning Project

Group Members Naqqash, Mussab


Dataset TMDB 5000 Movie Dataset
Algorithm Cosine Similarity and Nearest
Neighbour Algorithm
Sap-Ids 1980, 2024
Project Movie Recommendation
System

Movie Recommendation System


First, we need to know that there are three kinds of Movie Recommendation
System:
 Content Based Recommendation System (recommends movie based on
content)
 Popularity Based Recommendation System (recommends movie based on
popularity)
 Collaborative Recommendation System (recommends movie based on
user watching pattern)
In this project we will try to incorporate Content Based Recommendation
System and a certain amount of Popularity Based Recommendation System

Introduction:
As we know that today instead of buying DVDs users use OTT platforms like
Netflix, Amazon Prime and many more to watch their favorite movies and tv
series. Have you ever imagined why they are so popular? They are popular
because they have recommendation system that make simple for users to watch
their favorite movies which they want to. Movies are recommended to users on
their interest, popularity, or content. So, we will make a Movie
Recommendation System as our project.

Model:
Supervised Machine Learning (Classification)
Application:
User will search for his favorite movie then movies related to their favorite
searched movie will be recommended to the user. It just tells what
movies/items are most similar to the user’s movie choice.

Working
As we know that now a days everyone like to watch tv series or movies on OTT
platform like Netflix, amazon prime and many more they used recommendation
to system to attract many of their customers. So, we are also going to make a
movie recommendation system in this system we will try to incorporate the
things such as the title of the movie, the cast, director, and theme of the movie
so on. First step is of data collection, so we need to have the data of the movies
and several details about them. So, the details that we want are who is director
of the movie what is the genre of the movie furthermore description of the
movie after collecting the data we will process this data afterwards we will
extract the features. The main thing about this movie data set is that all the data
will be in the form of text. So, we cannot use this data directly we need to
convert this textual data into numerical so we will convert it into feature vectors
after that we will find the similarity score between the movies let’s say we have
5000 movies in our dataset we need to find which movies are like each other.
This will be done by using similarity score. Once this is done user will be asked
to give input so user will give the name of his/her favorite movie based on this
data and based on this input we are going to suggest them which movies they
can watch. For this we are going to use cosine similarity algorithm. This
algorithm will be used to find the similarity between the vectors. Each movie
converting into a kind of vector. User types of his favorite movie this movie
name will be compared and similar movie names will be displayed to the user in
form of list.

You might also like