Shri Vishnu Engineering College for
Women::Bhimavaram (Autonomous)
Department of IT
Academic Year: 2023-24 Class: IV B.Tech/II
Sem Project Work Abstract
Name of the class /Section IT-B
Batch number B8
Project Domain / Technology Machine Learning
Project Title Movie recommendation system
Guide Name M. Srinivasa Rao
Registered Student name Student
number Signature
20B01A1299 M. Vijaya Grace
Student Registered 20B01A12A1 M. Teertha Sree
20B01A1295 M. Varshini
20B01A1293 K. Keerthi Mala
Signature of Internal project Signature of Projects- Signature of Head of the
guide Coordinator Department
Problem Statement:
In today's digital age, there's a growing demand for a personalized movie
recommendation system that can make the movie selection process efficient
and enjoyable for users. The Movie Recommender System aims to do just that.
It utilizes a cosine similarity algorithm to predict and suggest movies based on
users' viewing history and preferences. What sets it apart is the introduction of a
Multi-Criteria Recommender System (MCRS) tailored specifically for movies.
This MCRS considers various criteria, including top casts and top 10 movie
recommendations, to offer more precise and diverse film suggestions. With this
system, we aim to revolutionize the way people discover and enjoy movies,
ensuring a tailored and satisfying experience in the world of digital
entertainment.
Objectives of the project:
The main objective of the 'Movie Recommendation System' project is to
develop a collaborative-based recommender system that tailors movie
recommendations toindividual users by analyzing their viewing history and
preferences, while also providing a similarity-based recommender system to
suggest the top 10 movies touser preferences, enhancing their movie-watching
experience
Abstract of the project (200 words):
In today's digital era, recommendation systems have become an integral part of
our daily lives, helping us discover content that suits our preferences across
various domains like movies, books, education, and online shopping. These
systems harness the power of machine learning to analyze user profiles and
predicttheir likelihood of enjoying specific items. Our project focuses on the
developmentof a movie recommendation system that recognizes the intricate
nature of movie preferences. It takes into account a wide range of attributes,
including genre, ratings, tags, feedback, and top cast members, among others.
We aim to predict and suggest movies that align with their unique tastes by
analyzing a user's search history.What sets our recommendation system apart is
its ability to provide personalized and diverse movie suggestions by considering
combinations of two ormore attributes. To overcome the limitations of existing
systems, we introduce a novel approach that merges deep similarity techniques with
collaborative filtering.
Our goal with this project is to provide users with a robust and efficient movie
recommendation system, simplifying the process of selecting movies and ensuring
anenjoyable experience tailored to individual interests.
.
Existing System (if any)-Features and Drawbacks:
1. Limited Attribute Variety: Many existing movie recommendation systems
primarily use a narrow range of characteristics like genre or ratings to
suggest films.This can lead to a lack of diversity in recommendations, as it
doesn't capture the fullcomplexity of user preferences.
2. Lack of Personalization and Explanation: Users often encounter
recommendationsthat lack personalization, meaning the suggested movies may
not align well with their individual tastes. Additionally, these systems often fail
to provide explanations for why a particular movie was suggested, leaving
users in the dark about the reasoning behind the recommendation.
Proposed System – Features:
Our system employs a multifaceted approach to movie recommendations, taking
into account a variety of attributes, including genre, ratings, tags, feedback, and
top cast members. This comprehensive consideration of attributes leads to richer
and more nuanced movie suggestions. To enhance the user experience, the
system emphasizes personalized recommendations by closely analyzing a user's
search history and individual preferences. By combining multiple attributes, it
ensures a diverse and highly personalized selection of recommended movies.
Furthermore, to further engage users and improve satisfaction, a special feature
is in place, suggesting the top10 movies based on user feedback. Additionally,
the system incorporates a unique approach that combines deep similarity
techniques with collaborative filtering, ensuring the quality and relevance of its
movie recommendations.
Software/Hardware Requirements:
Software requirements:
Programming Language: Python
IDE: Jupyter Notebook
Packages: Pandas, Numpy, Matplotlib, Pickle, Scikit-learn, NLTK
Operating System: Windows 10
Web Technologies: HTML, CSS, JavaScript
Framework: Flask
Hardware requirements:
Processor: Intel core i3
Hard Disk:1TB
RAM: 8GB
Architecture: 32bit
or 64bit
