ABSTRACT

Every year new movies are released with a varied story-line or a genre which could
be of potential interest to viewers. Various online movie or video streaming platforms
can keep the customers engaged by recommending movies of the viewer's
preference. Hybrid Recommendation System for Movies uses the combination of
collaborative and content based filtering in the context of web-based recommender
systems. Recommendation systems constitute a specific type of information filtering
that attempt to present items according to the interests expressed by a user , in order
to reduce the effect of such dependencies. Our project proposes a hybrid
recommendation system with sentiment analysis of movie tweets. The movie tweets
have been collected from micro blogging websites to understand the current trends
and user response of the movie. Through different investigations, we will show the
ability to expand utilization, just as the qualities of how to foresee blended longings.
Ideas for particular kinds of channel data endeavor to decide factors dependent on
the business' advantages to limit the effect of protection. Our venture requires an
arrangement of correspondence and passionate examination of tweets. The film's
tweets have been gathered on a microblog webpage to stay aware of the most recent
patterns and arrangements from film clients. Leading examination in the state files
gives positive outcomes.

v
 Table of Content

Chapter TITLE Page No.

No.

ABSTRACT v

1 INTRODUCTION 1

1.1 RELEVANCE OF THE PROJECT 1

1.2 PROBLEM STATEMENT 2

1.3 OBJECTIVES OF THE PROJECT 2

1.4 SCOPE OF THE PROJECT 2

2 LITERATURE REVIEW 3

2.1 RELATED WORK 3

3 METHODOLOGY 5

3.1 EXISTING WORK 5

3.2 PROPOSED WORK 5

3.3 SYSTEM REQUIREMENT SPECIFICATION 6

3.3.1 HARDWARE REQUIREMENTS 6

3.3.2 SOFTWARE REQUIREMENTS 6

3.4 SOFTWARE SPECIFICATION 7

3.4.1 FRONT-END SPECIFICATION 7

3.4.2 BACK-END SPECIFICATION 12

3.5 SYSTEM ANALSIS AND DESIGN 15

3.5.1 SYSTEM ARCHITECTURE 15

3.5.2 ACTIVITY DIAGRAM 16

3.5.3 DATA FLOW 16

vi
 3.6 RECOMMENDER SYSTEM TYPES AND TECHNIQUES 17

3.6.1 ENTITIES IN RECOMMENDATION SYSTEM 17

3.6.2 COLLABORATIVE FILTERING 20

3.6.3CONTENT-BASED RECOMMENDATION 21

3.6.4 HYBRID RECOMMENDATION AND APPROACH 23

3.7 DATA ANALYSIS 25

3.7.1 DATA COLLECTION 25

3.7.2 DATA PRE-PROCESSING 27

3.8 IMPLEMENTATION 27

3.8.1 COSINE SIMILARITY 27

3.8.2 KNN 28

3.8.3 SVD 29

4 RESULTS AND DISCUSSION 30

5 CONCLUSION AND FUTURE SCOPE 32

5.1 CONCLUSION 32

5.2 FUTURE SCOPE 32

APPENDICES 34

A.SOURCE CODE 34

B.SCREENSHOTS 39

C.PUBLICATION WITH PLAGARISM REPORT 42

vii
 LIST OF FIGURES

FIGURE TITLE PAGE NO

NO

3.1 ARCHITECTURE OF HYBRID APPROACH 15

3.2 ACTIVITY DIAGRAM 16

3.3 DATA FLOW DIAGRAM 16

3.4 EXAMPLE OF TAXONOMY OF THE RECOMMENDATION SYSTEMS 19

3.5 RECOMMENDATION TECHNIQUE 20

3.6 OVERVIEW OF RECOMMENDATION SYSYTEMS 21

3.7 CLASSIFICATION ABOUT NEW SAMPLE 28

B SCREENSHOTS 39

viii
 LIST OF TABLES

TABLE TITLE PAGE NO

NO

3.1 DIFFERENT HTTP METHODS 14

4.1 COMPARISON BETWEEN THE THREE APPLICATIONS 31

ix
 CHAPTER 1
INTRODUCTION
1.1 RELEVANCE OF THE PROJECT

A recommendation system or recommendation engine is a model used for

information filtering where it tries to predict the preferences of a user and provide
suggests based on these preferences. These systems have become increasingly
popular nowadays and are widely used today in areas such as movies, music, books,
videos, clothing, restaurants, food, places and other utilities. These systems collect
information about a user's preferences and behaviour, and then use this information
to improve their suggestions in the future.

Movies are a part and parcel of life. There are different types of movies like
some for entertainment, some for educational purposes, some are animated movies
for children, and some are horror movies or action films. Movies can be easily
differentiated through their genres like comedy, thriller, animation, action etc. Other
way to distinguish among movies can be either by releasing year, language, director
etc. Watching movies online, there are a number of movies to search in our most liked
movies .

Movie Recommendation Systems helps us to search our preferred movies

among all of these different types of movies and hence reduce the trouble of spending
a lot of time searching our favourable movies. So, it requires that the movie
recommendation system should be very reliable and should provide us with the
recommendation of movies which are exactly same or most matched with our
preferences. A large number of companies are making use of recommendation
systems to increase user interaction and enrich a user's shopping experience.
Recommendation systems have several benefits, the most important being customer
satisfaction and revenue. Movie Recommendation system is very powerful and
important system. But, due to the problems associated with pure collaborative
approach, movie recommendation systems also suffers with poor recommendation
quality and scalability issues.

1
1.2 PROBLEM STATEMENT

The goal of the project is to recommend a movie to the user. Providing related
content out of relevant and irrelevant collection of items to users of online service
providers.

1.3 OBJECTIVE OF THE PROJECTS

 Improving the Accuracy of the recommendation system.

 Improve the Quality of the movie Recommendation system.
 Improving the Scalability.
 Enhancing the user experience

1.4 SCOPE OF THE PROJECT

The objective of this project is to provide accurate movie recommendations to

users. The goal of the project is to improve the quality of movie recommendation
system, such as accuracy, quality and scalability of system than the pure approaches.
This is done using Hybrid approach by combining content based filtering and
collaborative filtering, To eradicate the overload of the data, recommendation system
is used as information filtering tool in social networking sites. Hence, there is a huge
scope of exploration in this field for improving scalability, accuracy and quality of
movie recommendation systems Movie Recommendation system is very powerful and
important system. But, due to the problems associated with pure collaborative
approach, movie recommendation systems also suffers with poor recommendation
quality and scalability issues.

2
 CHAPTER 2

LITERATURE SURVEY

2.1.RELATED WORK

 The work of Nandagawali and Patil [1] provides a solution for the various
recommender systems by using collaborative filtering algorithms with the
community based user domain model. The main purpose is to satisfy the
customer„s product needs by providing them recommendation based on
products. It deals with the Amazon dataset. Better hardware configuration may
result in better output.
 The work of Jatinder et al., [2] deals with the review of several techniques that
are used by a system for recommending electronic products. They have also
worked with the Amazon dataset. Time complexity of the whole model is pretty
high, making it less convenient.
 In the work of Eklaspur and Pashupatimath, [3] a formal representation of
social network is present where text mining is taken as a perspective. A
framework is proposed that can recommend friend using an efficient Algorithm.
Their work is based on Facebook data. The notable points in their work can be
that the accuracy is high, and works fine with any size of data. Analysis of the
classes and its classification is also highly précised. But it lacks to meet the
standard when it comes to time of execution.
 Haruna, et al., [4] in their work, have proposed a research paper recommender
system that transforms all the recommending papers into a paper-citation
relations matrix. The key points that makes their model better can be faster
results. Their algorithm which is Brute Force based, but fails to meet the
compatibility in real time application as it will be inefficient for very large
dataset, as the number of rows to column ratio will increase abnormally.
 In the research work carried out by Kumar et al., [5] they have proposed a
movie recommendation system named MOVREC. The system sorts the ratings
by implementing K-means clustering algorithm. The pros of the system include

3
 that the time taken is less, and almost every time the result of the classification
of data is accurate [21][25]. But the cons can be that creating classes amongst
closely related data which may belong to the same cluster can be redundant
and misleading.
 According to the work of Chen et al. [6], they have incorporated Singular Value
Decomposition (SVD) in user based collaborative filtering technique for movie
recommendation system. Although, they have got good accuracy with their
model, but additional user or item features can be applied along with a larger
dataset for better preparation of model.
 The work of Reddy, et al. [7] deals with genre correlation-oriented content-
based movie recommendation system, that only focusses on the genre of a
movie. On the contrary, for any person, choice might not be completely based
on one genre.
 In [8], C. S. M. Wu, D. Garg, and U. Bhandary proposed a recommendation
system using collaborative filtering where a user‟s rating is used to suggest the
list. The authors have used the Apache Mahout framework and essentially
compared the performances and efficiency of us er-based & item-based
recommendations.
 In [9],R. E. Nakhli, H. Moradi, and M. A. Sadeghi proposed the percentage
view approach for recommending movies to the users, it finds relevant movies
for the customer and then compares the performance with a random movie
recommendation system for showing the accuracy of the project.
 In [10], a content-based recommendation system is proposed by H. W. Chen,
Y. L. Wu, M. K. Hor, and C. Y. Tang using neural networks. In recent years,
these are top topics for the researchers to work on when they want to build a
movie recommendation system.

4
 CHAPTER 3

METHODOLOGY

3.1 EXISTING WORK

In the existing system, collaborative filtering and content based filtering have
been used.Collaborative filtering system maintains a database of many users ratings
of a variety of items. It makes use of user data, ignoring content/item data.Content
based filtering uses only the item data maintaining a profile for each item. Each user
assumed to operate independently . No need for data on other users.

Disadvantages of existing work:

Disadvantages of Collaborative filtering:

 Cold start.
 Sparsity.
 First rater.
 Popularity bias.

Disadvantages of content based filtering:

 Cannot filter items on some assessment of quality, style or viewpoint because

of lack of consideration of other people.
 Absence of personal recommendations.
 No serendipitous.

3.2 PROPOSED WORK

In the proposed hybrid recommender system takes advantage from both the
representation of the content as well as the similarities among users. The intuition
behind this technique is to use a content-based predictor to fill the user-rating matrix
that is sparsely distributed.

5
 A web crawler is used to download necessary movie content for our dataset.
After the preprocessing the movie content database is stored. The dataset consists of
a user-rating matrix.

Content-based predictions are used to train each user-rating vector in the user-
rating matrix and convert it into a pseudo rating matrix which combines actual rating
with the predicted ratings. Collaborative filtering is then applied to this full pseudo
user-rating matrix to make recommendation for an active user.The movie tweets have
been collected from microblogging websites to understand the current trends and
user response of the movie.

3.3 SYSTEM REQUIREMENT SPECIFICATION

This involves both the hardware and software requirements needed for the
project and detailed explanation of the specifications.

3.3.1 Hardware Requirements

 A PC with Windows/Linux OS
 Processor with 1.7-2.4gHz speed
 Minimum of 8gb RAM
 2gb Graphic card

3.3.2 Software Requirements

 Text Editor (VS-code/WebStorm)

 Anaconda distribution package
 Python libraries

Anaconda distribution:Anaconda is a free and open-source distribution of the Python

programming languages for scientific computing (data science, machine learning
applications, large-scale data processing, predictive analytics, etc.), that aims to
simplify package management system and deployment. Package versions are
managed by the package management system conda. The anaconda distribution
includes data-science packages suitable for Windows, Linux and MacOS.3