You are on page 1of 4

Book Recommendation System Using Machine

Learning
Dr. Rashmi Thakur, Saurabh Mishra, Danish Ali Ansari, Sachin Mishra
Computer Engineering Departmen,Thakur College of Engineering & Technology, Mumbai
rashmi.thakur@thakuredu,cation.org, mishrasaurabh847@gmail.com, ansaridanishalii@gmail.com,
sachinm1106@gmail.com
Abstract—The following section provides an overview of led to the problem of exposure to this huge data
the Book Recommendation System using Machine leading to information overloading. Hence for this very
Learning. Recommendation Systems today are used in purpose of personalization or customization,
most of the platforms to recommend services and products
Recommender systems came into the picture.
to the users. Most of the libraries and websites that
provide books are enriched with books i.e. they have a
huge number of books, but also the needs of the users are III. LITERATURE REVIEW
increasing, similarly the diversity is increasing and getting Collaborative filtering is one of the prominent
personalized. To match the user’s need and find the
techniques used in recommender systems.
desired resource at the correct time we need a
recommendation system. A recommendation system Collaborative filtering as the name itself suggests that
usually provides the users with the products and services it is a method of filtering the interests of a particular
they might like or gives a prediction on how much the user user based on collaboration among the various users,
might like the particular item. Choosing what book to various thoughts, etc. Collaborative filtering involves a
read is the question for many users. There are two large amount of user data and his/her preferences.
common approaches for solving this issue, they are Collaborative filtering is of two types namely model-
Collaborative Filtering and Content-Based Filtering. based and memory-based. Memory-based approach
So,this paper focuses on using a Hybrid approach to requires us to find the similarity measure i.e. we need
provide users with the personalized recommendation of
books. The Book Recommendation System recommends
to find the correlation between two users or a group of
books based on user parameters like age, location, user’s users while in Model-based approach a model has to be
preferred author, publication, and also similar content of created using various data mining and machine
the other books read by the user. learning algorithms.

Keywords— Book Recommendation System, Collaborative IV. OBJECTIVES


filtering, User-based filtering, Rating based filtering, Item-
based filtering, Content-Based filtering. To design an online book recommendation system that
will recommend a collection of books that the users
I. INTRODUCTION will likely wish to preview or purchase.
The popularity of recommendation systems has
increased a lot in recent times. Today many platforms V. PROPOSED SYSTEM
use recommendation systems to recommend products Collaborative filtering is been used very widely in
and services to the users. Because of the Internet, the recommendation systems to create a recommendation
needs of the users have been increased and also for the users. Our approach for the problem has four
diversified and personalized. To recommend books to steps which can be used parallelly, they are:
the users and to satisfy their needs we use a book User-Based collaborative filtering: In User-Based
recommendation system. There are two popular collaborative filtering we cluster the users based on
approaches for Book Recommendation System, and their common characteristics such as location, age, etc.
they are Collaborative and Content-Based Filtering. In After Clustering we will recommend the books read by
Collaborative filtering, the approach is to build a model the similar users in the cluster to our target user based
from a user's past behavior as well as similar decisions on some threshold frequency.
made by other users to predict items that the user may Item-Based collaborative filtering: In Item-Based
have an interest in. In Content-based filtering, it collaborative filtering we cluster the books read by the
approaches the characteristics of an item are analyzed to target user based on their common characteristics such
recommend items to the user. A recommender platform as author, publication, genre, etc. After Clustering we
with a combination of all these techniques is also will recommend the books having the same
possible. We are building a book recommendation characteristics to the target user based on some
system with a combination of all these techniques to threshold frequency.
fulfill the user’s personalized and diverse requirements. Rating Based Recommendation: In Rating Based
Recommendation we recommend the top-rated books or
the most popular books to the user based on the book's
ratings.
Content-based filtering: In Content-based filtering we
II. EXISTING SYSTEM will cluster the books based on the content and
There is a large growth in the amount of data available description of the books and then After Clustering we
on the internet coupled with the fact that there are a will recommend the books having the same
large number of internet users as well. This in turn has
EC-DSIC 2022
characteristics to the target user based on some VIII. RESULTS
threshold’s frequency.

VI. SYSTEM ARCHITECTURE


To develop a machine learning model that be flexible in
recommending books for both new and existing users.
Using this application would reduce the time required
for selecting a book to read.
Users may find new books that were previously
unknown to them. They can find what they should read
with the help of other user’s choices and making it Fig. 2 Graph showing Book Rating in terms of count
convenient for a reader to select a book of his choice
from this website.

Fig. 3 Recommendation based on Item-Based Collaborative Filtering

Fig. 1 Block Diagram of Recommendation System.

VII. ALGORITHM, TOOLS & TECHNOLOGY

A. Algorithm
I. Start
II. Get user record
III. Based on the record apply the following
filtering
IV. Apply all the types of filtering Fig. 4 Recommendation based on User-Based Collaborative Filtering
a. User-Based Collaborative Filtering
b. Item Based Collaborative Filtering
c. Rating Based Recommendation
d. Content-Based Filtering
V. Using the above recommendations from step 4
recommend books to the user
VI. Stop

B. Tools
 Colab,GCP
 Python Idle
 Web Tools Fig. 5 Rating Based Recommendation of Books

C. Technologies
 MachineLearning
 Python
 Data Science
 Web Technologies

Fig. 6 Recommendation based on Content of Title

2
Book Recommendation System Using Machine Learning
[2] J. Rathnavel, K. Kelkar, “Personalized Book Recommendation
System”, International Journal of Engineering And Computer
Science,vol. 6, no. 4, Apr. 2017.
[3] N. Kurmashov, K. Latuta and A. Nussipbekov, "Online book
recommendation system," 2015 Twelve International Conference
on Electronics Computer and Computation (ICECCO), Almaty,
2015, pp. 1-4, doi: 10.1109/ICECCO.2015.7416895.
[4] N. Ifada, I. Syachrudin, M. Sophan, S. Wahyuni, “Enhancing the
Performance of Library Book Recommendation System by
Employing the Probabilistic-Keyword Model on a Collaborative
Filtering Approach”, 2019, Procedia Computer Science, 157.
345-352.10.1016/j.procs.2019.08.176.
[5] S. Kanetkar, A. Nayak, S. Swamy and G. Bhatia, "Web-based
personalized hybrid book recommendation system," 2014
Fig. 7 Recommendation based on Content of Description International Conference on Advances in Engineering &
Technology Research (ICAETR - 2014), Unnao, 2014, pp. 1-5,
IX. CONCLUSION doi:10.1109/ICAETR.2014.7012952.
[6] K. Shah, A. Salunke, S. Dongare and K. Antala, "Recommender
This system aims to provide a personalized systems: An overview of different approaches to
recommendation of books to the user. The system tries recommendations," 2017 International Conference on Innovations
to predict the books by considering the item's similarity in Information, Embedded and Communication Systems
as well as the user's similarity so that a user can get (ICIIECS), Coimbatore, 2017, pp. 1-4.doi:
10.1109/ICIIECS.2017.8276172
recommendations for new books. This system will [7] Y. Liang and S. Wan, "The Design and Implementation of Books
effectively recommend quality books to old as well as Recommendation System," 2018 IEEE 9th International
new users and enhance their reading experience. Conference on Software Engineering and Service Science
(ICSESS), Beijing, China, 2018, pp. 305-308.doi:
10.1109/ICSESS.2018.8663914
ACKNOWLEDGMENT [8] R. Darekar, K. Dayma, R. Parabh and S. Kurhade, "A Hybrid
This paper on ‘Book Recommendation System’ has Model for Book Recommendation," 2018 Second International
been possible only because of kind co-operation lent by Conference on Inventive Communication and Computational
Technologies (ICICCT), Coimbatore, 2018, pp. 120-124.doi:
my teacher and project guide Dr. Rashmi Thakur 10.1109/ICICCT.2018.8473274
without which this would not have been possible. We [9] J. Li, T. W. Xu and J. Zhou, "A personalized Recommendation
would also like to thank our parents, who have provided Algorithm for College Books based on User Interest," 2020 IEEE
us with all possible resources to gain the best possible 2nd International Conference on Computer Science and
Educational Informatization (CSEI), Xinxiang, China, 2020, pp.
knowledge. At last, we would like to thank Dr. B.K. 144-149.doi: 10.1109/CSEI50228.2020.9142507
Mishra (Principal), Dr. HarshaliPatil (Computer [10]H. Liu and N. Jiao, "A Hybrid Book Recommendation Algorithm
department, HOD) for their guidance and support. Based on Context Awareness and Social Network," 2020 3rd
International Conference on Advanced Electronic Materials,
Computers and Software Engineering (AEMCSE), Shenzhen,
References
China, 2020, pp. 554-561.doi:
[1] Ms. S. Rajpurkar, Ms. D. Bhatt and Ms. P. Malhotra. "Book 10.1109/AEMCSE50948.2020.00122.
Recommendation System" International Journal for Innovative
Research in Science & Technology, Volume 1, Issue 11, April
2015, Page 314-316.

3
EC-DSIC 2022

You might also like