International Journal of Engineering Science and Computing, April 2017 10942 http://ijesc.org/
ISSN XXXX XXXX © 2017 IJESC
Hybrid Book Recommendation System
Yash Wani
1
, Darsh Bakshi
2
, Vinesh Desai
3
, Sheetal Pereira
4
UG Student
1, 2, 3
, Assistant Professor
4
Department of Computer Engineering
K.J Somaiya College of Engineering, Vidyavihar, Mumbai, India
Abstract:
Internet has been overwhelmed with mass amount of information. In order to provide necessary and accurate information to the
user filtering mechanism were used. Recommendation systems are filtering systems which provide personalization in terms of the
information coming to a user based on the similarities, interests, relevance of the information etc. Recommender systems are used
widely for recommending movies, articles, restaurants, places to visit, items to buy etc. Its main goal is the management of the
information overload. We propose a Book Recommendation system which integrates content based and collaborative filtering
approaches. In Content based approach system uses keywords in order to find similar books, while in Collaborative filtering data
is distributed evenly across all peers. In this system we attempt to present to the user various books the user is interested in
filtering various redundant books using above two methodologies.
Keywords: Collaborative filtering, content based filtering, Apriori algorithm, Karl Pearson algorithm
I. INTRODUCTION
With the rapid popularization of the web services, Internet has
begun overflowing with information. Among the many web
resources, users do not know how to quickly find the resources
they really need. Therefore, a user specific recommendation
system becomes an urgent demand product. Related research
has become an important topic in this field of research and has
gained wide attention of many research pioneers.
Recommendation system [1] gets the reader’s interest hobby by
online analyzing, recommendations. Collaborative filtering [1]
[2] is one of the most important technologies working towards
recommendations. eg grouplens, its can automatically provide
their users with information which comes from people with
similar interests. Users make evaluation to books and the
system also provides users with a list of books that comes from
other users. It is different from recommended system that needs
to show the user’s evaluation information which comes from
people with similar interests. Content-based filtering [2] 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.
II.LITERATURE SURVEY
According to survey, techniques used for recommendation are
classified on the basis of knowledge sources. For instance,
Collaborative Filtering technique works on user-item
preference data, Content-based technique [2] is based on item
features. Collaborative Filtering [1] [2] technique filters out the
recommendation with the help of user behavior in the form of
ratings. As mentioned in the paper we are going to implement
collaborative filtering with the help of finding Pearson
correlation ratio and applying nearest neighbor algorithm. This
technique generates rating for an unrated item for the user,
based on the commonalities among users and their ratings.
Recommendation quality is directly proportional to the size of
rating dataset. Content-based technique uses item features and
user preference to provide recommendations. In this technique,
attributes like genres, authors etc. are used to describe items,
while user rating indicates the items liked by the user. It
recommends items that are similar to the items preferred by the
user. As mentioned in the papers the attribute matrix formed
can be simplified by using matrix factorization by using
singular value decomposition.
Hybrid systems:
Combines two or more recommendation techniques to predict
recommendation. Using Hybrid technique, it is possible to
overcome the drawbacks set by one recommendation technique
and sum up advantages of different recommendation
techniques. For example, Collaborative Filtering technique
have problem when limited user-item ratings are available
whereas Demographic and Content-based technique do not use
rating data and therefore can overcome cold start problem. [1]
There are various ways to combine recommendation
techniques to achieve effective hybrid recommendation.
III.PROPOSED SYSTEM
We propose to build a website for book recommendation using
Content Based and Collaborative Filtering technique. We will
be making an attribute matrix for content based filtering and
apriori algorithm [3] for Collaborative filtering. We propose to
build a website for book recommendation using Content Based
and Collaborative Filtering technique. [2]
User Input:
While making an account the user will be asked to give his/her
details like name, age etc. Also the user can rate books that
they like.
Output:
According to the data and the system the user will get book
recommendations which will be personal.
Research Article Volume 7 Issue No.4