International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169

Volume: 4 Issue: 7 267 - 269

____________________________________________________________________________________________________________________
Recommendation Item Based on Keyword Search using Big Data

Jafar Sadik Kamadod Prof. Shrivatsa Koulgi

Mtech IT, SDMCET, Dharwad SDMCET, Dharwad
Email: m.jafarsk@gmail.com Email: shrivatsa.koulgi@gmail.com

Abstract- Design the methodology for piloting the Book Recommendation System. Provide responses on the suitability of Books with respect to
content and likes by users. Determine the suitability of existing technology whether there is need for other recommendations. Moreover, to
improve the scalability and efficiency of service recommendation system in Big Data environment, we have implemented it on a Map Reduce
framework in Hadoop platform. Finally, we recommend books to users based on similarity.

Keywords-web search, big data, mapreduce, collaborative filtering,recommendation system.

__________________________________________________*****_________________________________________________
I. INTRODUCTION
With the growing number of alternative online
services, recommending services that users preferred
effectively has become an important research issue. Service
recommender systems have shown to be valuable tools to help
users deal with services overload and provide appropriate
recommendations to them. Examples of such practical
applications include CDs, books, web pages and various other
products now use recommender systems [1], [2].
In recent years, the amount of data in our world has
been increasing explosively and analyzing large data setsso
calledBig Databecome a key basis of competition
underpinning new waves of productivity growth, innovation,
and consumer surplus [3]. Then, what is Big Data?, Big data
refers to data sets whose size is beyond the ability of current
technology, method and theory to capture, manage, and process
the data within a tolerable elapsed time. Today, Big Data
management stands out as a challenge for IT companies. The
solution to such a challenge is shifting increasingly from
providing hardware to provisioning more manageable software
solutions [4].
A search system is designed to search required
information like searching a book using book title, using the
searched data, recommend the book to the user with similar
taste. To recommend the books we use collaborative filtering
approach.
keyword to see if there are similar books. We believe that the
abandoned keywords cover a lot of book demands
information. These demands can serve as a reference for
purchasing books.
A. COLLABORATIVE FILTERING:
It is a technique used by some recommender systems.
Collaborative filtering has two senses, a narrow one and a more
general one. In general, collaborative filtering is the process of
filtering for information or patterns using techniques involving
collaboration among multiple agents, viewpoints, data sources,
etc. [5]. Applications of collaborative filtering typically involve
very large data sets. Collaborative filtering methods have been
applied to many different kinds of data including: sensing and
monitoring data, such as in mineral exploration, environmental
sensing over large areas or multiple sensors; financial data,
such as financial service institutions that integrate many
financial sources; or in electronic commerce and web
applications where the focus is on user data, etc. The remainder
of this discussion focuses on collaborative filtering for user
data, although some of the methods and approaches may apply
to the other major applications as well [6].
IV. METHODOLOGY/ PLANNING OF WORK:
The recommendation system based on keyword search works
solely on the basis of explicit user feedbacks and ratings.

II. RELATED WORK

In the recent years the amount of data has increased in World
Wide Web that yields to big data problem. In Earlier
approach we face problems of scalability and efficiency
because of increasing data. Earlier the recommendation system
is focused on ranking and rating of the user to provide the
services. Now purposed system focused on item of service. The
collaborative filtering technique is used to implement the
recommendation of the active user, The Map Reduce
framework is used to calculate ratings in parallel to increase the
scalability and efficiency of the system because of the large
dataset.

III. BOOK RECOMMENDATION MODEL

Recommendation system grounded itself on the traditional
bibliographic retrieval system. Readers used to look up a book
using the retrieval system. Sometimes the system will return an
empty set, which means there is no book that contains the
readers keyword. In this case, readers have to change the Fig.1 Architecture

267
IJRITCC | July 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 7
____________________________________________________________________________________________________________________
The main steps of Recommendation system based on keyword
search are depicted in Fig. 1, which are described in detail as
follows:
A. Capture user preferences by a keyword-aware
approach: In this step, the preferences of active users and
previous users are formalized into their corresponding
preference keyword sets respectively. An active user refers to a
current user needs recommendation.
1) Preferences of a current user. A current user can give
his/her preferences about candidate services by selecting
keywords from a keyword-candidate list or Search a book
2) Preferences of existing users. The preferences of a previous
user for a candidate service are extracted from his/her reviews
for the service according to the keyword-candidate list or
search a book for recommendation
B. Preprocessing
The reviews of previous users are stored into the
Dataset. The Dataset was downloaded from the informatics
site. Preprocessing is carried out on a Dataset and gets the
previous user rating.
C. Similarity computation. The second step is to identify the
reviews of previous users who have similar tastes to an active
user by finding neighborhoods of the active user based on the
similarity of their preferences. Before similarity computation,
the reviews unrelated to the active users preferences will be
filtered out by the intersection concept in set theory. If the
intersection of the preference keyword sets of the active user
and a previous user is an empty set, then the preference
keyword set of the previous user will be filtered out.
sim (APK, PPK) = Jaccard(APK, PPK) = |APKPPK|
|APKPPK|
D. Compute ratings and generate recommendations. The
rating can be calculated based on the similarity of the active
user and previous users, further filtering will be conducted and
calculate the personalized ratings and collects the highest
ratings using nearest neighbors algorithm.
Top S with nearest neighbors algorithm
G a,k= ra + S(ru,k-ru)xW(s1,s2)
S(s1,s2)
E. Items:
Items are the objects that are recommended. Items may be
characterized by their complexity and their value or utility. The
value of an item may be positive if the item is useful for the
user or negative if the item is not appropriate and the user made
a wrong decision when selecting it. We note that when a user is
acquiring an item she will always incur in a cost, which
includes the cognitive cost of searching for the item and the
real monetary cost eventually paid for the item.
F. Users:
Users of a recommended system may have very diverse goals
and characteristics. In order to personalize the
IJRITCC | July 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
ISSN: 2321-8169
267 - 269
recommendations and the human-computer interaction, RSs
exploit a range of information about the users. This information
can be structured in various ways and again the selection of
what information to model depends on the recommendation
technique.
G. Pearson correlation coefficient algorithm
Correlation between the set of important items is calculated
using the Pearson coefficient algorithm.
Sim(APK, PPK)=kk (ra,k ra)(rp,k rp)
kk (ra,k ra) kk (rp,k rp)
H. Map Reduce
It is used to process the input in parallel way. Here we compute
similarity and generate the highest rated item and calculate the
ratings and ranking in parallel. [7]
V. MODULES
A. Login and book details
The Login Form module presents site visitors with a form with
username and password fields. If the user enters a
validusername/password combination they will be granted
access to additional resources on your application.
Whichadditional resources they will have access to can be
configured separately [8].
In this module admin can add new Book titles and Images
details. These details will add tothe existing details. User can
select the books details added in this module and they will rate
the Book based on their reviews. These details are used for
cluster the data based on their ratings [9].
B. MapReduce Programming Model
1) Map, written by the user, takes an input pair and
produces a set of intermediate key/value pairs. The MapReduce
library groups together all intermediate values associated with
the same intermediate key I and passes them to the Reduce
function.
2) It processes the reviews of the existing user with their
preference and compute average rating for each user.
3) The Reduce function, also written by the user, accepts
an intermediate key I and a set of values for that key. It merges
together these values to form a possibly smaller set of
values.Typically just zero or one output value is produced per
Reduce invocation. The intermediate values are supplied to the
users reduce function via an iterator. This allows us to handle
lists of values that are too large in memory.
4) The Reduce task calculates the personalized ratings
and recommends the books with similar taste.
5) Map Reduce Specification Object. Contains names of
input/output less and optional tuning parameters. The user then
invokes the MapReduce function, passing it the specification
object. The users code is linked together with the MapReduce
library.
268
International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 4 Issue: 7 267 - 269
____________________________________________________________________________________________________________________
Example: In table1 the three users rated the books commonly,
which means they have similar taste, but the user not rated the
book (HTML), which means he/she hasnt saw that book. So
by the ratings of user A and C recommend the book to user B.

VII. CONCLUSION
In this paper, we propose Online Book Recommendation
System Based on Keyword Search Using Big Data Analysis,
keywords are used to indicate users preferences, and a user
based Collaborative Filtering algorithm is adopted to generate
appropriate recommendations. More specifically, The active
user gives his/her preferences by selecting the keywords from
the keyword-candidate list, and the preferences of the past
users can be extracted from their reviews and recommending
the most appropriate service(s) to the users. Moreover, to
improve the scalability and efficiency of our approach in big
data environment, we implement it in a Map-Reduce
framework on Hadoop platform.

Fig.2 MapReduce implementation model
6) MapReduce computation done parallel to improve the
performance of the system
7) MapReduce work on key-value pairs, Key i and value
i is the pairs of existing user reviews input to MapReduce
framework, where key j and value j are the pair input of the
current users. Using this MapReduce compute the rating in
parallel to increase the scalability and efficiency of the
recommendation system.
VI. REQUIREMENTS ON ONLINE BOOKS
RECOMMENDATION
New user must register first to get one users account ID. The
given information includes username, first name, last name,
email address and password. After registration the registration
details are stored in database, which we can retrieve the
information at the time of login. One user can only have only
one account.
The book details are identified by ISBN. For each book, the
system also needs to record its authors name, title, edition,
year, category, publisher etc.
After login the user search for books using keyword,
collaborating filtering technique is used to recommend the user
with similar taste that the user is looking for the record.
TABLE I.
USER- ITEM RATINGS
Books Bytes of Bokeh Pandas Ipython HTML
Users Python
VIII. REFERENCES
[1] G. Linden, B. Smith, and J. York, Amazon.com
Recommendations: Item to-Item Collaborative Filtering,
IEEE Internet Computing, Vol. 7, No.1, pp. 76-80, 2003.
[2] B. N. Miller, I. Albert, S. K. Lam, J. A. Konstan, and J.
Riedl, MovieLens Unplugged: Experiences with an
Occasionally Connected Recommender System, In IUI 03
Proceedings of the 8th International Conference on
Intelligent User Interfaces, pp.263-266, 2003.
[3] J. Manyika et al., Big Data: The Next Frontier for
Innovation,Competition, and Productivity, 2011.
[4] C. Lynch, Big Data: How Do Your Data Grow? Nature,
vol. 455,no. 7209, pp. 28-29, 2008.
[5] K.G.S. Venkatesan, Dr. Kathir. Viswalingam, N.G. Vijitha,
Associate Adaptable Transactions Information store in the
cloud using Distributed storage and meta data manager,
International Journal of Innovative Research in computer &
communication Engineering, Vo l . 3 , Issue3 , PP. 1548
15 55, March 2015
[6] M. A. Beyer and D. Laney, The importance of big data:
A definition, Gartner, Tech. Rep., 2012.
[7] Y. Zhu and Y. Hu. 2006. Enhancing search performance on
Gnutella-like P2P systems, IEEE Transactions on Parallel
and Distributed Systems, Vol. 17, No. 12, pp. 1482-1495.
[8] K.G.S. Venkatesan and M. Elamurugaselvam, Using the
conceptual cohesion of classes for fault prediction in object-
oriented system, International journal of Advanced &
Innovative Research, Vol. 2, Issue 4, PP. 75 80, April
2013.
[9] K.G.S. Venkatesan, R. Resmi, R. Remya, Anonymizimg
Geographic routing for preserving location privacy using
unlinkability and unobservability, International Journal of
Advanced Research in computer science & software Engg.,
Vol. 4, Issue 3, PP. 523 528, March 2014.

A 8 8 8 8 9
B 9 8 8 9 -
C 8 7 8 10 10

269
IJRITCC | July 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________