Keyword Based Service Recommendation system for Hotel System using Collaborative Filtering

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 02 | Feb-2016 www.irjet.net p-ISSN: 2395-0072
© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 788
Keyword Based Service Recommendation system for Hotel System
using Collaborative Filtering.
Ms. Anjali Landge1, Ms. Apurva Randive2, Ms. Ankita Ambi3, Ms. Priyanka Alagoudar4,
Prof.Mr. G.I.Rathod5
1Student of CSE, in Dr.J.J.Magdum College of Engineering Jaysingpur , Maharashtra, India
2 Student of CSE, in Dr.J.J.Magdum College of Engineering Jaysingpur , Maharashtra, India
5Assistant Professor, Dept. of Computer Science and Engineering, Dr.J.J.Magdum College of Engineering Jaysingpur ,
Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - To provide appropriate recommendation to the
user, service recommender system is a valuable tool. Fromlast
few years, increased number of customer, services and online
information has grown widely, so that big data analysis
problem has been occurred for service recommendation
system. In traditional service recommender systems often
suffer from scalability and inefficiency problems when
processing or analyzing such large-scale data. The existing
service recommender systems fails to meetusers‘personalized
requirements’ because of there is presence of the same ratings
and rankings of services to different users withoutconsidering
diverse users’ preferences.
In this paper, we propose a Keyword-Based Service
Recommendation system, to resolve the above challenges. It
provides a personalized service by consideration of not only
recommendation list but also recommending the most
appropriate services to the users effectively. To generate
appropriate recommendations, user’s preferences as well as
user-based Collaborative Filtering algorithm is adopted.
Key Words: Big data, big table, keywords,
recommendation system, thesaurus, user preference
1. INTRODUCTION
The amount of data in our world has been increasing
explosively and analyzing large data sets is called “Big Data”
become a key basis of competition underpinning newwaves
of productivity growth, innovation, and consumer
surplus[1]. Big data is a broad term for data sets so large or
complex that traditional data processing applications are
inadequate. Challenges include analysis,
capture, data creation, search, sharing, storage, transfer,
visualization, and querying and information privacy. 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 [2]. Big data also brings new
opportunities and critical challenges to industry and
academia [3] [4]. Similar to most big data applications, the
big data tendency also poses heavy impacts on service
recommender systems. With the increasing number of
alternative services, effectively recommending servicesthat
users preferred have become an important research issue.
Service recommender systems have been shownasvaluable
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[5][6][7]. Over the last decade, there has been much
research done both in industry and academia on developing
new approaches for service recommender systems.
To provide appropriate recommendation to the
user, service recommender system is a valuable tool. From
last few years, increased number of customer, services and
online information has grown widely, so that big data
analysis problem has been occurred for service
recommendation system. In traditional service
recommender systems often suffer from scalability and
inefficiency problems when processing or analyzing such
large-scale data. The existing service recommendersystems
fails to meet users ‘personalized requirements’ because of
there is presence of the same ratings and rankings of
services to differentuserswithoutconsideringdiverseusers’
preferences. Motivated by these observations, in this paper,
we address these challenges through the following
contributions:
 A keyword-based service recommendationmethod
is proposed in this paper, which is based on a user-
based Collaborative Filtering (CF) algorithm.

 In this, keywords extracted from reviews of
previous users are used to indicate their
preferences [13].
Recommendations are offered as ranked list of items. In
performing this ranking, recommender systems try to
predict what the most suitable products or services are,
based on the user’s preferences and constraints.
2. LITERATURE REVIEW
While processing or analyzing large scale data the
service recommendation system suffer from the scalability
and inefficiency. In the current existing recommender
system provides rating and ranking of services without
bothering of diverse user’s preference list .Because of this
reason they are failed to meet users personalized
requirement. Moreover, in most existing service
recommender systems, such as hotel reservation systems
and restaurant guides, the ratings of servicesandtheservice
recommendation lists presented to users are thesame.They
have not considered users’ different preferences, without
meeting users’ personalized requirements.
Recommendation carried out by different ways:
2.1 Content-based approaches.
Content based recommenders provide
recommendations by comparing representation of contents
describing an item or a product to the representation of the
content describing the interest of the user. They are
sometimes referred to as content based filtering[9]. Content
based recommendation system is nothing but to get form
from the content of item. In most of the cases they are the
words which portray the item.Thisrecommendationsystem
is able to introduce new item for the user. It recommending
item along with explanation of item to the user. The main
motto of this method is to analyze the bunch of document of
items, which is rated by the user and construct a model
according to the user interest which is based on feature of
the items. Information retrieval and information filtering
both are the source of content based recommending
approach. It provides only textual information like as news
webs and documents. No content-based recommendation
system can provide suitable suggestions if the analyzed
content does not contain enough information to
discriminate items the user likes from items the user does
not like[9]. Some representations get only certain views of
the content, but there are many others that would influence
a user’s experience.
2.2 Collaborative recommendation approaches.
Collaborative filtering approaches recommend
services [5] is well known Preferred by user in the past.
Collaborative filtering method is most important method in
recommendation system, in which it finds a set of people
who shares same interest with you. These people could be
determined by the similar ranking on items. These people
identified by neighborhoodofthecurrentuser.Collaborative
filtering use’s “Netflix”. If any recommendation systemgives
rating on item then it might be using collaborative filtering
method. First problem is that result of recommendation
system will be depends on presence of information. In the
relationship mining, new items not-yet-rated can be
abandoned in the recommendation processes. Second
problem is that, the collaborative filtering unable to cover
the extreme case. Similarity decisions are unable to be
established when the scales of the users are small or the
users have unique taste. Update frequency is the third
problem. If any new information of users has to be included
in the recommendation processes in real time, data latency
will increase the waiting time for the query result [5].
2.3 Big data.
Big data is nothing but one kind of the dataset whose
size is greater than that of the typical database which isused
to capture, store, manage and analyze the data. Sweating of
data is carried out in every industry and business so that it
becomes important factor of production. Now days, the use
of big data will be responsible for competition andgrowth of
individual firm [10]. Due to wide use of digital technology ,
there is rapidly growth in the digital data in every
organization .Digital network contain the large number of
users, devices and sensors which areinterconnectedtoeach
other and as time passes they are continuously risingsothat
ability to generate, communicate ,share and access the data
has been revolutionized.
2.4 Big Table
Big table is a distributed storage system for
managing structured data that is designed to scale to a very
large size: peta bytes of data across thousandsofcommodity
servers. Many projects at Google store data in Big table,
including web indexing, Google Earth, and Google Finance
[8]. Data in the table is organized in three dimensions which
are row, column and timestamp.
3. METHODOLOGY
3.1 Introduction
Keyword- based service recommendation method
keywords are used to indicate bothofusers’preferencesand
the quality of candidate services. A user-based CF algorithm
is adopted to generate appropriate recommendations. It
aims at calculating a personalized rating of each candidate
service for a user, and thenpresentinga personalizedservice
recommendation list and recommending the most
appropriate services to user. Just consider service as hotel
reservation system. As shown in Fig 1 while reserving any
hotel we are considering so many things like room service,
quality of food, cleanliness of hotel and its environment, etc.
Requirements of hotels are depending on customer, as
person changes there requirement also changes. Some
people more concern about cleanliness of hotel not
bothering about value, transportation facilities. But on the

other hand some people interestinvalue.Rectanglescontain
main keywords and oval contain related words of that
keyword. The user may puttherequirementas “he/she want
mall near to hotel” instead of “shopping”. So that, we are
applying domain treasures on the keyword. Because of that
we can easily get the related words of keywordwhichwill be
our candidate list1.
Fig -1: Hotel System
3.1.1 Keyword-candidate list:
The keyword-candidate list is a set of keywords aboutusers’
preferences and multicriteria of the candidate services,
which can be denoted as K= {k1; k2;…kn}; n is the numberof
the keywords in the keyword-candidate list. Anexampleofa
simple keyword-candidate list of the hotel reservation
system is described. Keywordsinthekeyword-candidate list
can be a word or multiple words related with the quality
criteria of candidate services[13].
3.1.2 Domain thesaurus:
A domain thesaurus is a reference work of the
keyword-candidate list that lists words grouped together
according to the similarity of keyword meaning, including
related and contrasting words and antonyms[12][13]. An
example of a simple domain thesaurus of hotel reservation
system .The words in the red rectangle are the keywords in
the corresponding keyword-candidate list, and the words in
the ovals are the related words of the keywords. Often,
domain thesauruses are updated regularly to ensure the
timeliness of the words.
3.2 Proposed Work
In this project, we proposea keyword-basedservice
recommendation method,. In this, keywords are used to
indicate users’ preferences, and a user based. Collaborative
Filtering algorithm is adopted to generate appropriate
recommendations. More specifically, a keyword- candidate
list and domain thesaurus is provided to help obtain users’
preferences. The active user gives his/her preferences by
selecting the keywords from the keyword candidatelist,and
the preferences of the previous users can be extracted from
their reviews for services according to the keyword-
candidate list and domain thesaurus. Our method aims at
presenting a personalized service recommendation list and
recommending the most appropriate service to the users.
Finally, the experimental results demonstrate that KBSR
significantly improves the accuracy andscalabilityofservice
recommender systems over existing approaches.
3.3 Proposed Methodology
3.3.1 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. In this project, an
active user refers to a current user needs recommendation.
Preferences of an active user: An active user can give
his/her preferences about candidate services by selecting
keywords from a keyword-candidate list, which reflect the
quality criteria of the services he/she is concerned about.
Preferences of previous 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 and domain thesaurus .And a review of the
previous user will be formalized into the preference
keyword set of him/her.
Fig -2: Main steps of KBSR System
The main steps of KBSR are depicted in Fig. 2, which are
described in detail as follows: , Our system provide
candidate keyword list so current users can select keyword

like food, transport, hospitality etc. from given preference
list as shown in Fig. 3.
Fig -3: Current user can select keyword from given list of
services
In Fig 4 , keyword selected by active user will goes to the
text area which will consider as active user preference list.
Fig -4 Select service from given list
In Fig 5, keyword selected by active user from preference
list can search for synonyms using word net dictionary.
Fig -5: Synonymns of selected keyword
In Fig 6 in first text area the extracted keyword with their
count is display which was taken by previous users review.
In second text area jaccard coefficient displayed with the
help of jaccard coefficient similarity computation algorithm
Fig -6: Review extraction and computing similarity
3.3.2 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 user’s
preferences will be filtered outbytheintersectionconceptin
set theory. We are using word net dictionary for finding
synonymous of keyword which are presenting active user
preference list 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 [14].

3.3.3 Calculatepersonalizedratingsandgenerate
recommendations.
Based on the similarity of the active user and
previous users, further filtering will be conducted. Once the
set of most similar users are found, the personalized ratings
of each candidate service for the active user can be
calculated. Finally, a personalized service recommendation
list will be presented to the user and the service(s) with the
highest rating(s) will be recommended to him/her [14]. .
4. CONCLUSIONS
We have proposed a keyword-based service
recommendation method, named KBSR. In KBSR, keywords
are used to indicate users' preferences, and a user based
Collaborative Filtering algorithm is adopted to generate
appropriaterecommendations.Morespecifically,a keyword-
candidate list and domain thesaurus are provided to help
obtain users' preferences. The active user gives his/her
preferences by selecting the keywords from the keyword-
candidate list, and the preferences of the previous users can
be extracted from their reviews for services accordingtothe
keyword-candidate list and domain thesaurus. Our method
aims at presenting a personalized service recommendation
list and recommending the most appropriate service to the
users.
REFERENCES
[1] J. Manyika, M. Chui, B. Brown, et al, “Big Data: The
next frontier for innovation, competition, and
productivity,” 2011.
[2] Lynch, “Big Data: How do your data grow?” Nature,
Vol. 455, No. 7209, pp. 28-29, 2008
[3] F.Chang, J.Dean,SGhemawat,andWCHsie,“Bigtable:A
distributed storage systemforstructureddata”ACM
Transactions on Computer Systems,Vol. 26, No.
2(4), 2008.
[4] W. Dou, X. Zhang, J. Li,J. Chen, “HireSome-II:Towards
Privacy-AwareCross-CloudServiceCompositionfor
Big Data Applications,”IEEE Transactions on
Parallel and Distributed Systems,2013.
[5] 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.
[6] M. Bjelica, “Towards TV Recommender System
Experiments with User Modeling,” IEEE
Transactions on Consumer Electronics, Vol. 56,
No.3, pp. 1763-1769, 2010.
[7] M. Alduan, F. Alvarez, J. Menendez, and O. Baez,
“Recommender System for Sport Videos Based on
User Audiovisual Consumption,” IEEE Transactions
on Multimedia, Vol. 14, No.6, pp. 1546-1557, 2013
[8] F. Chang, J. Dean, S. Ghemawat, and W.C. Hsieh, “Big
table: ADistributed Storage System for Structured
Data,” ACM Trans. Computer Systems, vol. 26, no. 2,
article 4, 2008.
[9] Simon Philip, P.B. Shola (PhD),
AbariOvyeJohn“Application of Content-Based
Approach in Research Paper Recommendation
System for Digital Library”, 2014.
[10] J.Manyika et al., “Big Data: The Next Frontier for
Innovation, Competition, and Productivity,” 2011
[11] Z. Luo, Y. Li, and J. Yin, “Location: A Feature for
Service Selection in the Era of Big Data,” Proc. IEEE
20th Int’l Conf. Web Service,pp. 515-522, 2013.
[12] H. Schutze and J. O. Pedersen, “A Cooccurrence-
Based Thesaurus and Two Applications to
Information Retrieval,” Information Processing &
Management, vol. 33, no. 3, pp. 307-318, 1997.
[13] Y. Jing and W. Croft, “An Association Thesaurus for
Information Retrieval,” Proc. IntelligentMultimedia
Retrieval Systems and Management Conf. (RIAO),
vol. 94, pp. 146-160, 1994.
[14] KASR:A Keyword-Aware ServiceRecommendation
Method on MapReduce for BigData Applications
Shunmei Meng, Wanchun Dou, Xuyun Zhang, and
Jinjun Chen, Senior Member, IEEE

Keyword Based Service Recommendation system for Hotel System using Collaborative Filtering

More Related Content

What's hot (19)

Similar to Keyword Based Service Recommendation system for Hotel System using Collaborative Filtering (20)

More from IRJET Journal (20)

Recently uploaded (20)

Keyword Based Service Recommendation system for Hotel System using Collaborative Filtering