International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________

Cluster Optimization for Improved Web Usage Mining

Roshni Ali
Department of Computer Science & Engineering
RCERT
Chandrapur, India
roshniali27@gmail.com
Abstract Now days, World Wide Web (WWW) has become rich and most powerful source of information. Conversely, it has become tricky
and critical task to retrieve actual information due to its continuous expansion in dimensions. Web Usage Mining is a step-wise technique of
extracting useful access patterns of the user from web. Web personalization makes use of web usage mining techniques, for knowledge
acquisition process done by analyzing the user navigational patterns. The web page personalization involves clustering of different web pages
having similar navigation patterns for an individual. Since cluster size expands due to the frequent access, optimization or shrinking the size of
clusters becomes a chief consideration. This paper proposes a tactic of cluster optimization based on concept of swarm intelligence techniques.
Later on based on the recognition of user access patterns, clustering is implemented using neural fuzzy approach i.e. NEF Class algorithm and
cluster optimization is implemented using Ant Nest Mate Approach.
Keywords-Web Usage Mining, Web personalization,NEF Class, Cluster optimization, Swarm intelligence, Pheromone Values, Scouts, Searchers

__________________________________________________*****_________________________________________________
I.

INTRODUCTION

With about 30 million new web pages posted daily, the

WWW is the prime, most used, knowledge source and the most
perspective marketplace. In order to profitably hold on to users
in this hastily rising surroundings, a web site must be build in
such a way that it supports user personalization. To achieve
this, an organization focuses on keeping track of user activities
while browsing their web sites. Although there are many tools
that helps to analyze this data using some of the web statistics
methods, but the data provided is a sufficient information only
for the web site and not for the designer. One of the ways to
overcome this shortcoming is by applying data mining
techniques on the Web. Data mining is a powerful new
technology with great potential
to help companies focus on
the most important information in the data they have collected
about the behavior of their potential clients. It identifies valued
information within the data that queries and reports can't
successfully divulge. Data mining, or knowledge discovery, is
the computer-assisted process of digging through and analyzing
enormous sets of data and then extracting the meaning of the
data[1]. Data mining tools foresee behaviors and upcoming
trends, allowing businesses to build proactive, knowledgedriven decisions.
Web mining - is the application of data mining techniques
to discover patterns from the World Wide Web. Web mining is
the integration of information gathered by traditional data
mining methodologies and techniques with information
gathered over the World Wide Web. Web mining is the use of
data mining techniques to automatically discover and extract
information from Web documents and services. Web mining
can be divided into three different types Web usage
mining, Web content mining and Web structure mining [2].
The rapid e-commerce growth has made both business
community and customers face a critical condition. Powerful
& influential competition and the customers option to choose
from several different alternatives, business community has
realized the necessity of intelligent marketing strategies and
relationship management. Web servers record and accumulate
data about user interactions whenever requests for resources

are obtained. Analysis of Web access logs can helps to

understand the user behavior and the web structure. From the
point of view of business and applications, knowledge
acquired by the Web usage patterns can be directly applied to
efficiently manage activities related to e-business, e-services,
e-education and so on . Accuracy Web usage analysis helps to
attract new customers, preserve existing customers, improvise
cross marketing/sales, efficiency in promotional campaigns,
tracking parting customers and find the most effective logical
structure for their Web space. User profiles could be built by
combining users navigation paths with other data features,
such as page viewing time, hyperlink structure, and page
content [4]. As different brands morph into content providers,
clients have become more and more habituated of seeing
content that's customized, hence laying a basis for the need
importance of personalization.
In the coming sections, in the paper we will come across
detailed process of Web Usage Mining in Section II,Section
III will illustrate basics of Clustering in web & Neuro-Fuzzy
approach for the same. Section IV will focus on area of Swarm
Intelligence & Ant Nest Mate method that has been used for
cluster optimization. Section V will give detailed experimental
results of the work done for this paper. Performance Analysis
for same is illustrated in next Section VI.
II.

WEB USAGE MINING

Web usage mining refers to the automatic discovery and

analysis of patterns in click streams, user transactions and
other associated data collected or generated as a result of user
interactions with web resources on one or more Web sites [7].
With the continual growth and propagation of e-commerce,
Web services, and Web-based information systems, the
volumes of click stream, transaction data, and user profile data
collected by Web-based organizations in their daily operations
has reached astronomical proportions [8]. Proper Analysis of
such web data helps the organizations to determine the lifetime value of clients, also helps in designing cross-marketing
strategies for products and services, evaluating effectiveness
6394

IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________
of promotional campaigns, optimizing the functionality of
III. CLUSTERING
Web-based applications, provides more tailored content to
Clustering is a method of data mining that collectively
viewers, and helps in finding the most effective logical
groups set of items having similarities in there characteristics
structure for Web space. This kind of analysis helps in
revealed. In usage domain, we have observed that there are
automatic discovery of significant patterns and associations
broadly two basic clusters i.e. user clusters and page clusters
among huge collection of chiefly semi-structured data stored
[12]. Clustering on user records (sessions or transactions) is
in Web servers and applications server access logs and related
referred as most common analysis done in Web usage mining
operational data sources. The goal is to confine, model, and
and Web analytics. Clustering of users together tends to create
examine the behavioral patterns and profiles of users
groups exhibiting similar kind of browsing or access patterns.
interacting with a Web site. The discovered patterns are
This knowledge is especially helpful for inferring user
usually represented as collections of pages, objects, or
demographics in order to perform market segmentation in eresources that are frequently accessed or used by groups of
commerce applications or provide personalized Web content
users with common needs or interests [10].
to the users with similar interests [8]. Further on, analysis of
Web Usage mining Consists of three phases, mainly preuser groups based on their demographic attributes can lead to
processing, pattern discovery, and pattern analysis [11]. Fig 1.
the discovery of valuable business intelligence. Moreover,
Below shows the sequence of Web Usage Mining process.
Usage-based clustering has also been used to create Web-

Fig 1: Web Usage Mining Process

1.Pre-Processing.
It is the process of converting the unstructured data into
useful information by applying some algorithm. Web usage
data sources must be integrated, filtered, cleaned, and
transformed, such that gaps will be possibly filled, irrelevant
information will be thrown away, and user sessions and
transactions will be identified. These sources of data are
mainly Web server log files, agent logs and other interfaces.
The data present in the log file cannot be used as it is.
Therefore, contents of the web log file should be cleaned in
this preprocessing step. The unwanted data are bumped of and
a minimized log file is obtained.
2.Pattern Discovery
After converting the data in log file into a formatted data
the pattern discovery process is done. With the existing data of
the log files many useful patterns are identified either with user
ids, session details, time outs etc. It is the key component for
analyzing the pre-processed data. In this phase the process is
done through various algorithm and knowledge discovery
techniques. It can be done using various techniques such as
association rules, classification, clustering, sequential pattern
and statistical analysis.
3. Pattern analysis
This process eliminates the irrelevant rules or patterns that
were generated. They extract the interesting rules or patterns
from the output of the pattern discovery. The most familiar
form of pattern analysis comprises of a knowledge query
mechanism such as SQL (Structured Query Language) or
loads the usage data into a data cube toper form OLAP (Online
analytical processing) operations.

based user communities reflecting similar interests of users,

and to study user patterns that can be used to provide dynamic
recommendations in Web personalization applications.
One straightforward approach in creating an aggregate view
of each cluster is to compute the centroid (or the mean vector)
of each cluster. The dimension value for each page view in the
mean vector is computed by finding the ratio of the sum of the
page view weights across transactions to the total number of
transactions in the cluster. If page view weights in the original
transactions are binary, then the dimension value of a page
view p in a cluster centroid represents the percentage of
transactions in the cluster in which p occurs. Thus, the
centroid dimension value of p provides a measure of its
significance in the cluster. Page views in the centroid can be
sorted according to these weights and lower weight page views
can be filtered out [8]. The resulting set of page view-weight
pairs can be viewed as an aggregate usage profile
representing the interests or behavior of a significant group of
users.
Since a fuzzy systems describes the knowledge it encodes
better but it cant learn or adapt its knowledge from training
examples. On contrary neural network learns from training
examples but cannot elucidate what it has learnt hence its not
viable to infer the end result in natural language. Neural
networks and fuzzy systems both have their own strengths and
weaknesses. The unification of neural networks and fuzzy
logic in neuro-fuzzy models provides learning & readability.
Hence, researchers have made many attempts to assimilate
these specific methods to craft hybrid models which will unite
the advantages/merits of both methods. In the conventional
approach to fuzzy clustering the model designer based on a
prior knowledge fixes the membership functions and the
consequent models. However, in cases where this set is
unavailable and instead a set of input-output data is observed
from the process, the components of fuzzy system i.e.
membership and consequent models can be represented in a
parametric form and the parameters are tuned with the help of
neural networks. In such situations the fuzzy methods turn into
neurofuzzy methods. Neuro-fuzzy methods combine the
uncertainty handling capability of fuzzy systems and the
learning ability of neural networks. Thus, neurofuzzy (NF)
computing has become a popular framework for solving
complex problems in general and clustering problems in
particular. In case the knowledge about clustering or any
6395

IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________
general problem can be expressed in linguistic rules, then a
As the size of the cluster goes on increasing due to increase
fuzzy inference system (FIS) can be built, and if it is in data,
in users or growth of interest of users it has become inevitable
or can be learned from a simulation or training then artificial
need to optimize the clusters. Here we introduces a cluster
neural networks (ANNs) can be applied[16][17].
optimizing methodology based on ants nest mate recognition
ability and is used for eliminating the data redundancies that
IV. OPTIMIZATION THROUGH SWARM INTELLIGENCE
may occur after the clustering done by the web usage mining
methods. Ant Nest Mate approach for cluster optimization is
Particle Swarm Optimization (PSO) was originally
presented to personalize web page clusters of target users.
designed and introduced by Eberhart and Kennedy. The PSO
Hierarchy relationship exists within groups. These complex
algorithm is a population base seek algorithm based on social
behaviors can be instantiated with a fact that ants can
behavior of birds, bees or a school of fishes [20]. Originally
distinguish between nest mates and non-nest mates. The level
swarm intelligence focuses on graphically simulating the
of interaction and cooperation among ants of different colony
graceful and unpredictable choreography of bird folk. Every
is nearly nil as to protect the exploitation of the colony from
single individual is represented as vector in multidimensional
outsiders. Ants can distinguish nest mates from non-nest
search area. Thus same vector have one assigned vector that
mates, which allow them to limit altruism and cooperation to
can determine the subsequent progress of the particle called as
members of their own colony and protect their colony from
velocity vector. The PSO then determines methods to revise
exploitation by outsiders.
the velocity of a particle. Each particle then updates its
velocity based on present velocity and the finest arrangement
V. EXPERIMINTAL RESULTS
explored so far [20].The PSO practice is then iterated for some
fixed number of times till minimum error based on preferred
1) Creating Web log File.
The Windows Firewall log allows advanced users to
performance index is attained. It has been shown that this
collect and identify inbound traffic. You can log dropped
simple model can deal with difficult optimization problems
packets and successful connections. Once logging is turned
efficiently. The PSO, in the beginning, was developed for real
on all of the information is written to a file called,
valued spaces but many troubles are, however, defined for
pfirewall.log. The log file is stored in the %system
discrete valued spaces where the domain of the variables is
root%\Windows directory. This log file contains fields like
finite.
date, time, action, protocol src-ip ,dst-ip ,src-port ,dst-port
Recently a family of nature have inspired lots of
,size ,tcpflags, tcpsyn ,tcpack ,tcpwin, icmptype ,icmpcode
technical algorithms, known as Swarm Intelligence (SI).It has
& info path.This log file is filtered & pre-processed
fascinated number of researchers from the areas of pattern
initially.
recognition and clustering [21]. Various clustering techniques
that are based on this have allegedly presented many classical
Step 1: Take pfirewall.log file as an input.
methods of partitioning a complex real world dataset. This
Step 2: Parsing the pfirewall.log.
area of Swarm Intelligence is a relatively new interdisciplinary
(Selecting the only required attributes.i.e.Src ip,dest
field of research that has gained huge popularity now a day.
ip & size(no. of . Packets)
Different algorithms resembling to the domain portray
Step3: Store in array.
inspiration from the collective intelligence emerging from the
Step 4: Check which entry have same destination &
behavior of a group of social insects (like bees, termites and
source ip and apply the accumulation filter /
wasps). When acting as a community together, these insects
Discretion filter.
with very limited individual capability cooperatively perform
Step 5: Display output in console.
many complex tasks necessary for their continued existence.
Troubles of finding and storing foods, selecting and picking up
materials for future usage need a thorough planning, and are
solved by insect colonies without any kind of supervisor or
controller. Particle Swarm Optimization (PSO) is another very
popular SI algorithm for global optimization over continuous
search spaces.
The complex social behavior of ants and other social
insects requires multiple levels of recognition. Thus, Ant Nest
mate approach suggests that ants can distinguish nest mates
from non-nest mates, which allow them to limit altruism and
cooperation to members of their own colony and protect their
colony from exploitation by outsiders. Ants that have the same
odor will be in the same nest. The clusters obtained are feed
into an ant based clustering approach that checks for the
similarity of the pheromone values of the artificial ants. This is
done on the fact that ants belonging to the same nest will have
similar odor. In this algorithm clusters are considered as the
ants nest and the url combinations in each cluster is considered
as the artificial ants.

Fig 2: Filtered data from Web log File

2) Implementing NEF CLASS.
Out of this filtered data,Src & Dest IP are separated & its
total access of time, total accessed data packets & total hits on
link is calculated & displayed in console. As NEF CLASS
suggests, using Src & Dest IP & Size of data, total accessed
data packets & total hits on link is calculated. We had group of
inputs that had 3 data values. After implementation of NEF
Class we had obtained more 2 attributes i.e. Total accessed
6396

IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________
time & Total accessed counts. Now using these inputs values,
Clusters as a whole that are referred as Centroid are
decision will be taken to add data to clustering.
divided in Scouts (Cluster 1) showing similar access patterns
of user & Searchers (Cluster 2) showing dissimilar ones.

Fig 3: Implementation of NEF-CLASS

Fig 5:Cluster as Scouts.

Once we have calculated mean access time and mean

access count, entries are then added for clustering by applying
condition.
Condition Applied.
If Access Time & Access Count Mean
Access Time & Access Count..
Then, ADD ENTRY TO CLUSTER.
Else..Reject Or Discard.

Fig 6.Cluster as Searchers

Once the Clusters are optimized and divided into Scouts &
Searchers, then on basis of this optimized data, User IPs are
tracked ,access to which other user has been done is identified
& mean access, mean time & mean size are calculated &
displayed.

Fig 4:Centroid Cluster.

Once the cluster have been created, then entries of these
clustered are parsed again to identify the similarities and
difference of data accessed. This parsing is done by applying
Swarm intelligence optimization technique on clusters & Ant
Nest mate approach is used for same.
Centroid = Cluster of data from web log file obtained
by pattern matching & Analysis
Pheromone Values = Navigation Links that user
follows by continuous browsing.
Scouts = Similarities of Links that has all relevant
matching contents of users need.
Searchers = Dissimilarities in Links has all irrelevant
items that dont matches up to data that was searched.
Fig 7: Tracked User Profiles
6397
IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________
VI. PERFORMANCE ANALYSIS
Once the user profiles have been tracked, Pie-charts are
generated for the same analyzing the behavior of each user on
the various ip address.Fig 8,below represent the user access
times showing that which ip destination is accessed for how
much time span.In the result below green color represent the
maximum time of access & ip address related for same was
179.60.192.7.and remaining other ips were not accessed for
much longer duration as compared to mentioned one.
Fig 10:Pie Chart for User Access Sizes
CONCLUSION

Fig 8:Pie chart for User Access Times

User access counts Pie chart represented below in Fig
9,mentiones the number of times particular ip address is
visited on an on by the user. As from the below Pie chart we
can say ip address 31.13.95.8 have got maximum number of
hit counts by the user,in the time span observed. Next to it was
10.47.176.28 represented by bit broad area in green color.
Other ip addresses like 10.163.122.233, 30.45.61.160,
23.35.39.14 & others as can been seen have fair number of
access counts.

Internet is the broad media for the users today, where

ample amount of data is available to them on a single click,
this data is updated day by day and growing larger in size.
This paper deals with the phenomena of providing user a
better access opportunity with less complexity in minimum
amount of time. This paper proposes optimization technique of
clusters based on swarm intelligence technique of Ant Nest
Mate. As user access data on web, enteries of its detailed
activity gets logged into web log file. Web log file tracks the
record of each and every event of customer. The web log is
accessed and performs data cleaning to remove crawlers
request and request to graphics. Accumulation filtering is done
for log file filtering. This filtered data is then used for
clustering. Clustering is done using NEF Class algorithm
which is a Neuro-Fuzzy approach. Thus filtered input pattern
is used for clustering & clusters thus formed are feed to Ant
Nest Mate Algorithm, where cluster optimization is performed
on ips accessed. Optimization is done on the basis of swarm
intelligence properties, where user following the similar
navigational links will be grouped together in one cluster
referred as Scouts and the other one with dis-similar behavior
will be placed in Searchers. Thus the cluster or centroid gets
optimized in more better way. Then we perform tracking user
profilesby extracting user profiles from each cluster as a set of
relevant urls and the user profiles discovered in certain period
of time is compared with the user profiles discovered in later
period. Based on the user profiles the web page is
personalized. As a future enhancement scalability can be taken
into consideration. Concept of Big data or thin data may come
across as for the storage of data in clusters. Hence improved
algorithm for same could be implemented for better storage.
REFERENCES
[1]
[2]

Fig 9:Pie chart for User Access Counts

[3]

User access size pie chart in Fig 10.below mentions amount of

data accessed on particular ip address.Here ip adress 10.34.189
can be observed to be have largest access size,next to this
comes ip,s like 10.163.122..203,10.45.61.160,10.47.176.28
etc,and other that can be seen has average access size

[4]
[5]
[6]

http://www.laits.utexas.edu/~anorman/BUS.FOR/course.mat/Ale
x/
Lin, C.-W. and Hong, T.-P. (2013), A survey of fuzzy web
mining. WIREs Data Mining Knowl Discov, 3: 190199.
doi: 10.1002/widm.1091
Srivastava J, Desikan P and V Kumar, Web Mining-Concepts,
Applications & Research Direction in 2002 Conference.
Abraham, Ajith, He Guo, and Hongbo Liu. Swarm intelligence:
foundations, perspectives and applications. Springer Berlin
Heidelberg, 2006.
Srivastava J, Desika& n P and V Kumar , Web MiningAccomplishment Future Directions in 2004 Conference.
R. Kosala, and H. Blockeel, Web Mining Research: A Survey,
SIGKDD Explorations, Newsletter of the ACM Special Interest
6398

IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________

International Journal on Recent and Innovation Trends in Computing and Communication

Volume: 3 Issue: 11

ISSN: 2321-8169
6394 - 6399

________________________________________________________________________________________________________
Group on Knowledge Discovery and Data Mining, Vol. 2, No. [26] C. W. Cleverdon The Cranfield Tests on Index Languages
[7]

[8]
[9]
[10]
[11]
[12]
[13]

[14]

[15]

[16]
[17]
[18]
[19]

[20]
[21]

[22]

[23]

[24]

[25]

1 pp 1-15, 2000.
Puzis, Yury, et al. "Predictive web automation assistant for
people with vision impairments." Proceedings of the 22nd
international conference on World Wide Web. International
World Wide Web Conferences Steering Committee, 2013.
Mobasher, Bamshad. "Data mining for web personalization."
The adaptive web. Springer Berlin Heidelberg, 2007. 90-135.
Qingtian Han, Xiaoyan Gao, Wenguo Study on Web Mining
Algorithim based on usuage Mining, Computer Aided
Industrial design and Conceptual design, 2008 CAID/CD 2008.
Facca, Federico Michele, and Pier Luca Lanzi. "Mining
interesting knowledge from weblogs: a survey." Data &
Knowledge Engineering 53.3 (2005): 225-241.
Jaideep shrivastav, Robert Colley, Mukund Deshpande, PangNing Tan, Web Usage Mining: discovery and Application of
usage pattern from web data,ACM SIGKDD,jan2000.
www.springer.com/cda/content/.../cda.../9783642539640-c1.pdf
WANG Tong HE Pi-lian, Web Log Mining by an Improved
AprioriAll Algorithm, proceedings of world academy of
science, engineering and technology volume 4 February 2005
ISSN 1307-6884, 2005 WASET.ORG.
Mohd Helmy Abd Wahab, Mohd Norzali Haji Mohd, Hafizul
Fahri Hanafi, and Mohamad Farhan Mohamad Mohsin, Data
Pre-processing on Web Server Logs for Generalized Association
Rules Mining Algorithm, World Academy of Science,
Engineering and Technology ,2008.
Kobra Etminani, Mohammad-R. Akbarzadeh-T, and Noorali
Raeeji Yanehsari, Web Usage Mining: users' navigational
patterns extraction from web logs using Ant-based Clustering
Method, in Proc. IFSA-EUSFLAT ,2009.
Roohi, Farhat. "NEURO FUZZY APPROACH TO DATA
CLUSTERING: A FRAMEWORK FOR ANALYSIS."
European Scientific Journal 9.9 (2013).
Cordn, Oscar. Genetic fuzzy systems: evolutionary tuning and
learning of fuzzy knowledge bases. Vol. 19. World Scientific,
2001.
R. Cooley, B. Mobasher, and J. Srivastava,Web Mining:
Information and Pattern Discovery on the World Wide Web,
IEEE Computer Society,2009, pp. 558
R. Cooley, B. Mobasher, and J. Srivastava, Data Preparation
for Mining World Wide Web Browsing Patterns,
KNOWLEDGE AND INFORMATION SYSTEMS, vol.
1,1999.
He, Jie, and Hui Guo. "A modified particle swarm optimization
algorithm." TELKOMNIKA Indonesian Journal of Electrical
Engineering 11.10 (2013): 6209-6215.
Ali, Yasir Hassan, Roslan Abd Rahman, and Raja Ishak Raja
Hamzah. "Acoustic emission signal analysis and artificial
intelligence techniques in machine condition monitoring and
fault diagnosis: a review." Jurnal Teknologi 69.2 (2014).
D.Vasumathi, and A.Govardan,BC-WASPT : Web Acess
Sequential Pattern Tree Mining, IJCSNS International Journal
of Computer Science and Network Security., Vol.9,June-2009,
pp. 569571.
S.Vijayalakshmi V.Mohan, S.Suresh Raja,Mining Constraintbased Multidimensional Frequent Sequential Pattern in Web
Logs, European Journal of Scientific Research., Vol.36, pp
.480-490,2009.
Ming-Syan Chen, Jong Soo Park, Philip S. Yu, Efficient Data
Mining for Path Traversal Patterns, Ieee Transactions On
Knowledge And Data Engineering, Vol. 10, No. 2, March/April
1998.
F.M. Facca, P.L. Lanzi Mining interesting knowledge from
Weblogs: a survey, Data and Knowledge Engineering Vol. 53,
No. 3,June 2005, pp 225-241.

[27]
[28]
[29]
[30]
[31]

[32]
[33]
[34]

[35]
[36]
[37]
[38]

[39]
[40]
[41]
[42]
[43]

[44]

Devices. In Readings in Information Retrieval, Morgan

Kaufmann Publisher. Inc, San Francisco, California, pp.4760.1997.
Qingtian Han, XiaoyanGao, Wenguo Study on Web Mining
Algorithim based on usuage Mining, Computer Aided
Industrial design and Conceptual design, 2008 CAID/CD 2008.
Jaideepshrivastav, Robert Colley, MukundDeshpande, PangNing Tan, Web Usage Mining: discovery and Application of
usage pattern from web data,ACM SIGKDD,jan2000.
Bhaskaran, V.S.a.V.M., "Data Preparation Techniques for Web Usage
Mining in World Wide Web-An Approach,"International Journal of
Recent Trends in Engineering, Vol 2, No.4, 2009.
C, G., M. M, and D. K, "Preprocessing of Web Log Files in Web
Usage Mining,"The Icfai Journal of Information Technology, 35J2008-03-06-01(35J-2008-03-06-01): pp. 55-66, 2008.
Mohd Helmy Abd Wahab, M.N.H.M., Hafizul Fahri
Hanafi,Mohamad Farhan and Mohamad Mohsin."Data Preprocessing on Web Server Logs for Generalized Association Rules
Mining Algorithm,"World Academy of Science, Engineering and
Technology 48, 2008.
L.K. Joshila Grace, V.M.a.D.N., "ANALYSIS OF WEBLOGS AND
WEB USER IN WEB MINING," International Journal of Network
Security & Its Applications(IJNSA), Vol.3, No.1, 2011.
K, D.a.V.v.M., "Session Reconstruction in Web Usage Analysis," The
Icfai Journal of Information Technology, 2008.
R.Krishnamoorthi, K.R.S.a.,"Identifying User Behavior by
Analyzing Web Server Access Log File," IJCSNS International
Journal of Computer Science and Network Security, Vol.9, No. 4,
2009.
V.Chitraa, A.S.D., "A Survey on Preprocessing Methods for Web
Usage Data," (IJCSIS) International Journal of Computer Science and
Information Security, Vol. 7, No. 3, 2010.
Arshi Shamsi, R.N., Pankj Pratap Singh and Mahesh Kumar
Tiwar,"Web Usage Mining by Data Preprocessing," International
Journal of Computer Science And Technology, Vol.3, No. 1, 2012.
R, C.R.S.J.a.D., "Web Usage Mining: Discovery and Application of
Interesting Patterns from Web Data," Ph.D Thesis, University of
Minnesota, 2000.
M.-S. Chen, J. S. Park, and P. S. Yu. Data mining for path traversal
patterns in a web environment. In Proc. of the 16th International
Conference on Distributed Computing Systems,pages 385392, May
1996.
T. Nakayama, H. Kato, and Y. Yamane. Discovering the gap between
web site designers expectations and users behavior. In Proc. of the
Ninth Intl World Wide Web Conference,Amsterdam, May 2000.
J. Pei, J. Han, B. Mortazavi-asl, and H. Zhu. Mining access patterns
efficiently from web logs. In Proc. of the 4th Pacific-Asia Conf. on
Knowledge Discovery and Data Mining,pages 396407, April 2000.
M. Perkowitz and O. Etzioni. Adaptive web sites:Automatically
synthesizing web pages. In Proc. of the Fifteenth National Conf. on
Artificial Intelligence (AAAI), pages 727732, 1998.
M. Perkowitz and O. Etzioni. Towards adaptive sites: Conceptual
framework and case study. In Proc. of the Eighth Intl World Wide
Web Conf, Toronto, Canada, May 1999.
C. Shahabi, A. M. Zarkesh, J. Abidi, and V. Shah. Knowledge
discovery from users web-page naviagtion. In Proc. of the 7th IEEE
Intl. Workshop on Research Issues in Data Engineering (RIDE), pages
2029, 1997.
M. Spiliopoulou, L. C. Faulstich, and K. Wilkler. A data miner
analyzing the navigational behaviour of web users. In Proc. Of
the Workshop on Machine Learning in User Modelling of the
ACAI99, Greece, July 1999.

6399
IJRITCC | November 2015, Available @ http://www.ijritcc.org

_______________________________________________________________________________________