Professional Documents
Culture Documents
Textbook Computational Intelligence Theories Applications and Future Directions Volume Ii Icci 2017 Nishchal K Verma Ebook All Chapter PDF
Textbook Computational Intelligence Theories Applications and Future Directions Volume Ii Icci 2017 Nishchal K Verma Ebook All Chapter PDF
https://textbookfull.com/product/computational-intelligence-
theories-applications-and-future-directions-volume-i-nishchal-k-
verma/
https://textbookfull.com/product/advances-in-computer-
communication-and-computational-sciences-proceedings-of-
ic4s-2017-volume-1-sanjiv-k-bhatia/
https://textbookfull.com/product/computational-intelligence-and-
its-applications-abdelmalek-amine/
https://textbookfull.com/product/solidworks-2017-black-book-
gaurav-verma/
https://textbookfull.com/product/cognitive-work-analysis-
applications-extensions-and-future-directions-1st-edition-
neville-a-stanton/
https://textbookfull.com/product/computational-intelligence-
applications-in-business-intelligence-and-big-data-analytics-1st-
edition-vijayan-sugumaran/
https://textbookfull.com/product/structural-dynamics-
fundamentals-and-advanced-applications-volume-ii-volume-ii-1st-
edition-alvar-m-kabe/
https://textbookfull.com/product/solidworks-
simulation-2017-black-book-gaurav-verma/
Advances in Intelligent Systems and Computing 799
Computational
Intelligence: Theories,
Applications and
Future Directions—
Volume II
ICCI-2017
Advances in Intelligent Systems and Computing
Volume 799
Series editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
The series “Advances in Intelligent Systems and Computing” contains publications on theory,
applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all
disciplines such as engineering, natural sciences, computer and information science, ICT, economics,
business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the
areas of modern intelligent systems and computing such as: computational intelligence, soft computing
including neural networks, fuzzy systems, evolutionary computing and the fusion of these paradigms,
social intelligence, ambient intelligence, computational neuroscience, artificial life, virtual worlds and
society, cognitive science and systems, Perception and Vision, DNA and immune based systems,
self-organizing and adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics including
human-machine teaming, knowledge-based paradigms, learning paradigms, machine ethics, intelligent
data analysis, knowledge management, intelligent agents, intelligent decision making and support,
intelligent network security, trust management, interactive entertainment,Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are primarily proceedings
of important conferences, symposia and congresses. They cover significant recent developments in the
field, both of a foundational and applicable character. An important characteristic feature of the series is
the short publication time and world-wide distribution. This permits a rapid and broad dissemination of
research results.
Advisory Board
Chairman
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Members
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing, Universidad Central “Marta
Abreu” de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science and Electronic Engineering, University of Essex, Colchester,
UK
László T. Kóczy, Faculty of Engineering Sciences, Department of Information Technology, Széchenyi
István University, Győr, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information, University of Technology, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute of Technology, Tijuana,
Mexico
Nadia Nedjah,Department of Electronics Engineering, State University of Rio de Janeiro, Rio de Janeiro,
Brazil
Ngoc Thanh Nguyen, Wrocław University of Science and Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering, The Chinese University of Hong
Kong, Shatin, Hong Kong
Editors
Computational Intelligence:
Theories, Applications
and Future Directions—
Volume II
ICCI-2017
123
Editors
Nishchal K. Verma A. K. Ghosh
Department of Electrical Engineering Department of Aerospace Engineering
Indian Institute of Technology Kanpur Indian Institute of Technology Kanpur
Kanpur, Uttar Pradesh, India Kanpur, Uttar Pradesh, India
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface
v
Contents
vii
viii Contents
xiii
Part I
Intelligent Informatics
Fog Computing-Based Autonomic
Security Approach to Internet of Things
Applications
1 Introduction
IoT has changed the phase of our living style by providing ubiquitous and pervasive
environment using network of physical devices embedded with sensors and software
[1]. The involvement of connected devices in the intimate spaces generates immense
of privacy and security threats. The wireless communication between the connected
devices in IoT should meet the security requirement of Internet and sensor network
along with the security requirement of IoT to ensure safe and reliable operations [2].
The involvement of evolving revolutionary technologies such as cloud computing,
software-defined network, big data analytics in IoT application has increased the
vulnerabilities because most of these technologies are at developing stage with non-
standard architecture and protocols.
IoT applications need self-protect capability because most of the connected
devices are not installed in managed environment [3]. Some IoT applications need
the devices to be installed on unmanaged environment such as climatic forecast-
ing, environmental monitoring applications. Another major factor that influences the
need for self-protect characteristics is increasing number of connected devices [4].
As the number of connected devices increases in IoT application, the attack surface
exponentially increases and it is difficult to handle such massive attack. To protect
the IoT application, each connected device has to protect itself from cyberattack by
using its own capability. Majority of the IoT devices are simple sensors with limited
computational power and memory, where it is not feasible to implement self-protect
mechanism. IoT application is made of devices from simple sensors to high-end sys-
tem; hence, a common security mechanism cannot be applied to all devices. Most of
the IoT applications process the data from the connected devices at the cloud. At the
same time, implementing self-protect mechanism at cloud side of the IoT application
leads to centralized security approach which is subjected to single-point failure and
latency in detection. Therefore, the self-protect mechanism of IoT applications are
implemented using fog computing (edge computing) [5].
Fog computing provides processing and storage capabilities at the edge with
lower latency [6]. Fog nodes connect the end devices of IoT application to cloud
computing resources to provide fast, actionable decisions to be made based on vast
amount of data generated from the IoT end devices. The self-protect mechanism of
IoT application can be implemented at the distributed fog nodes to forecast, detect,
and respond to attack at faster rate. In this paper, autonomic approach is used to
protect the IoT applications using fog computing. The proposed system is suitable
for IoT applications which connects and process the data from end devices at the
cloud such as environmental monitoring, weather forecasting, smart home. The major
contribution of the paper is
1. Fog computing-based self-protect mechanism for IoT applications is proposed
2. The forecasting module uses Gaussian process regression to predict the attack
3. The detection module uses two kinds of techniques based on the requirement of
IoT application
• Cognition-based approach
• Correlation-based approach
4. The response module uses fuzzy logic to generate response which can handle the
attack efficiently.
Fog Computing-Based Autonomic Security Approach to Internet … 5
The rest of the paper organized as follows. Section 2 provides existing research
performed so far in this domain. In Sect. 3, the detailed architecture and functionalities
of proposed system is presented. Section 4 presents the experimental setup to evaluate
the proposed work. Finally, we concluded the paper by providing possible extensions
in proposed autonomic approach to protect IoT applications in Sect. 5.
2 Related Work
IoT applications are made up of smart devices which can sense their surround-
ings and act accordingly [7]. Hence, self-protection capability can be imparted into
these devices to proactively protect themselves without human intervention. Many
self-protection systems have been proposed [8], but these are not specific to IoT
applications which contain heterogeneous devices from simple sensors to high-end
servers. A framework for self-protect mechanism was designed for IoT ecosystem
[9]. It is based on centralized control mechanism which is not suitable for large-scale
IoT applications that needs a distributed approach with lightweight implementation.
The proposed approach uses Gaussian process regression for forecasting the
anomaly. Gaussian process is mainly used for prediction in many applications [10].
In [11], Gaussian process is used to detect the anomaly in smart grid but it does not
use fog computing technique with edge intelligence to implement the learning model.
Many user behavior analytics-based anomaly detections have been proposed [12].
The proposed system uses a combination of user and behavior analytics to detect
anomaly. Similarly, statistical techniques are used for cyberattack detection [13], but
do not use the fog computing technology. The fuzzy logic is used in the response
module of the proposed system. In existing IoT security, fuzzy logic is mainly used
for providing access control [14] to IoT devices and not for detecting or handling the
attack.
To the best of our knowledge, this is the first autonomic approach using fog
computing to provide self-protect mechanism to IoT application.
In this section, autonomic approach to protect IoT applications using fog computing
is proposed. The IoT application of the proposed method made of end devices, fog
nodes, and cloud server. Figure 1 shows the architecture of the proposed approach.
In the proposed approach, self-protect method is implemented in the fog node con-
sisting of attack forecasting module, detection module, and response module. The
forecasting module uses Gaussian process regression to forecast the cyberattacks in
IoT application. The detection module contains two different techniques to detect
the attacks (i) cognitive analytics-based detection method and (ii) correlation-based
detection method. The detection techniques are chosen based on the application
6 S. Prabavathy et al.
requirement and capability. Finally, the response module consists of fuzzy logic
to identify the most certain response to handle attack scenario. When an attack is
detected by detection module or predicted by the forecasting module, alerts are gen-
erated and sent to the response module to select correct response to handle the attack.
The self-protect mechanism is implemented at the fog node. The traffic from end
devices flows to the cloud server through the fog nodes. Similarly, the data from
cloud reaches the end devices through fog node. Since all traffic passes through the
fog node, it is efficient to perform the analysis and decision at the distributed fog
nodes for large-scale IoT applications. Therefor, self-protection mechanism can also
be efficiently performed by the fog nodes.
Forecasting module uses the Gaussian process regression at fog nodes to forecast
anomaly in IoT application. The features of data traffic that represents anomaly are
considered as random variables to formulate the Gaussian process. Any subset of
these random variables can have joint multivariate Gaussian distribution which is
y f(x) + ε (1)
The covariance function k(x, x ) models the dependence between the function
values at different input points x and x . The choice of covariance function is very
import for optimal prediction. A composite covariance function is used because of
the heterogeneous nature of the IoT application. The sum of covariance function
considering short-term and long-term trends, the fluctuations in observation length
and periodicity. The short-term and long-term trends are represented by two isotropic
squared exponential covariance functions. An isotropic quadratic covariance function
is used to represent the fluctuations in the observation lengths. The periodicity is
represented by smooth periodic covariance function.
The detection module detects the anomaly using two different techniques (i) cog-
nitive analytics-based detection and (ii) correlation-based detection. The detection
technique is based on the IoT application requirement. If the fog nodes of IoT applica-
tion are resource constrained devices then complex machine learning algorithms can-
not be used; therefore, correlation-based techniques are used. If the fog nodes are not
resource constrained devices, then cognitive analytics-based detection can be used.
8 S. Prabavathy et al.
Fig. 3 Correlation-based
detection method
it is considered as anomaly. To reduce the false alarm average of three sample’s coef-
ficient is used to detect the anomaly. Figure 3 shows the flow of correlation analysis.
Response module generates the highly certain response necessary to handle the attack
scenario in the IoT application under two different output from the detection module
(i) if cloud or edge devices found abnormal and (ii) if the cloud or edge devices or
silent. Based on the result of the detection module, the response is generated using
the fuzzy logic controller. When there is uncertainty in the value of input parameters,
then fuzzy logic controllers can be used [16]. The result of the detection module
gives the state of the cloud and edge devices and it is given as input to the fuzzy
inference engine. The output of fuzzy inference system is the incident response for
IoT application to handle the attack scenario. Linguistic variables are defined for all
input and values are assigned to these linguistic variables. The rule base of fuzzy logic
controller stores all the details about input, membership functions, and corresponding
output.
The fuzzy inference system in the response module uses Mamdani model consist
of two input and one output as shown in Fig. 4. Triangular membership function is
used to define the linguistic variables. The two inputs are resulted from the detection
module that gives the state of the cloud and edge devices, and the values of two inputs
are normal anomaly and silent. The fuzzification process converts the individual
input into fuzzy set membership degree to identify highly appropriate response. The
rules are written on the rule base based on the result of detection module and the
expected response to handle the attack scenario. The centroid method is used for
defuzzification to convert the fuzzy output into crisp output (Table 1).
10 S. Prabavathy et al.
To demonstrate the proposed system, a simple smart home IoT application is simu-
lated using Cooja [17] simulator as shown in Fig. 5. It uses Contiki [18] operating
system on a system with configuration Core i5 3470M processor and 8 GB RAM hav-
ing GCC 4.9 and Ubuntu 14.04 environment. The simulation uses 6LoWPAN [19]
network made of Tmote sky sensor nodes connected to 6BR which acts as fog node.
The 6BR is connected to Microsoft Azure cloud service with the computing resource
4x Dual-Core AMD Opteron 2218 @ 2.6 GHz, 8 core, 32 GB RAM, 6 * 146 GB
HDD. The 6BR is not a constrained node; therefore, a laptop with DUALCORE
processor, 2 GB RAM, 500 GB HDD configuration is used. DDoS attack is gener-
ated by generating flooding attack in the 6LoWPAN simulated network to study the
performance of the proposed method. As a preliminary work, only correlation-based
detection is evaluated in the experiment.
The forecasting module uses Gaussian process regression to predict the attack
at faster rate. The prediction time is shown in Fig. 6. The probabilistic prediction is
depicted as gray area showing the higher and lower values of energy consumption due
to attack. The blue line shows the training part, and the red line shows the testing part
Fog Computing-Based Autonomic Security Approach to Internet … 11
of the forecasting module. The fog node evaluates the prediction time and activates
the response module.
The average of Pearson correlation coefficient is computed for the simulated smart
home application traffic at the fog node. Fig. 7 shows the lack of correlation between
the normal and malicious traffic. The detection accuracy is measured in terms of true
positive rate and false positive rate as shown in Fig. 8.
Figure 9 shows the defuzzification of the response module where X-axis and Y-
axis denote the cloud and end devices malicious traffic and Z-axis represents response
generated based on the rule in the fuzzy inference engine at fog node.
12 S. Prabavathy et al.
Fig. 9 Defuzzification in
response module
To study the efficiency of fog node in terms of latency, the proposed module is
implemented in cloud alone without fog nodes. The same experiment is repeated
under cloud scenario and delay is measured in terms of average response time. The
fog node forecasts, detects, and responds to attacks at faster rate when compared to
cloud implementation as shown in Fig. 10.
Fog Computing-Based Autonomic Security Approach to Internet … 13
5 Conclusion
In this paper, an autonomic approach to protect IoT applications using fog computing
was proposed. This approach implements the self-protection mechanism at the fog
nodes in IoT application to forecast, detect, and respond to the attacks at faster rate.
The proposed method uses online linear regression to forecast the attack. Cognitive
analytics and correlation analysis are used to detect the attack, and fuzzy logic is
used to identify the most appropriate response to handle the attack scenario.
The future work to test the approach in real-time IoT applications is to validate
the results and to increase the performance by tuning the proposed approach.
References
1. Wortmann, F., Flüchter, K.: Internet of things. Bus. Inf. Syst. Eng. 57(3), 221–224 (2015)
2. Sun, X., Wang, C.: The research of security technology in the internet of things. In: Advances
in Computer Science, Intelligent System and Environment, pp. 113–119. Springer, Berlin,
Heidelberg (2011)
3. Zhao, K., Ge, L.: A survey on the internet of things security. In: 9th International Conference
on Computational Intelligence and Security (CIS), pp. 663–667 (2013)
4. Dobson, S., Sterritt, R., Nixon, P., Hinchey, M.: Fulfilling the vision of autonomic computing.
Computer 43(1) (2010)
5. Bonomi, F., Milito, R., Zhu, j., Addepalli, S.: Fog computing and its role in the internet of
things. In: Proceedings of the first edition of the MCC workshop on Mobile cloud computing,
pp. 13–16. ACM (2012)
6. Yi, S., Li, C., Li, Q.: A survey of fog computing: concepts, applications and issues. In: Pro-
ceedings of the 2015 Workshop on Mobile Big Data, pp. 37–42. ACM (2015)
7. Xia, F., Yang, L.T., Wang, L., Vinel, A.: Internet of things. Int. J. Commun. Syst. 25(9),
1101–1109 (2012)
8. Tesauro, G., Chess, D.M., Walsh, W.E., Das, R., Segal, A., Whalley, I., Kephart, J.O.,
White, S.R.: A multi-agent systems approach to autonomic computing. In: Proceedings of
the Third International Joint Conference on Autonomous Agents and Multiagent Systems, vol.
1, pp. 464–471. IEEE Computer Society (2004)
14 S. Prabavathy et al.
9. Chen, Q., Abdelwahed, S., Erradi, A.: A model-based validated autonomic approach to self-
protect computing systems. IEEE Internet Things J. 1(5), 446–460 (2014)
10. Alecu, T.I., Voloshynovskiy, S., Pun, T.: The Gaussian transform of distributions: definition,
computation and application. IEEE Trans. Signal Process. 54(8), 2976–2985 (2006)
11. Fadlullah, Z.M., Fouda, M.M., Kato, N., Shen, X., Nozaki, Y.: An early warning system against
malicious activities for smart grid communications. IEEE Netw. 25(5) (2011)
12. Stanton, J.M., Stam, K.R., Mastrangelo, P., Jolton, J.: Analysis of end user security behaviors.
Comput. Secur. 24(2), 124–133 (2005)
13. Hurley, N., Cheng, Z., Zhang, M.: Statistical attack detection. In: Proceedings of the Third
ACM Conference on Recommender Systems, pp. 149–156 (2009)
14. Mahalle, P.N., Thakre, P.A., Prasad, N.R., Prasad, R.: A fuzzy approach to trust based access
control in internet of things. In: 3rd International Conference on Wireless Communications,
Vehicular Technology, Information Theory and Aerospace & Electronic Systems (VITAE),
pp. 1–5 (2013)
15. Benesty, J., Chen, J., Huang, Y., Cohen, I.: Pearson correlation coefficient. In: Noise Reduction
in Speech Processing, pp. 1–4. Springer Berlin Heidelberg (2009)
16. Harris, J.: An Introduction to Fuzzy Logic Applications, vol. 23. Springer Science & Business
Media (2001)
17. Osterlind, F., Dunkels, A., Eriksson, J., Finne, N., Voigt, T.: Cross-level sensor network sim-
ulation with cooja. In: 31st IEEE Conference on Local Computer Networks, pp. 641–648
(2006)
18. Dunkels, A., Gronvall, B., Voigt, T.: Contiki-a lightweight and flexible operating system for
tiny networked sensors. In: 29th Annual IEEE International Conference on Local Computer
Networks, pp. 455–462 (2004)
19. Mulligan, G.: The 6LoWPAN architecture. In: Proceedings of the 4th Workshop on Embedded
Networked Sensors, pp. 78–82 (2007)
Self-adaptive Frequent Pattern
Growth-Based Dynamic Fuzzy
Particle Swarm Optimization
for Web Document Clustering
Abstract Web documents being dynamic and vague, any system to cluster these
documents needs to be self-adaptive to these dynamic situations. For this, the system
requires the capability of capturing dynamicity. Dynamicity takes into account any
updates happening in the search space. If any new potential solution arises, the sys-
tem needs to identify and reinitialize the particle lists to the newly updated potential
solutions. The traditional particle swarm optimization fails in accounts of conver-
gence speed and maintaining diversity. The experimental results for the proposed
algorithm show that frequent pattern growth-based dynamic fuzzy particle swarm
optimization algorithm performs better than existing conventional approaches in
matters of convergence speed and in maintaining diversity.
R. V. Pamba (B)
Department of Computer Science and Engineering, LBS Institute of Technology for Women,
Trivandrum, Kerala, India
e-mail: pambaraj@gmail.com
R. V. Pamba
School of Computer Sciences, Mahatma Gandhi University, Kottayam, Kottayam, India
E. Sherly
VRCLC, Indian Institute of Information Technology Management, Kerala, Technopark Campus,
Kazhakkoottam, Kerala, India
e-mail: sherly@iiitmk.ac.in
K. Mohan
Payszone LLC LTD, Dubai, UAE
e-mail: me@kiranmohan.com
1 Introduction
Recent studies and research in the domain of document clustering are spearhead-
ing in the direction of inducing computational intelligence to information retrieval
[1–4]. To achieve this goal, researchers are trying to evolve efficient methodologies
for imparting dynamic optimization in clustering. The mechanism behind clustering
strives to partition set of similar items to one group and dissimilar items to another
group, finding its applications in information retrieval, document organization,
summarization, and classification [5]. In the domain of web documents clustering
especially being unstructured and dynamic in nature, the search space is volatile to
any minute fluctuations in the parameters and in the optimal centroids computed.
The proposed paper discusses a novel method of dynamic web document clustering
that takes into account convergence speed as well as divergence issues which are the
common delimiting factors while using particle swarm optimization. The proposed
approach not only find optimal cluster centroids using frequent pattern growth (FP
Growth) but also help find best optimal solutions as it changes over time [6] using
dynamic fuzzy particle swarm optimization. In order to identify optimal solutions
which change over time, it is imperative to maintain huge solution diversity using
dynamic optimization. Too much diversity would lead to lack in exploitation of best
solutions while less diversity would lead to faster convergence. This situation is taken
care of by the proposed approach of dynamic frequent pattern growth-based fuzzy
particle swarm optimization in providing a balanced search so that the documents
can cluster to its relevant entities.
The rest of the paper is structured as follows with Sect. 2 gives an insight on the
proposed methodology. While Sect. 3 discusses the results, Sect. 4 deals with the
conclusions.
2 Proposed Methodology
The problem definition for dynamic clustering in web document is phrased as below:
d = (d1 , d2 , ....,
˙ dD ) (2)
Here, we consider f (d, α(t)) as the value of fitness function of the particles in
the search space solution identified by the frequent pattern growth-based dynamic
fuzzy particle swarm optimization. The parameter α at time t captures the change in
environment. And the number of dimensions for the search space represented by D.
The key objective is to minimize the fitness value of the particle. The fitness function
tries to minimize the distances between data objects and its selected cluster centroids,
whereas d in Eq. 2 is the particle solutions or vector representation of documents.
Self-adaptive Frequent Pattern Growth-Based Dynamic Fuzzy … 17
Authors in their previous work [7] have developed a reference model that generates
particles through frequent pattern growth and tries to cluster those particles to its
nearest cluster centroids uisng fuzzy particle swarm optimization. In this paper,
emphasis is given on aspects of inculcating a self-adaptive process for the particles
to detect any positive changes happenings in their surroundings as discussed in [8]
and to adapt to the better position using the proposed algorithm of frequent pattern
growth-based dynamic fuzzy particle swarm optimization. Frequent pattern growth
algorithm extracts the swarms and particles for fuzzy particle swarm optimization.
2.1 FP Growth
Frequent pattern growth (FP Growth) [9] work in two scans, first scan deals with
finding all frequent item sets and second scan creates nodes with updated frequencies
for each items being visited in every transaction.
With the frequent patterns received as output from the FP Growth algorithm [10,
11], the proposed approach retrieves all possible combination of terms matching
to the user search context effectively. For the selection of frequent patterns, the
threshold is set to mean-squared residual error (MSRE) as the criteria. Table 1 gives
a reference for the concept translation between the output of FP Growth and the
inputs for modified DFPSO as discussed in Sect. 2.6.
Table 1 Translating frequent mining concepts to dynamic fuzzy particle swarm clustering
FP Growth concepts Dynamic fuzzy PSO clustering
Items Population/Dimension
Item sets Semantically related terms
Frequent item sets Swarms for dynamic FPSO
Cluster centroids Average of each frequent item sets
18 R. V. Pamba et al.
q
μi j = 1, ∀i = 1, 2, . . . p (4)
j=1
q
p
Jm = μimj si j where si j = oi − z j (5)
j=1 i=1
p
i=1 μimj oi
z j = p (6)
i=1 μimj
1
μi j = 2 (7)
q si j m−1
k−1 sik
The objects in the search space to be fuzzy clustered are embedded in μi j . Each
matrix has p rows and q columns in which p is the number of data objects and q is
the number of clusters. The values at ith row and jth column in the matrix μi j give
the degrees of membership that the particular element has to its respective clusters.
In the proposed methodology as shown in Table 1, the parameters for dynamic fuzzy
particle swarm optimization are generated by frequent pattern growth algorithm. The
frequent item sets generated acts as the swarms or the particles. The cluster centroids
are calculated by finding the average of all frequent item sets under every item sets.
The dynamic fuzzy particle swarm optimization (FPSO) begins with parameters
generated as in Sect. 2.1. The various parameters derived are, population of particles,
initial number of tentative cluster centroids. These cluster cenroid positions indicate
the potential solutions for the documents. The position matrix, X , shows the mem-
bership degrees between the particles spread across the search space in columns and
its related cluster centroids in rows as in Eq. 8. The position matrix is equivalent to
the fuzzy matrix. Each column shows how each particle is related to the respective
cluster centroids. The position matrix is given below:
⎡ ⎤
μ11 · · · μ1c
⎢ ⎥
X = ⎣ ... . . . ... ⎦ (8)
μn1 · · · μnc
Another random document with
no related content on Scribd:
1.C. The Project Gutenberg Literary Archive Foundation (“the
Foundation” or PGLAF), owns a compilation copyright in the
collection of Project Gutenberg™ electronic works. Nearly all the
individual works in the collection are in the public domain in the
United States. If an individual work is unprotected by copyright
law in the United States and you are located in the United
States, we do not claim a right to prevent you from copying,
distributing, performing, displaying or creating derivative works
based on the work as long as all references to Project
Gutenberg are removed. Of course, we hope that you will
support the Project Gutenberg™ mission of promoting free
access to electronic works by freely sharing Project
Gutenberg™ works in compliance with the terms of this
agreement for keeping the Project Gutenberg™ name
associated with the work. You can easily comply with the terms
of this agreement by keeping this work in the same format with
its attached full Project Gutenberg™ License when you share it
without charge with others.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside
the United States, check the laws of your country in addition to
the terms of this agreement before downloading, copying,
displaying, performing, distributing or creating derivative works
based on this work or any other Project Gutenberg™ work. The
Foundation makes no representations concerning the copyright
status of any work in any country other than the United States.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if
you provide access to or distribute copies of a Project
Gutenberg™ work in a format other than “Plain Vanilla ASCII” or
other format used in the official version posted on the official
Project Gutenberg™ website (www.gutenberg.org), you must, at
no additional cost, fee or expense to the user, provide a copy, a
means of exporting a copy, or a means of obtaining a copy upon
request, of the work in its original “Plain Vanilla ASCII” or other
form. Any alternate format must include the full Project
Gutenberg™ License as specified in paragraph 1.E.1.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.