The implementation of terrorist web miner for the terrorist networks
1 like98 views
The document discusses the development of a 'Terrorist Web Miner', an automated tool designed to collect, classify, and analyze fragmented information about terrorist networks available on the internet. This system aims to replace manual data collection methods, facilitating the extraction of useful links related to terrorist organizations, which can help in understanding their communication patterns and relationships. The study highlights the importance of social network analysis and the challenges of mining data from diverse and covert sources in order to support counter-terrorism efforts.
![IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________
Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 268
THE IMPLEMENTATION OF TERRORIST WEB MINER FOR THE
TERRORIST NETWORKS
R. D. Gaharwar1
, D. B. Shah2
, G.K.S. Gaharwar3
1,2
G. H. Patel Department of computer Science and Technology, Sardar Patel University, Vallabh Vidyanagar, India
3
School of Business and Law, Navrachana University, Vadodara, India
Abstract
Terrorist web information is very valuable for Law enforcement agencies. This type of information is available on World Wide
Web but is in fragments. These fragments of information are to be collected and analyzed for the terrorist network mining. This
process when done manually may result into inefficient and error prone mining of data. Hence during this study the attempt is
made to automate this difficult task of collecting, classifying and analysis of the terrorist web data. Developed Terrorist Web
Miner will search terrorist information available on World Wide Web, it will fetch all the web pages related to terrorist web data
and will extract the useful links. These links are stored into databases for future reference. This system can replace the manual
system for collecting terrorist web data. The web pages related to terrorist attacks can be used in future to extract the links
between terrorist organizations and hence a communication pattern, linkage patterns, role and relationship analysis can be
derived for terrorist networks.
Keywords: Terrorist Networks, Terrorist Network Miner, SNA, TNM.
---------------------------------------------------------------------***---------------------------------------------------------------------
1. INTRODUCTION
In modern era, Social Network Analysis (SNA) has emerged
as a novel tool for Terrorist Network Mining. SNA can be
used effective and efficiently for Terrorist Network Mining
only if the investigators are well equipped with adequate
information about Terrorist Network. This process of
information collection about terrorist networks is a difficult
task as these networks work in a covert manner. For this we
have to rely on media reports, press reports etc to collect
such data from open source. For this task search engines can
be used.
2. WEB MINING
The World Wide Web is a gigantic collection of different
types of homogenous as well as heterogeneous documents.
The web pages are themselves a collection of different types
of tags like anchor tags, header tags etc and they may be
human generated or machine generated. These documents
can be in different languages i.e. Human or programming
languages. The heterogeneous web documents can be a
collection of text, html, images, sound clips, pdf files etc.
The unique thing about web documents is that they also
have meta information associated with them .i.e.
information about information contained in these web
documents. [1]
From the World Wide Web, investigator tries to focus only
on those web documents which are related to particular topic
and then carefully extract useful links from these web pages.
This task can be performed by a web miner. Moreover a
web miner detects the unusual patterns that exist amongst
the web pages which initially seems unrelated.
Web mining is one of the data mining technique to dig out
useful information from web which includes web
documents, hyperlinks between documents, web site visited
logs, etc. Internet has became an indispensable part of our
lives now a days so the techniques which are helpful in
extracting data present on the web is an interesting area of
research. These techniques help to extract knowledge from
Web document, in which at least one of structure or usage
(Web log) data is used in the mining process (with or
without other types of Web). According to analysis targets,
web mining can be divided into three different types, which
are Web usage mining, Web content mining and Web
structure mining. [2]
3. TERRORIST WEB MINING
Terrorist Web mining is a challenge for detecting patterns of
terrorists/ terrorist organizations from the vast collection of
heterogeneous web pages available on internet. In a way
terrorist web mining encompasses web mining as one has to
mine all the data on the web as well as mine the structure
and usage patterns. Information on the web includes
structured text as well as unstructured text data. Therefore
the tools developed for web mining apply for terrorist web
mining also. In addition, we need tools to extract the
structure of the terrorist network as well as the
communication patterns.
Information about terrorists/ terrorist organizations is rarely
available. This is because of that these terrorists/ terrorist
organizations work in a secret manner. The complete
terrorist network working depends on their secrecy. The
terrorist networks are sometimes also referred as dark/covert
networks because of their pattern of working in a secret
manner.](https://image.slidesharecdn.com/theimplementationofterroristwebminerfortheterroristnetworks-160919091852/85/The-implementation-of-terrorist-web-miner-for-the-terrorist-networks-1-320.jpg)
![IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________
Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 269
4. RELATED RESEARCH
Valdis E. Krebs in “Mapping Networks of Terrorist Cells”
described a connection between relationship of individual in
Terrorist Networks and Data Sources for example
Trust/Dependence relationship can only be confirmed
amongst terrorists if terrorists have family contacts, has been
in neighborhood, school, military or organization. This type
of information can be gathered from public and court
records and hence this information may be available in
individual’s native country only. Task/mission relationship
can only be drawn from phone calls, electronic mail, chat
messages, web site visits or board messages records. Travel
records are also useful in collecting such information. Valdis
built a matrix showing links between terrorists. He decided
to draw linkage map of three strengths. Valdis also
concluded that the binding strength amongst the terrorists
will depend largely on the amount of time spend either at
school, learning institutes, clubs, work place, or at training
program. [3]
Xaing Y. et al have developed a prototype for criminal
relationship visualize tool called COPLINKS. Authors have
successfully developed an efficient, user friendly
information analysis and sharing system. Authors have used
data and information from various sources including law
enforcement agencies data. Hence COPLINK database
contains 2 million criminal case reports from the criminal
departments. This visualizing tool creates hierarchical lists
and hyperbolic tree from this database. After creating
hierarchical lists and hyperbolic tree of criminal data authors
have done empirical study to compare effectiveness and
efficiency of both these views. During this empirical study
they found out that both views can be useful in criminal
relationship visualization. [4]
Javal. et. al. has created a high-level architecture for
criminal mining, they have used a scalable universal crawler
for this architecture. This crawler begins its procedure by
listing all the unvisited URLs. Thereafter the pages
correlated to these URLs are fetched and the URLs from the
fetched web pages are added to the list. Moreover to
upgrade the performance of this crawler Javal et. al. have
used multithreaded architecture. Hence the request to web
pages is loaded in different threads and all these threads will
run simultaneously and the system performance is enhanced.
[5]
Gaharwar R. et al [6]has mentioned that all the terrorist
organizations interact to generate covert networks and they
operate in a secret manner which is also their biggest
strength. This secrecy help these terrorist organizations to
work efficiently without being at the surface, Hence
collection of information about such covert networks can be
a challenge. However if the information from open source
can be taken as authentic, applying effective data mining
techniques can generate the useful information. Such
information can be farther used for counter terrorist
activities.
Ding Z. et al described the importance of semantic
information on social networks compared to previous work
in SNA where prime focus was on social network (SN)
topology. Authors have suggested two different semantic
community discovery models in social networks. These are
Bayesian models which combine probabilistic modeling
with community detection in social networks. The authors
mainly focus on documents as the much of the
communication in social networks (SN) is done through
exchanging documents. These documents may be emails,
blogs, instant massages, chat history, boards, posts etc. The
authors have assumed that the users having related
communication interests will be associated with their
communication of a semantic society in SN. The
communication is done through document called
communication document which is generally taken as email.
Hence emails are considered as carrier document of
communication because they have valuable information
regarding SN infrastructure and semantic shared knowledge.
Here the authors study the semantics of SN. [7]
The following figure is the architecture of Terrorist Web
Miner. The Terrorist Web Miner shown here will search on
internet for social networking data of terrorist attacks in
India. Once the web pages related to the terrorist attacks in
India are fetched, these web pages will be parsed for
extracting the links of the related web pages. These web
pages related to Terrorist Networks will be stored in
Terrorist Database. [8]
Internet with social
networking data
Terrorist Tracker
Web Miner
Parsing for Terrorist
Organization
.
.
Terrorist
Database
Web Pages
Figure 1: the architecture of Terrorist Web Miner
Terrorist Database contains all the web pages which farther
will be used to extract terrorist links among terrorist
organizations. These web pages will be parsed to extract the
terrorist links. The terrorist links can be used to create a
linkage map of any terrorist attack. Linkage map can also be
used to study the relationships to each organization/terrorist
in the network as well as its can be used to study the role of
to each organization/terrorist in the network.](https://image.slidesharecdn.com/theimplementationofterroristwebminerfortheterroristnetworks-160919091852/85/The-implementation-of-terrorist-web-miner-for-the-terrorist-networks-2-320.jpg)
![IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________
Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 271
CONCLUSION
The counter terrorism war can be fought and won only by
superior knowledge and effective SNA tools. Terrorist
Networks can be destabilized only when we have enough
knowledge about terrorist networks and its topologies.
Terrorist Web Mining aids in searching, extracting and
collecting this knowledge. During this study we successfully
created a terrorist web miner which crawls through different
search engines and returns the related web pages. These web
pages are then parsed for extracting the terrorist links from
the web pages which may contain information about
terrorist organizations. This links can be farther stored to
create linkage map of terrorist networks.
REFERENCES
[1]. Sergey Brin and Lawrence Page, "The Anatomy of a
Large-Scale Hypertextual Web Search Engine," in
7th World Wide Conference(WWW7), Brisbane,
Australia, April 1998
[2]. Monika Yadav and Pradeep Mittal, "Web Mining:
An Introduction," International Journal of Advanced
Research in Computer Science and Software
Engineering, vol. 3, no. 3, pp. 683-687, March 2013.
[3]. V. E. Krebs, "Mapping networks of terrorist cells,"
Connections 24, pp. 43–52, March 2001
[4]. Yang Xiang, Michael Chau, Homa Atabakhsh, and
Hsinchun Chen, "Visualizing criminal relationships:
comparison of a hyperbolic tree and a hierarchical
list," Decision Support Systems 41, pp. 69–83, 2005.
[5]. Javad Hosseinkhani, Suriayati Chaprut, Hamed
Taherdoost, Morteza Harati Cool, and Sadegh
Emami Korani, "Criminal Communities Mining on
the Web," International Journal of Advanced
Computer Science and Information Technology
(IJACSIT), vol. 2, no. 2, pp. 13-22, April 2013.
[6]. R. D. Gaharwar, D. B. Shah, and G.K.S. Gaharwar,
"Terrorist Network Mining: Issues and Challenges,"
International Journal of Advance Research in Science
and Engineering, vol. 4, no. 1, pp. 33-37, 2015.
[7]. D Zhou, R Manavoglu, J Li, C L Giles, and H Zha,
"Probabilistic models for discovering e-
communities," in 15th international conference on
world wide web (WWW), 2006, pp. 173–182.
[8]. R. D. Gaharwar, D. B. Shah, and G.K.S. Gaharwar,
"Proposed Architecture for Terrorist Web Miner,"
International Journal of Computer Applications, vol.
128, no. 9, pp. 18-20, October 2015.](https://image.slidesharecdn.com/theimplementationofterroristwebminerfortheterroristnetworks-160919091852/85/The-implementation-of-terrorist-web-miner-for-the-terrorist-networks-4-320.jpg)
The implementation of terrorist web miner for the terrorist networks
- 1. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 268 THE IMPLEMENTATION OF TERRORIST WEB MINER FOR THE TERRORIST NETWORKS R. D. Gaharwar1 , D. B. Shah2 , G.K.S. Gaharwar3 1,2 G. H. Patel Department of computer Science and Technology, Sardar Patel University, Vallabh Vidyanagar, India 3 School of Business and Law, Navrachana University, Vadodara, India Abstract Terrorist web information is very valuable for Law enforcement agencies. This type of information is available on World Wide Web but is in fragments. These fragments of information are to be collected and analyzed for the terrorist network mining. This process when done manually may result into inefficient and error prone mining of data. Hence during this study the attempt is made to automate this difficult task of collecting, classifying and analysis of the terrorist web data. Developed Terrorist Web Miner will search terrorist information available on World Wide Web, it will fetch all the web pages related to terrorist web data and will extract the useful links. These links are stored into databases for future reference. This system can replace the manual system for collecting terrorist web data. The web pages related to terrorist attacks can be used in future to extract the links between terrorist organizations and hence a communication pattern, linkage patterns, role and relationship analysis can be derived for terrorist networks. Keywords: Terrorist Networks, Terrorist Network Miner, SNA, TNM. ---------------------------------------------------------------------***--------------------------------------------------------------------- 1. INTRODUCTION In modern era, Social Network Analysis (SNA) has emerged as a novel tool for Terrorist Network Mining. SNA can be used effective and efficiently for Terrorist Network Mining only if the investigators are well equipped with adequate information about Terrorist Network. This process of information collection about terrorist networks is a difficult task as these networks work in a covert manner. For this we have to rely on media reports, press reports etc to collect such data from open source. For this task search engines can be used. 2. WEB MINING The World Wide Web is a gigantic collection of different types of homogenous as well as heterogeneous documents. The web pages are themselves a collection of different types of tags like anchor tags, header tags etc and they may be human generated or machine generated. These documents can be in different languages i.e. Human or programming languages. The heterogeneous web documents can be a collection of text, html, images, sound clips, pdf files etc. The unique thing about web documents is that they also have meta information associated with them .i.e. information about information contained in these web documents. [1] From the World Wide Web, investigator tries to focus only on those web documents which are related to particular topic and then carefully extract useful links from these web pages. This task can be performed by a web miner. Moreover a web miner detects the unusual patterns that exist amongst the web pages which initially seems unrelated. Web mining is one of the data mining technique to dig out useful information from web which includes web documents, hyperlinks between documents, web site visited logs, etc. Internet has became an indispensable part of our lives now a days so the techniques which are helpful in extracting data present on the web is an interesting area of research. These techniques help to extract knowledge from Web document, in which at least one of structure or usage (Web log) data is used in the mining process (with or without other types of Web). According to analysis targets, web mining can be divided into three different types, which are Web usage mining, Web content mining and Web structure mining. [2] 3. TERRORIST WEB MINING Terrorist Web mining is a challenge for detecting patterns of terrorists/ terrorist organizations from the vast collection of heterogeneous web pages available on internet. In a way terrorist web mining encompasses web mining as one has to mine all the data on the web as well as mine the structure and usage patterns. Information on the web includes structured text as well as unstructured text data. Therefore the tools developed for web mining apply for terrorist web mining also. In addition, we need tools to extract the structure of the terrorist network as well as the communication patterns. Information about terrorists/ terrorist organizations is rarely available. This is because of that these terrorists/ terrorist organizations work in a secret manner. The complete terrorist network working depends on their secrecy. The terrorist networks are sometimes also referred as dark/covert networks because of their pattern of working in a secret manner.
- 2. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 269 4. RELATED RESEARCH Valdis E. Krebs in “Mapping Networks of Terrorist Cells” described a connection between relationship of individual in Terrorist Networks and Data Sources for example Trust/Dependence relationship can only be confirmed amongst terrorists if terrorists have family contacts, has been in neighborhood, school, military or organization. This type of information can be gathered from public and court records and hence this information may be available in individual’s native country only. Task/mission relationship can only be drawn from phone calls, electronic mail, chat messages, web site visits or board messages records. Travel records are also useful in collecting such information. Valdis built a matrix showing links between terrorists. He decided to draw linkage map of three strengths. Valdis also concluded that the binding strength amongst the terrorists will depend largely on the amount of time spend either at school, learning institutes, clubs, work place, or at training program. [3] Xaing Y. et al have developed a prototype for criminal relationship visualize tool called COPLINKS. Authors have successfully developed an efficient, user friendly information analysis and sharing system. Authors have used data and information from various sources including law enforcement agencies data. Hence COPLINK database contains 2 million criminal case reports from the criminal departments. This visualizing tool creates hierarchical lists and hyperbolic tree from this database. After creating hierarchical lists and hyperbolic tree of criminal data authors have done empirical study to compare effectiveness and efficiency of both these views. During this empirical study they found out that both views can be useful in criminal relationship visualization. [4] Javal. et. al. has created a high-level architecture for criminal mining, they have used a scalable universal crawler for this architecture. This crawler begins its procedure by listing all the unvisited URLs. Thereafter the pages correlated to these URLs are fetched and the URLs from the fetched web pages are added to the list. Moreover to upgrade the performance of this crawler Javal et. al. have used multithreaded architecture. Hence the request to web pages is loaded in different threads and all these threads will run simultaneously and the system performance is enhanced. [5] Gaharwar R. et al [6]has mentioned that all the terrorist organizations interact to generate covert networks and they operate in a secret manner which is also their biggest strength. This secrecy help these terrorist organizations to work efficiently without being at the surface, Hence collection of information about such covert networks can be a challenge. However if the information from open source can be taken as authentic, applying effective data mining techniques can generate the useful information. Such information can be farther used for counter terrorist activities. Ding Z. et al described the importance of semantic information on social networks compared to previous work in SNA where prime focus was on social network (SN) topology. Authors have suggested two different semantic community discovery models in social networks. These are Bayesian models which combine probabilistic modeling with community detection in social networks. The authors mainly focus on documents as the much of the communication in social networks (SN) is done through exchanging documents. These documents may be emails, blogs, instant massages, chat history, boards, posts etc. The authors have assumed that the users having related communication interests will be associated with their communication of a semantic society in SN. The communication is done through document called communication document which is generally taken as email. Hence emails are considered as carrier document of communication because they have valuable information regarding SN infrastructure and semantic shared knowledge. Here the authors study the semantics of SN. [7] The following figure is the architecture of Terrorist Web Miner. The Terrorist Web Miner shown here will search on internet for social networking data of terrorist attacks in India. Once the web pages related to the terrorist attacks in India are fetched, these web pages will be parsed for extracting the links of the related web pages. These web pages related to Terrorist Networks will be stored in Terrorist Database. [8] Internet with social networking data Terrorist Tracker Web Miner Parsing for Terrorist Organization . . Terrorist Database Web Pages Figure 1: the architecture of Terrorist Web Miner Terrorist Database contains all the web pages which farther will be used to extract terrorist links among terrorist organizations. These web pages will be parsed to extract the terrorist links. The terrorist links can be used to create a linkage map of any terrorist attack. Linkage map can also be used to study the relationships to each organization/terrorist in the network as well as its can be used to study the role of to each organization/terrorist in the network.
- 3. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 270 5. IMPLEMENTATION OF TERRORIST WEB MINER The following figure shows the implementation of Terrorist Web Miner. This web miner is developed using Java and MySql database. Internet with social networking data Search.java Terrorist_Source.java . . UrlTable Common links SearchEngineUrl TableSearch Engine Name SourceTable Top 5 links Web Pages Figure 2: the architecture of Terrorist Web Miner There are 3 tables in Terrorist Tracker Database namely SearchEngineUrl, UrlTable and SourceTable. SearchEngineUrl Table contains the URLs of the search engines from which investigator wants to search results and may contain URLs of popular search engines like Yahoo! Search, Ask, Bing and Dogpile etc. Each of these search engine uses its own internal indexing system to rank the page for making searching effective. The task of mining terrorist networks is divided into two steps: Step 1: Extraction There is a java program named Search.java that will run the search on all the search engines stored in SearchEngineUrl Table. Search.java will have a search keyword for example “Terrorist attacks in India”. This search keyword will be used to the useful web pages. As a output of this keyword, these search engines will return number of links contain in a web page. Extracting these links from the web pages is a difficult task because a web page is a collection of header tags, anchor tags, links etc. Hence, parsing a web page is far more difficult than parsing a text document. Search.java will parse the web page returned by these search engines as a result of search. The links from the web page will be extracted and these links will be stored in UrlTable table along with their ranks. Ranks are allocated by the Terrorist Web Miner. As the frequency of the link URL increases its rank also increases i.e. In UrlTable table, ranks of the web pages changes dynamically. Hence the URL of the link that appears in almost all search engines will have highest rank. Step 2: Filtration In search, investigator may face the problem of Web Spamming. Web Spamming means increasing the rank of any page illegitimately. If the page have low rank it will not be returned as a search result irrespective of whether it has appropriate content or not. Hence to reduce this web spamming only 5 highest rank URLs are selected for farther study by Terrorist _Source.java program. This program will fetch the contents of the top 5 ranked URLs/ links stored in UrlTable table. If the content of these links/URLs contains the name Terrorist Organization then this web page is stored in SourceTable table. Hence SourceTable table will contain only those web pages which have Terrorist Organization information in them. Later on these web pages can be used to study the relationship between different Terrorist Organizations. 6. EVALUATION The following table shows the content of UrlTable table in MySql Database. UrlTable table have fields like SiteUrl and Rank. As described in implementation, Search.java will look for Terrorist attack information on popular search engines like Yahoo! Search, Ask, Bing and Dogpile etc. This program will parse the contents of the web pages and will look for the useful links. When the links is found, it is added in this table and if this link already exists in this table then simply rank is increased. This will reduce the level of the web spamming. When UrlTable table is ready Terrorist_Source.java will select top five links from this table. These top 5 links will be mined to get all the others links related to this topic because in web mining we assume that if current web pages are useful then other pages covering the related topic are likely to provide a like to it. Hence the implementation of Terrorist Web Miner is successfully done. Table: UrlTable Sr No SiteUrl Rank 1 www.satp.org/satporgtp/countries/india/database /index.html 5 2 www.ndtv.com/topic/india-terror-attacks 5 3 www.theguardian.com/world/mumbai-terror- attacks 4 4 en.wikipedia.org/wiki/Terrorism_in_India 3 5 www.india.com 3 6 en.wikipedia.org/wiki/List_of_terrorist_incident s_in_India 2 7 www.ibtimes.com/major-terrorist-attacks-india- over-last-20-years-timeline-1752731 2 8 www.chacha.com 2 10 en.wikipedia.org/Terrorist_incidents_in_India 2
- 4. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 04 Issue: 11 | Nov-2015, Available @ http://www.ijret.org 271 CONCLUSION The counter terrorism war can be fought and won only by superior knowledge and effective SNA tools. Terrorist Networks can be destabilized only when we have enough knowledge about terrorist networks and its topologies. Terrorist Web Mining aids in searching, extracting and collecting this knowledge. During this study we successfully created a terrorist web miner which crawls through different search engines and returns the related web pages. These web pages are then parsed for extracting the terrorist links from the web pages which may contain information about terrorist organizations. This links can be farther stored to create linkage map of terrorist networks. REFERENCES [1]. Sergey Brin and Lawrence Page, "The Anatomy of a Large-Scale Hypertextual Web Search Engine," in 7th World Wide Conference(WWW7), Brisbane, Australia, April 1998 [2]. Monika Yadav and Pradeep Mittal, "Web Mining: An Introduction," International Journal of Advanced Research in Computer Science and Software Engineering, vol. 3, no. 3, pp. 683-687, March 2013. [3]. V. E. Krebs, "Mapping networks of terrorist cells," Connections 24, pp. 43–52, March 2001 [4]. Yang Xiang, Michael Chau, Homa Atabakhsh, and Hsinchun Chen, "Visualizing criminal relationships: comparison of a hyperbolic tree and a hierarchical list," Decision Support Systems 41, pp. 69–83, 2005. [5]. Javad Hosseinkhani, Suriayati Chaprut, Hamed Taherdoost, Morteza Harati Cool, and Sadegh Emami Korani, "Criminal Communities Mining on the Web," International Journal of Advanced Computer Science and Information Technology (IJACSIT), vol. 2, no. 2, pp. 13-22, April 2013. [6]. R. D. Gaharwar, D. B. Shah, and G.K.S. Gaharwar, "Terrorist Network Mining: Issues and Challenges," International Journal of Advance Research in Science and Engineering, vol. 4, no. 1, pp. 33-37, 2015. [7]. D Zhou, R Manavoglu, J Li, C L Giles, and H Zha, "Probabilistic models for discovering e- communities," in 15th international conference on world wide web (WWW), 2006, pp. 173–182. [8]. R. D. Gaharwar, D. B. Shah, and G.K.S. Gaharwar, "Proposed Architecture for Terrorist Web Miner," International Journal of Computer Applications, vol. 128, no. 9, pp. 18-20, October 2015.