SPECIAL ISSUE ON COMPUTING AND COMMUNICATION
JUNE 2012
REVIEW OF APRIORI ALGORITHM FOR FINDING ASSOCIATION RULES
Supreet Singh, Munish Saini, Harpreet Singh
Association rule mining is the process of finding interesting and useful relationship between various data elements of database. As the size of database is growing so rapidly, efficient methods are required for finding association rules effectively. This paper presents a review of classical Apriori Algorithm and recent work done by various researchers to improve the performance of Apriori Algorithm.
A TREE BASED ALGORITHM FOR WEB PAGE CHANGE DETECTION SYSTEM
Srishti Goel, Rinkle Rani Aggarwal
People are using internet actively for exchange of information across the world resulting in uploading of information on web pages and updating of new web pages very frequently. The contents of web page changes continuously & rapidly. Hence it becomes very difficult to observe the changes made in web pages and retrieve the original web pages .For efficient retrieval and monitoring the changes made in web pages and compare the difference between refreshed page and old page efficiently that too in minimum browsing time, an effective monitoring system for the web page change detection based on user profile is needed. The web page change detection system can be implemented by using various Tools or Algorithms. In this paper, we will explain the algorithms to detect the changes in content and structure of a web page.
A TOKEN STRING BASED ALGORITHM FOR DATA EXTRACTION IN WEB USAGE MINING
Surbhi Anand, Rinkle Rani Aggarwal
World Wide Web is a massive repository of web pages which provides abundance of information for the Internet users due to which the size of web has become tremendously large. Web Usage Mining applies mining techniques in log data to extract the behaviour of users which is used in various applications like personalization, adaptive web sites, customer profiling, creating attractive web sites etc. A Web Usage Mining process consists of three phases: data preprocessing, patterns discovery and patterns analysis. Data preprocessing tasks must be performed prior to the application of mining algorithms to convert the raw data collected from server logs into the data abstraction. The appropriate analysis of web log file proves beneficial to manage the websites effectively from the administrative and users prospective. Data preprocessing results also have a direct impact on the later phases. Therefore, the preprocessing of web log file plays an essential part in the web usage mining process. This paper emphasizes on the mining of web access logs and makes some exploration in the field of data preprocessing.
CACHE OBLIVIOUS ALGORITHMS
Upasna Sharma Himanshu Aggrawal
This paper presents the approaches which help us to analyze the running time of an algorithm on a computer with a memory hierarchy with limited associativity, in terms of various cache parameters. The performance of an algorithm when implemented depends on many parameters: its theoretical asymptotic performance, the programming language chosen, choice of data structures, the configuration of the target machine. The cache efficient and disk efficient algorithms and data structure is the notion of cache oblivious, introduced by Frigo, Lieserson, Prokop, and Ramachandran in 1999. In this paper, we study cache resources efficiently. The hierarchy of memory models in modern computers.
.
N-GRAM BASED LANGUAGE IDENTIFICATION FOR WRITTEN TEXT
Navdeep Kaur, Mandeep Kaur, Nishi Sharma
Language identification technology is widely used in the domains of machine learning and text mining. Many researchers have achieved excellent results on a few selected European languages. However, the majority of African and Asian languages remain untested. The primary objective of this paper is to evaluate the performance of our new n-gram based language identification algorithm on languages written with Arabic script. In this paper we have worked for automatically identifying language for Arabic script documents. Our approach to the language identification problem is based on the n-gram analysis method. This technique achieves good performance, even with relatively small training sets. It works for three languages namely Urdu, Arabic and Shahmukhi. Then the system classifies the language of the given document. An overall accuracy of about 95.33% is achieved with test documents.
A Neural Network Implementing Back Propagation Sensing Jatropha’s Maturity Level
Kestina Rai Maitreyee Dutta Sunil Aggarwal
Abstract: Jatropha curcas ( Sanskrit : danti pratyanshrani) is a species of flowering plant in the spurge family, In 2007 Goldman Sachs cited Jatropha curcas as one of the best candidates for future biodiesel production. It is resistant to drought and pests, and produces seeds containing 27-40% oil, averaging 34.4%. The remaining press cake of jatropha seeds after oil extraction could also be considered for energy production. Traditionally, human experts perform the identification of Jatropha curcas. Its quality depends on type and size of defects as well as skin color and fruit size. Then a Grading System of Jatropha (GSJ) by using color histogram method was developed to distinguish the level of ripeness of the fruits based on the color intensity. Although this automated approach was better than the human expert identification but it only deals with one aspect of the fruit, that is, color. In this paper we propose an artificial neural network approach to build an expert system to measure the quality of the fruit based not only on the color intensity but also on other features of the fruit like fruit size and texture of the fruit, etc. Because this type of a system can learn from examples like humans and therefore can give better results.
IMPLEMENTATION OF LOW INTERACTION HONEYPOT TO IMPROVE THE NETWORK SECURITY
Gurdip Kaur, Dr. Gurpal Singh, Jatinder Singh Saini
Honeypots are increasingly used to provide early warning of potential intruders, identify flaws in security strategies, and improve an organization's overall security awareness. Honeypots cansimulate a variety of internal and external devices, including Web servers, mail servers, database servers, application servers, and even firewalls. Among the universe of security tools that have been developed to protect our networks, as Firewalls, IDS (Intrusion Detection Systems), IPS (Intrusion Prevention Systems), etc., there is a relative new kind of security tool called Honeypot. Spitzner, L. (2003) defines a Honeypot as follows: A Honeypot is a security resource whose value lies in being probed, attacked, or compromised. This security resource is so flexible, that the organization can use it to detect intrusion, unethical behavior of employees, delay attack to their networks, forensics, gather information of attacks to know how they were made, virus researching, etc. Honeypots are a highly flexible security tool with different applications for security. They have multiple uses, such as prevention, detection, or information gathering. Honeypots all share the same concept, a security resource that should not have any production or authorized activity. In other words, deployment of honeypots in a network should not affect critical network services and applications. A honeypot is a security resource whose value lies in being probed, attacked, or compromised.
PERFORMANCE ANALYSIS OF SERVICE BROKER AND LOAD BALANCING POLICY IN CLOUD COMPUTING
Mandeep Devgan KVS Dhindsa Mandeep Singh
Cloud computing is an on demand service in which shared resources, information, software and other devices are provided according to the clients requirement at specific time. The performance of an application gets affected with service broker policy and the load balancing policy used across different virtual machines in a single Data Center (DC). In this paper using different combinations of various algorithms of load balancing and service broker policy we analyzed the effect on performance of a cloud based application. In this paper we also explained two policies named service broker and load balancing in single DC, we used a CloudSim based analysis and modeling tool named CloudAnalyst.
ADVANCEMENT IN NETWORK SECURITY USING IPSEC
Mohit Kumar Parmpreet Singh Priya Joshi Reetika Aggarwal Ashish Jalota
IPsec (IP security) is a standardized framework for securing Internet Protocol (IP) communications by encrypting and/or authenticating each IP packet in a data stream. There are two modes of IPsec operation: transport mode and tunnel mode. In transport mode only the payload (message) of the IP packet is encrypted. It is fully-routable since the IP header is sent as plain text; however, it can not cross NAT interfaces, as this will invalidate its hash value. In tunnel mode, the entire IP packet is encrypted. It must then be encapsulated into a new IP packet for routing to work Internet Protocol security (IPsec) uses cryptographic security services to protect communications over Internet Protocol (IP) networks. IPsec supports network-level peer authentication, data origin authentication, data integrity, data confidentiality (encryption), and replay protection. The Microsoft implementation of IPsec is based on Internet Engineering Task Force (IETF) standards.
BUILDING A SUCCESSFUL CRM USING DATA MINING TECHNIQUES
Gaurav Gupta, Himanshu Aggarwal
Customer Relationship Management (CRM) refers to the methodologies and tools that help businesses manage customer relationships in an organized way. Advancements in technology have made relationship marketing a reality in recent years. Technologies such as data warehousing, data mining, and campaign management software have made customer relationship management a new area where firms can gain a competitive advantage. Particularly through data mining organizations can identify valuable customers, predict future behaviors, and enable firms to make proactive, knowledge-driven decisions. Data Mining is the process that uses a variety of data analysis and modeling techniques to discover patterns and relationships in data that may be used to make accurate predictions. Various techniques exist among data mining software, each with their own advantages and challenges for different types of applications. It can help you to select the right prospects on whom to focus. Variable selection and class distribution on the effects the performance in building a successful CRM.
AN EFFICIENT AND SECURE AUTHENTICATION SYSTEM- USING ENHANCED SERPENT TECHNIQUE
Raman Kumar
With the upcoming technologies available for hacking, there is a need to provide users with a secure environment that protect their resources against unauthorized access by enforcing control mechanisms. To counteract the increasing threat, enhanced serpent technique has been introduced. It generally encapsulates the enhanced serpent technique and provides client a completely unique and secured authentication tool to work on. This paper however proposes a hypothesis regarding the use of enhanced serpent technique and is a comprehensive study on the subject of using enhanced serpent technique (EST). This forms the basis for a secure communication between the communicating entities. Several password authentication protocols have been introduced each claiming to withstand to the several attacks, including replay, password file compromise, denial of service, etc. We introduced a new technique through which the people can communicate to each other securely using EST. In this technique, very small size key size of 128, 192 and 256 bits are needed for both the sender and receiver and then they can have a very secure communication between them, so the server which is providing the communication between sender and receiver will also do not have any knowledge about the way to decrypt the text. Therefore, the proposed scheme is secure and efficient against notorious conspiracy attacks.
CONCEPT MINING FROM DATA BASE TO FIND EXPERTS TOPICS
Ankita Dwivedi, Vridhi Madaan Batra, Archana Singh
This paper explores the concept extraction task which is an important step in natural language processing. Now a days need of automation in locating important terminology among a specific domain becomes crucial. The manual process for finding the concepts and their further refining requires extensive labor and is time consuming. Automation of the entire process reduces manual involvement in concept extraction. This journal mainly focuses on the application of finding domain experts by locating interesting topics and assigns those to the experts. A profile of experts is being maintained which gives the information about the expertise knowledge. After locating the concepts, manual verification is required as all the results are not relevant and experts of domain validate them.
A COMPREHENSIVE STUDY OF CLASSIFICATION OF NOSQL DATABASE
In this paper, we examine a number of NoSQL data stores designed to scale simple OLTP style application loads over many servers. Originally motivated by Web 2.0 applications, these systems are designed to scale to thousands or millions of users doing updates as well as reads, in contrast to traditional DBMSs and data warehouses. The NoSQL movement began early 2009 and is growing rapidly. Different NoSQL databases take different approaches. What they have in common is that they’re not relational. Their primary advantage is that, unlike relational databases, they handle unstructured data such as word-processing files, e-mail, multimedia, and social media efficiently.
STUDY OF DDOS ATTACKS USING DETER TESTBED
Daljeet Kaur, Monika Sachdeva, Krishan Kumar
In present era, the world is highly dependent on the Internet and it is considered as main infrastructure of the global information society. Therefore, the Availability of information and services is very critical for the socio-economic growth of the society. However, the inherent vulnerabilities of the Internet architecture provide opportunities for a lot of attacks on its infrastructure and services. Distributed denial-of-service (DDoS) attack is one such kind of attack, which poses an immense threat to the availability of the Internet. These attacks not only congest a Server by their attack, but also affect the performance of other Servers on the entire network also, which are connected to Backbone Link directly or indirectly. To analyze the effect of DDoS attack on FTP services, repeated research in cyber security that is vital to the scientific advancement of the field is required. To meet this requirement, the cyber-DEfense Technology Experimental Research (DETER) testbed has been developed. In this paper, we have created dumb-bell topology and generated background traffic as FTP traffic. Different types of DDoS attacks are also launched along with FTP traffic by using attack tools available in DETER testbed. Throughput of FTP server is analyzed with and without DDoS attacks.
VIRTUALISATION WITH CLOUD COMPUTING
Ms. Sagrika, Dr.Sandeep Sharma, Dr.G.S. Brar, Mr. Sunny Chanday
Cloud computing stands on decades of research in virtualization, distributed computing, utility computing, and, more recently, networking, web and software services. It implies a service-oriented architecture, reduced information technology overhead for the end-user, great flexibility, reduced total cost of ownership, on demand services and many other things. Virtualization is quickly becoming a vital technology across all parts of the IT environments since last couple of years. Virtualization is now in use across nearly all enterprises, and future plans to move some applications to Cloud Computing, because cloud computing will just compound the problems of virtualization or we can say both technologies are catalyzing each other. The similarities in both strategies are like both helps to reduce the size and control the expansion of data center to reduce the cost of hardware, power and cooling, space, management and disaster recovery but their initial and ongoing costs are differ. This paper focuses on the concept of cloud computing with the virtualization technology. Our study also concentrates on, if virtualization really helps to make the cloud better or both the technologies are making the work of IT industries complex.
COMPARISON OF DIFFERENT WAVELETS FOR SPECKLE DENOISING BY APPLYING HARD AND SOFT THRESHOLDING
Sandeep Kaur, Ranjit Singh
An image is often corrupted by noise in its acquisition and transmission. Image denoising is used to remove the noise while retaining as much as possible the important image features. In this paper, denoising of speckle noise is discussed with discrete wavelet transform (DWT) using different wavelets such as haar, daubechies, symlets etc. The two very basic thresholding techniques are applied on each of them and the comparison is done using parameters such as PSNR and MSE.
CLONE COST EFFECTS ?
Er. Amanpreet Kaur Chela, Er. Dhavleesh Rattan
The demand of softwares has increased with the development of technology and communication systems. With this the maintenance effects which is a vital factor. The softwares are not identical if we contrast them with past, present and future, due to development of new programming languages and their principles. To improve the quality of any software, maintenance is must. Cloning in source code files makes it difficult to modify. Several models are designed to overcome this problem. This paper presents the extension of analytical cost model to evaluate the cloning. As the size and the complexity of software increase, it also becomes essential to develop high-quality software, cost-effectively within a specified period. This paper presents a study on the cloned code, the large open source systems are used, various other new parameters are added to calculate clone.
AN ANALYSIS OF THE ENGINEERING STUDENTS ATTITUDE TOWARDS TECHNOLOGY ENHANCED LANGUAGE LEARNING
Dr.Gurleen Ahluwalia
This study investigates the students perceptions of using technological tools like blogs, wikis, Interactive software and the researchers language learning website as a means to supplement in-class language learning activities. This study evaluates a language laboratory program in which forty first year engineering students from a college in Punjab were introduced to these tools and instructed to use them for their laboratory work. The data collected reveals that despite encountering some difficulties, students had an overall positive attitude towards using these tools in their learning of English. The students find that learning English through technology is interesting and effective.
DESIGN OF RECONFIGURABLE DUAL-BEAM CONCENTRIC RING ARRAY ANTENNA USING BIOGEOGRAPHY BASED OPTIMIZATION
Urvinder Singh Dr. T.S. Kamal Ms. Gurkamal Kaur
This paper describes a method of designing a reconfigurable dual-beam antenna array using a novel Biogeography Based Optimization (BBO) algorithm. The objective is to generate a dual pencil beam patterns with two different pre-defined sidelobe levels from a concentric ring array of isotropic antennas. The two patterns differ only in radial phase distribution while sharing a common radial amplitude distribution. The pattern with lower sidelobe level is obtained by switching the phase distribution from zero value to the optimum value. The phases and amplitudes for elements in each ring is same but is varied radially. BBO is used to obtain optimum set of phase and amplitude distribution that will generate dual pencil beam.
IMPLEMENTATION OF PATTERN RECOGNITION SYSTEM USING NI VISION
Mohinder Pal Joshi, R.S Uppal, Livjeet Kaur
The emerging biometric personal identification systems have increased the need for accurate and efficient ways of Pattern recognition. Pattern recognition is the backbone of any biometric identification system and is widely used these days for the biometric personal identification systems. Pattern matching compares the user template with templates from the database using some matching metric. This paper presents a new and simple approach for Pattern recognition based on template matching. Pattern Matching is implemented in NI Vision Assistant and NI LabVIEW.
SEGMENTATION OF PROSTATE BOUNDARY FROM ULTRASOUND IMAGES USING ANT COLONY OPTIMIZATION
Vikas Wasson, Baljit Singh
Prostate Cancer & disease is one of the leading causes of death in adult & elderly man. For the success of the treatment, it is very important to detect the disease at early stages. Till now, manual contouring is the only reliable method used for this purpose. However manual contouring is a tedious & very time consuming task. In this paper, an automatic multi-stage algorithm for Prostate Boundary Detection from ultrasound images is presented. In the first stage, a Sticks Filter is used to enhance the contrast of the image & to reduce the speckle from it. Next, Initial Contour is determined from this enhanced image. The final step is to determine the volume of the Prostate by segmenting the Prostate Boundary using Ant Colony Optimization. In the last section, the performance of the present research is demonstrated by comparing it with Genetic Algorithms & Manual Outlining.
ASPECT ORIENTED SOFTWARE DEVELOPMENT- A FRAMEWORK FOR SOFTWARE REUSABILITY AND MAINTAINABILITY
Er. Mandeep Singh, Er. Balwinder Kumar, Prof. Satwinder Singh
Aspect -oriented software development (AOSD) is widely used in industries and in research work. It provides new abstractions and complexity dimensions for software engineering. As a result, Aspect oriented software development poses new problems to empirical software engineering. So it requires new frameworks which specifically measure the reusability and maintainability degrees of aspect-oriented systems. This paper presents a framework for aspect oriented software, which is contains two types of components: a suite of metrics and a quality model which are based on the principles and existing metrics of the software engineering. The AOSD framework has different characteristics and properties, varying control levels and different complexity degrees. So Based on this analysis, the advantages and drawbacks of the AOSD framework components are discussed.
IMPROVED APPROACH FOR SEMANTIC WEB SEARCH ENGINES AND AUTOMATIC DISCOVERY OF PERSONAL NAME ALIASES FROM THE WEB
In this paper we are presenting the new approach for efficient web search engines and extracting the personal name aliases from the web for social networking websites. Given a personal name, the proposed method first extracts a set of candidate aliases. Second, we rank the extracted candidates according to the likelihood of a candidate being a correct alias of the given name. The novel, automatically extracted lexical pattern-based method is proposed in order to efficiently extract a large set of candidate aliases from snippets retrieved from a web search engine. We define numerous ranking scores to evaluate candidate aliases using three approaches: lexical pattern frequency, word co-occurrences in an anchor text graph, and page counts on the web. To construct a robust alias detection system, we integrate the different ranking scores into a single ranking function using ranking support vector machines. In addition to this, there are many search engines techniques proposed in order answer the user queries efficiently and effectively, but they are vulnerable in answering intelligent queries from the user due to the dependence of their results on information available in web pages. The main focus of these search engines is solving these queries with close to accurate results in small time using much researched algorithms. However, it shows that such search engines are vulnerable in answering intelligent queries using this approach. They either show inaccurate results with this approach or show accurate but (could be) unreliable results. Thus we proposed layered model of Semantic Web which provides solution to this problem by providing tools and technologies to enable machine readable semantics in current web contents. For practical work, we build the proposed semantic web search engine and evaluate the proposed method of personal name aliases.
PERFORMANCE ANALYSIS OF HBASE ON HDFS- A COLUMN-ORIENTED DATABASE APPROACH
Jatinder Kaur Gurleen Kaur Dhaliwal
HBase is the open source distributed storage system for the management of large volumes of unstructured data. It is based on Google’s BigTable. It is non-SQL database system which provides an alternative way of storage as compared to traditional RDMS systems. This paper explores and evaluates HBase data storage on Hadoop Distributed file system (HDFS). The current works purpose is to convert simple relational schema into HBases column-oriented data store and then evaluate the performances of random writing and random reading of rows.
SECURITY OF LFSR BASED STREAM CIPHERS USING GENETIC ALGORITHM
Prof Rupinder Singh,Prof Jagjit Singh, Prof Lovejeet Singh
Stream Ciphers: It is a symmetric key cipher where stream of plaintext are mixed with a random cipher bit stream (key stream), typically by any logical operation. In the case of stream cipher one byte is usually encrypted at a particular time. In this paper, we combine two technologies, one is Linear Feedback Shift Register (LFSR) and other is Genetic Algorithm (GA). In this research we are proposing to develop an algorithm which encrypts our plain text into cipher text by using Genetic Algorithm. A cipher is a secret method of writing, whereby plaintext is transformed into cipher text. The process of transforming plaintext into cipher text is called encryption; the reverse process of transforming cipher text into plaintext is called decryption. Both encryption and decryption are controlled by cryptographic key parameters. Hence we are using linear feedback shift register with Genetic Algorithm. In this paper we propose a scheme wherein utilization of GA is explained with the help of LFSR. Here we have to develop two algorithms, one for sender side and another for receiver side.