Arun Kumar B R;Reddy C;Hiremath P S
000960 Arun Kumar B R;Reddy C;Hiremath P S (MCA Sir MVIT Dep, , Bangalore, Email: arunanjan@yahoo.com) : Cross layer design in wireless AD-HOC netwoks. J Analysis Computn 2008, 4(2), 121-8.
The performance of wireless systems is extremely sensitive to the mobility, to the dynamicity and to the environment. A number of researchers have looked at specific aspect of network performance applaying cross-layer design. They have presented different cross-layer design proposals which break away from traditional network design. In this article, we take a step to put together various individual efforts be put into perspective and a more holistic view be taken. This paper presents a survey of the literature in the area of cross-layer design. We present understanding, definition, motivation, example, disadvantages and open challenges of cross-layer design.
3 illus, 10 ref
Vidyarthi D P
023432 Vidyarthi D P (School of Computer and Systerms Sciences, Jawaharlal Nehru University, New Delhi) : Pre-allocation directives in parallel/distributed computing systems. Int J Inf Comput Sci 2009, 12(1), 13-24.
Parallel/Distributed Computing System caters to the high speed computing need and is the backbone of the modem high performance computing system. Job scheduling is one of the most important activities for any computing system, be it a uniprocessor or a parallel/distributed computing system. For a uniprocessor system, the scheduling decisions may be trivial but for a Parallel/Distributed System it is quite cumbersome as there are many factors to affect. In such systems, job (task) allocation onto the computing nodes is required for the scheduling purposes. The task allocation models in parallel/distributed computing system require some parameters based on which the allocation decision is made. Most of the models assume the parameters as pre-supplied i.e. provided by the applications itself. However, it is difficult for the applications to produce such parameters. The work, proposed here, elaborates the prior requirements of the allocation and the possible ways to determine these parameters on which the allocation decisions are made. These together form the pre-allocation directives.
4 illus, 3 tables, 17 ref
Umamaheswari J;Ravichandran S
023431 Umamaheswari J;Ravichandran S (Computer Science Dep, Govt., Arts and Science College for Women, Pudukkottai-622 001, Email: rajgokul2002@yahoo.com) : Quick mining algorithm (QMA) using pattern mining techniques for roomy data base. Scient Trans Envir Technovation 2009, 3(1), 25-9.
Pattern mining plays an essential role in many data mining tasks such as mining sequential patterns, multi-dimensional patterns, max-patterns, and emerging patterns. Pattern mining techniques can also be extended to solve many other problems involving computation and classification. Pattern mining is a significant research problem. This paper discusses the problem of efficient and effective pattern mining with an orientation towards a fast mining method.
10 ref
Ranjan D;Tripathi A K
023430 Ranjan D;Tripathi A K (Computer Science Dep, Faculty of Science, Banaras Hindu University, Varanasi-221 005, Email: ranjan_divya@yahoo.co.in) : Enhancing the efficiency of software testing with reuse: the roadmap. Int J Inf Comput Sci 2009, 12(1), 53-9.
Suggests the roadmap for incorporation of reuse specific activities into software testing processes which allows reuse of all kinds of software testing related experiences and thereby provides the means to achieve the desired magnitude of reduction in testing effort and enhancement in the test quality. A discussion is presented here on the need of defining reuse oriented software testing processes and a roadmap for the same.
^ssc10 ref
Pandey S K;Mishra R B
023429 Pandey S K;Mishra R B (Computer Science Dep, SMS, Varanasi, U.P., Email: Skphind@rediffmail.com) : Optimising web information discovery with modern techniques. Int J Inf Comput Sci 2009, 12(1), 25-40.
The Web is constantly changing and growing, and even the best es;imates can only approximate its extent and composition. Investigations of the typical user S preferences and search behavior have strongly influenced Web Information Discovery. Search engines are the most popular examples of Web Information Discovery tools. They contain four major components that interact together namely, the Crawler, Storage module, Parser, Indexer Query Processor and Ranking module. Efficient algorithms and data structures can make the difference between an average and an exceptional search engine. Search engines today have to index millions of pages. Studies text indexing in the context of web search engines for fast and relevant web information discovery. In particular the inverted file indexing algorithm for indexing is studied in detail. Different Data structures are compared in terms of the time required to create index, the time required to query the index and the space footprint.
8 illus, 21 ref
Dhaka V S;Rao M;Singh M P
023428 Dhaka V S;Rao M;Singh M P (Computing Science Dep, Maharaja Surajmal Inst of Tech, New Delhi) : Pattern classification for handwritten curve scripts of three alphabets using image density feature and conjugate descent technique of neural networks. Int J Inf Comput Sci 2009, 12(1), 41-52.
The image of handwritten text has quite many features like slanting, skewing, height, width, density etc. The pixel based intensity of a very important feature and this can be explored to get good recognition results. Discussed and worked upon a recognition system that takes density of an image into consideration with the conjugate descent of the error to obtain the optimized weight change. It has been proven that these techniques produce the efficient result for the classification of characters from the handwritten curve scripts of three letters. The intent behind this analysis is to process the image as a whole and recognize the letters from English alphabet that it contains. The density values of96 equally divided segments of the image are contributors to the input pattern for neural network. The row based and column based arrangement of these density values are two major differentiating experiments. The training algorithm of the ANN has been tweaked in-order to achieve higher classification results. The objective is to obtain the performance comparison of the ANN while processing the word image using varying training algorithms. Besides the gradient based and density based learning the weight update mechanism of these algorithm have also been altered to include-the second order momentum term. The network used is a feed forward - back propagating one. Introduction of additional second order momentum term in weight adjustment mechanism enhances the learning capacity of network. Significant results in achievement of reduced training iterations for the training patterns have already been described. This work is a step enhancement in the same direction. Remarkable achievements in meeting the learning goals are apparently evident as number of training iterations get reduced with row based input pattern with the data sets we tested.
3 illus, 3 tables, 14 ref
Datta M;Shrivastava A;Shrivastava R K
023427 Datta M;Shrivastava A;Shrivastava R K (Mathematics Dep, Agra College, Agra) : Queueing analysis of partial message discarding policy in an interdependent communication network. Int J Inf Comput Sci 2009, 12(1), 1-12.
Studies and analyze the Partial Message Discarding (PMD)policy to alleviate or avoid congestion in a packet switch communication network. In PMD policy, once a packet is dropped at a node, all subsequent packets from the message to which the packet belongs are dropped. Applies an Interdependent queueing model to a packet -switch communication network with the assumption that the arrival and the transmission processes at the network elements are correlated and follow a bi-variate Poisson process having the joint probability mass function of the form given by Milne [5] and Srinivas Rao.K. et al [6]. Expressions for the stationary queue length distribution and the goodput (defined as the ratio between the network node and the total exiting packets at the input) are presented.
3 tables, 7 ref
Ahmed A;Khan A A;Ahmad S P
023426 Ahmed A;Khan A A;Ahmad S P (Statistics Dep, University of Kashmir, Srinagar-190 006, Email: ruma_sheikh@yahoo.co.in ) : Applications of s-plus in data analysis. Int J agric statist Sci 2009, 5(1), 231-50.
S-PLUS is one of the standard softwares specially created for exploratory data analysis and statistics. In this paper, an attempt has been made to apply this software for statistical and exploratory data analysis. Simple as well as advanced features of data analysis along with graphics are covered for common users.
9 illus, 3 tables, 5 ref
Sridhar K;Sivaramakrishnan T R
022321 Sridhar K;Sivaramakrishnan T R (Shanmugha Arts, Science, Technology and Research Academy [Sastra] Sastra Univ, Thanjavur, Tamil Nadu) : Performance analysis of hash algorithms in software and reconfigurable FPGAs. Int J Comp Applic 2008, 3(1), 15-22.
In the digital world, the message digest algorithms play an important role in message authentication and data integrity checking. As the volume of data transmission and storage is increasing every day, the load on every CPU increases. In the next generation Internet, the IPV6, the IPSec is a mandatory security mechanism to be used to provide security to the data transmitted which needs the encryption and message authentication. Also, the data integrity assurance and data origin authentication are the two essential security services required in electronic commerce, software distribution and data storage. Hence the interest towards designing dedicated hardware to perform these operations is growing. Such a coprocessor can relieve the CPU from the routine operations and at the same time increase the throughput. Describes the implementation of the message digest algorithms MD5, SHA1 and SHA-2 in software and in FPGA and compares their performance in terms of throughput.
3 illus, 4 tables, 12 ref
Kharb L;Rajender Singh
022320 Kharb L;Rajender Singh (Computer Science & Applications Dep, M.D. Univ, Rohtak, Haryana) : PLHIM: a new object-oriented metric on hierarchical inheritance. Int J Comp Applic 2008, 3(1), 23-32.
Software metrics has been utilized to evaluate inheritance as well as to facilitate the designers in order to address cost estimation and product-quality across all life-cycle stages of developing the final product. By applying measurement through the different levels of class hierarchy, one could evaluate inheritance as well as reuse, in order to take into account the greater number of abstraction levels inherant in object-oriented systems (OOS). A new metric (PLHIM) of hierarchical inheritance is proposed that measures the quality of the program through different levels of "Object-Orientedness". The main idea behind proposed metrics and research work was to make use of measurement as a criterion for improvement in software development at different levels to minimize risk and this has been done by taking the problems of C++ and Java.
2 illus, 1 table, 9 ref
Jothi Venkateswaran C;Clementking A; Subramanian K
022319 Jothi Venkateswaran C;Clementking A; Subramanian K (Computer Science Dep, Presidency College, Chennai) : Natural language processing in multilingual email using universal networking languages. Int J Comp Applic 2008, 3(1), 43-52.
Describes about the design and implementation concept of Natural Language Processing in E-maii service application using Universal Networking Languages. Adopts the model that defines 'context' for natural language, which sends our mail as a query and gets the output in a desired language according to the users' desire. This proposed method consists of a dictionary, a thesaurus, relations and values between different regional languages across the world.
12 ref
Indira B;Sudha T
022318 Indira B;Sudha T (Computer Science Dep, Kasturba Gandhi Degree & PG, College for Women, West Marredpally, Sec'bad 26) : Applications of automatic number plate recognition. Int J Comp Applic 2008, 3(1), 53-9.
10 ref
Dhanalakshmi Y;Ramesh Babu I
022317 Dhanalakshmi Y;Ramesh Babu I (Computer Science & Engineering Dep, Acharya Nagarjuna Univ, Naarjuna Nagar, Guntur) : Mining optimized fuzzy association rules using genetic algorithms. Int J Comp Applic 2008, 3(1), 33-41.
A multi-objective genetic algorithm was proposed for optimizing association rules based on the fitness function and threshold value. The proposed algorithm is used for optimizing association rules having fuzzy characteristics. A fuzzy threshold value is used for identifying the correct fitness function and its members so that the number of iterations at every stage of Genetic Algorithms is reduced.
1 illus, 5 ref
Ananthapadmanaban K R;Jayasree S
022316 Ananthapadmanaban K R;Jayasree S (Computer Application Dep, S.R.M. Arts and Science College, Kattankulathur) : Bio-inspired for writer identification. Int J Comp Applic 2008, 3(1), 1-7.
Writer identification is one of the areas in pattern recognition that attract many researchers to work in, particularly in forensic and biometric application, where the writing style can be used as biometric features for authenticating an identity. The challenging task in writer identification is the extraction of unique features, in which the individualistic of such handwriting styles can be adopted into bio-inspired generalized global shape for writer identification. The feasibility of generalized global shape concept of complimentary binding in Artificial Immune System (AIS) for writer identification is explored. An experiment based on the proposed framework has been conducted to proof the validity and feasibility of the proposed approach for off-line writer identification.
2 illus, 2 tables, 10 ref
Abbas T P;Sanavullah M Y
022315 Abbas T P;Sanavullah M Y (A, B, C) : Identifying and removing the people-centered chaos in the software development. Int J Comp Applic 2008, 3(1), 61-7.
Development in the field of Software Engineering (SE) in the form of principles, methodologies, best practices, and tools, all have a single objective of setting order to the chaos inherent in the software development process. Despite impressive advancements in SE, studies have revealed that software development in most of the organizations is still chaotic. In the early years ofSE research, researchers were in pursuit of a silver technology bullet but it is now realized that technology is neither the problem nor the solution. The problem - and the solution-lay in people, processes and products. Extensive researches were reported to tackle the process-and product-centered chaos where as people-centered chaos was often ignored. This paper seeks to, analyze the root causes of the people-centred chaos in the software development with respect to a multiple case study of 100 software projects undertaken by various software firms in India during the year 2007. The Goal-Question-Metric approach is employed to guide the data collection procedure. The data collected is analysed using Pareto Analysis to identify the major contributors of people-centered chaos. A root cause analysis was performed to trace the root cause of the people-centered chaos. Finally, some remedial measures were suggested for reducing the people-centered chaos.
1 illus, 1 table, 4 ref
Sajja P;Sharda M
021276 Sajja P;Sharda M (Dep of Computer Science, Sardar patel University, Vallabh Vidyanagar-388 120) : Knowledge scouting in E-learning systems using semantic agents. Prajna 2008, 16(Oct), 30-6.
The education has taken a giant leap and paradigm shift from class room learning to semantic tutoring. Attempt in this research is to identify the problem with class room find the feasibility of e-learning system using semantic tutors, to pin point barriers and concerns of the students aspiring to go for e-learning using semantic tutors. The research attempts to identify, evaluate and present he outcomes of the study undertaken using standard techniques and statistical tools. Proposed a model for semantic e-learning and future implications for the further research in the area.
2 illus, 10 ref
Sah S K;Singh R S;Tripathi A K
021275 Sah S K;Singh R S;Tripathi A K (Dep of Computer Engineering IT-BHU, , Varanasi-221 005) : Static scheduling based on availability of resources in heterogeneous distributed computing system. Int J Inf Comput Sci 2008, 11(2), 1-5.
Presents a method for allocation of tasks of an application onto a set of heterogeneous distributed computing resources with prior knowledge of availability of some processors after some time interval from start time. In literature of static scheduling of an application in heterogeneous distributed computing, assumption of no internal job queue for the processors is followed. Considers the constraint of internal job queue for some processors i.e. availability of some processors after some time from start time.
2 illus, 5 ref
Jain M;Sharma G C;Singh N
021274 Jain M;Sharma G C;Singh N (NO, St. John's College, Agra-282 002, Email: jadouneetu@yahoo.co.in) : Performance analysis of double ended Markovian quene for machining system with spares. Int J Inf Comput Sci 2008, 11(1), 7-16.
Considers a machining system with M operating and S spare machines under the care of K2 servers. Spare machines are considered to be either cold-standby, warm-standby or hot-standby. The life-time distributions of operating and spare machines are assumed to be exponentially distributed with rates A and Ct,, respectively. The system may work in degraded mode in case of shortage of operating machines. There are tvo input queues, (i) queue of failed machines and (ii) the queue of servers; such that the failed machines feed one queue (Q/) and the server population feeds the other queue (QJ. As soon as the failed machine (server) joins the empty queue Q, (Q2) when the other queue Q2 (Qi) has at least one server (failed machine), the failed machine is attended by the server. Obviously the two queues can never coexist. Derives the system characteristics viz. average length of both queues, effective arrival rate, expected waiting times in the queues etc.. Finally, gives numerical examples to illustrate the effect of various parameters on performance measures.
6 illus, 19 ref
Gupta S K;Parveen Kumar;Chauhan R K
021273 Gupta S K;Parveen Kumar;Chauhan R K (NO, , BCET Gurdaspur, Punjab, Email: skgbcet1965@rediffmail.com) : Minimum-process coordinated checkpointing protocol for deterministic mobile distributed system. Int J Inf Comput Sci 2008, 11(1), 28-37.
A checkpoint algorithm for mobile computing systems needs to handle many new issues like: mobility, low bandwidth of wireless channels, lack of stable storage on mobile nodes, disconnections, limited battery power and high failure rate of mobile nodes. These issues make traditional checkpointing techniques unsuitable for such environments. Minimum-process coordinated checkpointing is an attractive approach to introduce fault tolerance in mobile distributed systems transparently. This approach is domino-free, requires at most two checkpoints of a process on stable storage, and forces only a minimum number of processes to checkpoint. But, it requires extra synchronization messages, blocking of the underlying computation or taking some useless checkpoints. Proposes a minimum-process coordinated checkpointing algorithm for deterministic mobile distributed systems, where no useless checkpoints are taken, no blocking of processes takes place, and anti-messages of very few messages are logged during checkpointing. Also address the related issues like: failures during checkpointing, disconnections, concurrent initiations of the algorithm and maintaining exact dependencies among processes.
2 illus, 29 ref
Guha D;Mandal S
021272 Guha D;Mandal S (Dr H C Roy Engineering College, , P.O.-Fuljhore, Jemua Road, Durgapur-713 206, Email: DGUHA69@Rediffmail.com) : News corpus development. Int J Inf Comput Sci 2008, 11(2), 1-5.
Corpus is generally defined as a collection of documents. Corpus can be developed from news. There are several types of corpora presently available. Researches are going on to develop news corpus. Because everything is available from web. Everyday news is available from the web. But problem is one can not process the news directly from the web because there are some other extra documents like links, text boxes etc. In webpage, there are some special tags called meta tags which are very much useful to retrieve the necessary information from the web. So if you have a news corpus it will be very useful for several application.
5 ref
Ghosh D N;Ghosh K
021271 Ghosh D N;Ghosh K (NO, Dr. B.C. Roy Engineering College, Durgapur-713 206) : Time variations of the monthly sunspot data by using wavelet transform. Int J Inf Comput Sci 2008, 11(1), 1-6.
It has been used Simple Denoising Algorithm using Wavelet Transform on the monthly sunspot data from January 1749 to February 2005 and the method exhibits periodicity around 132, 149, 168, 374, 483, 687, 819, 987, 1056, 1186, 1339, 1457, 1621, 1736, 1898, 2024, 2172, 2275, 2381, 2506, 2643, 2769, 2900 and 3019 months.
3 illus, 13 ref
Aggarwal R;Aggarwal H
021270 Aggarwal R;Aggarwal H (Dep of computer Sc & Engg, Thapar University, Patiala-147 004, Email: raggarwal@thapar.edu) : On performance analysis of regular and irregular MINs. Int J Inf Comput Sci 2008, 11(2), 18-26.
Regular and irregular MINs are analyzed and compared in terms of performance parameters namely bandwidth, probability of acceptance, throughput and processor utilization. Regular networks considered are Extra Stage Cube (ESC) and Augmented Shuffle Exchange (ASEN-2(, and Irregular networks are Four Tree (FT) Network and Modified Four Tree Network (MFT).
5 illus, 4 table, 7 ref
Yadav S K;Sarvesh Kumar;Joshi L
020275 Yadav S K;Sarvesh Kumar;Joshi L (Library & Information Science Dep, U.P. Rajshri Tandon Open Univ, , Allahabad, Uttar Pradesh) : Use of internet services on central library, Indian Institute of Technology, New Delhi. Libr Prog 2008, 28(2), 251-66.
Study aim studying the use of Internet by Teaching staff, researchers, students and library staff in IIT Library, New Delhi. Structured questionnaire was designed to collect the data. Out of 75 questionaires, 75 were received back. The survey found that 85.3% respondents use internet for email followed by 86.7% for by faculty, 84.6% by research purpose, 60% respondents say Internet added value to library for accessing information.
16 ref
Singh A;Sharma G
020274 Singh A;Sharma G (JMIT Radaur Dep, , Haryana) : Bried survey of network security & intrusion detection systems. Acta Cienc Indica-Math 2008, 34(3), 1175-86.
"Information security has become a very critical aspect of modern computing systems. With the global acceptance of the Internet, virtually every computer in the world today is connected to every other. While this has created tremendous productivity and unprecedented opportunities in the world we live in, it has also created new risks for the users of these computers. The users, businesses and organizations worldwide have to live with a constant threat from hackers and attackers, who use a variety of techniques and tools in order to break into computer systems, steal information, change data and cause havoc." [Rajesh Hukku, Chairman and Managing Director, i-flex solutions, India]. Intrusion Detection Systems (IDSs) are an important, evolving technology for protecting computer networks. Although firewalls have traditionally been seen, as the "first line of defense" against would be attackers, intrusion detection software is rapidly gaining ground as a novel but effective approach for making networks more secure. In this Paper basics of different network security tools are covered with emphasis on Intrusion Detection Systems. Here efforts are being made to throw light on various aspects of basic elements of Intrusion detection systems (IDSs):
2 illus, 21 ref
Pradeep Kumar P B;Sivanarayana G;Ramesh Babu C;Srinivasulu R
020273 Pradeep Kumar P B;Sivanarayana G;Ramesh Babu C;Srinivasulu R (Extension Education Dep, Agricultural College, Bapatla-522 101) : Study on constraints encountered while browsing internet by the students of agricultural college, Bapatla - an appraisal. Andhra Agric J 2008, 55(3), 415-16.
^ssc1 table, 2 ref
Pant D;Lohani M C
020272 Pant D;Lohani M C (Computer Science Dep, SSJ Campus, Uttar Pradesh) : Security problems in wireless local area networks with the suggested possible solution. Acta Cienc Indica-Math 2008, 34(3), 1369-74.
Wireless LAN is one of the best ways to improve data connectivity in an existing building without the expense of installing a structured cabling scheme to every desk. Problems with the physical aspects of wired LAN connections (locating live data outlets, loose patch cords, broken connectors, etc.) generate a significant volume of helpdesk calls. Use of WLAN can effectively reduce these problems. Only obstacle for the advancement of WLAN is the problem of security flaws. This paper presents the basic techniques used in WLAN and analyzes the security features of it. This analysis brings few solutions to the security related problems of the WLANs.
12 ref
Mishra M K;Ngene C U
020271 Mishra M K;Ngene C U (Computer Engineering Dep, Faculty of Engineering, Maiduguri Univ, Maiduguri, Borno State, Nigeria, Email: mishrasoft1@gmail.com) : Awareness and importance of computer in the university of Maiduguri. Libr Prog 2008, 28(2), 243-9.
Computer is the most versatile electronic machine man has ever created. Computers have made a great impact pn our everyday life. The aim of this study is to know the awareness and importance of computer in the University of Maiduguri. Structured questionnaire were designed to collect the data. Out of 200 questionnaires, 150 were received back. The survey found that 81.34% respondents are computer literate, 98%are interested in computer and 50.67 respondents have computer in their home. Computer is very important for 51.34% respondents and 42% respondents use computer for internet. The results from the analysis of the responses showed that 32% respondents are interested in programming area of computer and 41.33 respondents want to do degree level course in computer.
7 tables, 3 ref
Fayaz K;Ramachandra G A;Akthar P M;Sriharsha A V
020270 Fayaz K;Ramachandra G A;Akthar P M;Sriharsha A V (CS&T Dep, S.K. Univ, Anantapur, Andhra Pradesh, Email: skim_fayaz70@rediffmail.com) : Efficient imputation method on large data: information-hash-theoretic approach. Libr Prog 2008, 28(2), 267-80.
Missing data causes a variety of problems in data analysis. First, lost data decrease statistical power. Statistical power refers to the ability of an analytic technique to detect a significant effect in a data set. Second, missing data produce biases in parameter estimates and can make the analysis harder to conduct and the results harder to present. To overcome the missing data problem, a Component-based Framework for imputation is proposed, which efficiently searches the most plausible value for replacement (Jin Zhou 2006). The algorithm developed ensures complete imputation, the first phase of imputation is based on the complexity of 1) finding the missing value entry, 2) select the category of attribute set, 3) generate the characteristic weight, 4) search the characteristic weight, 5) find the location of missing value and 6) replace the plausible value with missing value.
9 ref
Dharmadhikari V
020269 Dharmadhikari V (e-Governance Group, Ministry of Communication and Information Technology, Government of India, New Delhi, Email: vinay2000@gmail.com) : Education for capacity building: new strategies to include ICT. Inf Stud 2007, 13(3), 131-8.
The social development commitments, variously endorsed by the Government of India, the World Summits and the UNESCO/UN- sponsored forum (E9, EFA, Health for all etc.) are achievable. But these can be achieved if only fragmentation / compartmentalization within an Department / Ministry of the Government (State/Federal/ Local) is avoided; and is totally re-engineered to develop as a New Education System that integrates across all relevant compartments, channels of communication, and developmental facilities, and is focused on strengthening community development endeavors. Furthermore, the electronic communication media (satellite, TV, FM radio, VCP, PC-network with modems, etc.) make it now possible to design and deliver such a system in the distance-education mode, to the unreached rural masses in the remotest corners of the country. Prof. G. Ram Reddy, noted that Distance Education is like the "Panchayat Raj" in governance: it enables decentralization, and hence localization; and empowerment of the common man. What is urgently needed is a design for creating an empowering system of education that is freed from the single-minded pursuit of wealth and is dedicated to nurturing the hidden potential within every individual member of society, a system designed to usher in a new and ever-advancing civilization. The paper highlights the inadequacies of the current education system and of education administration, and concludes by proposing a few principles to guide the efforts to bring about a radical transformation within the Indian society and open new vistas of opportunities for the advancement of the nation.
6 ref
Parimala;Nallaswamy R
019236 Parimala;Nallaswamy R (NO, National Institute of Technology, Tiruchirappalli-620 015) : Machine learning by support vector machine techniques in liver disorder database. Bull pure appl Sci-Sect E 2008, 27(2), 421-30.
"Machine learning by support vector machine techniques in liver disorder database" frames the analysis of support vector machine to predict the onset of liver disorder. The liver disorder dataset is taken from the Machine learning Repository of U.G.Irvine. The experiment using different conditions by varying the kernel functions and their parameter value is investigated.
5 illus, 1 table, 10 ref
Paranjothi N;Rajagopalan V;Kalyanaraman R
019235 Paranjothi N;Rajagopalan V;Kalyanaraman R (Statistics Dep, Annamalai Univ, Annamalai Nagar-608 002) : Simple analysis for a discrete-time two-phase queueing system with exhaustive batch service. Bull pure appl Sci-Sect E 2008, 27(2), 307-19.
Discrete-time, two-phase queueing system with exhaustive batch service is studied. The customers arrive at the system according to a Bernoulli process and receive service in the two phases. The first phase is of batch type seivice and the second phase is of individual type service. The probability generating function's (PGF's) of the system size and sojourn times are obtained. Also we show that the PGF of the system type decomposed into two PGFs, one of which is the PGF of the system size in the standard discrete-time Geo/G/1 queue without vacations. Some performance measures are also derived.
1 illus, 2 tables, 17 ref
Su Cheng H;Chien Sing L
017228 Su Cheng H;Chien Sing L (Faculty of Information Technology, Multimedia University, Malaysia Cyberjaya, Malaysia) : Efficient preprocesses for fast storage and query retrieval in native XML database. IETE Tech Rev 2009, 26(1), 28-40.
XML (extensible mark-up language) has emerged as one of the popular data representation standards for information storage and exchange. Proposes an extended INLAB architecture, INLAB2, focusing on preprocessing the XML document for fast native storage and accurate query retrieval. Firstly, the authors proposes their xParse parser to check the well-formedness of an XML document. Next, uses a (self-end) labeling scheme to encode each element in the XML database, by its positional information, to establish parent-child (P-C) or ancestor-descendant (A-D) relationships between nodes. Subsequently, the TwigINLAB2 algorithm is used to optimize query retrieval. TwigINLAB2 is a generalization of TwigStack, the stack-based algorithm for matching twig query. However, the TwigStack algorithm is efficient for A-D relationship queries only. Thus, in order to overcome this limitation, enhances query retrieval by utilizing indices to speed up the matching and merging phases. Experimental results indicate that TwigINLAB2 can, on an average, process twig queries 23% better than the TwigStack algorithm and 10% better than TwigINLAB1, in terms of execution time.
11 illus, 5 tables, 27 ref
Sharma N K;Singh M P;Sanjeev Kumar
015029 Sharma N K;Singh M P;Sanjeev Kumar (Computer Science Dep, , Knowledge Park II, Greater Noida-201 306, Email: nivan101@gmail.com) : Mathematical formulation for the second derivative of back propagation error with non-linear output function in feed forward neural networks. Asian J expl Sci 2009, 23(1), 77-94.
The feed forward neural network architecture is particularly uses back propagation learning algorithm to determine the weights between different interconnected layers. This learning procedure uses gradient decent technique to apply on a sum of square error function for the given patterns. It evolves iterative procedure to minimize an error function by adjusting the weights in a sequence of steps. The first derivate of the error with respect to the weights identifies the local error surface in decent directions. Therefore, for every distinct presented pattern, the network exhibits distinct local error surfaces and the weights modify to minimize the current local error surface. Determination of optimal weights is only possible when the global error is supposed to be minimized. In this paper, we are providing the generalized mathematical formulation for the second derivative of the error function for arbitrary feed forward neural network topology. The local error is back propagated among the units of hidden layers and the second derivatives of the error with respect to the weights of hidden and output layer independently as well as combined1y, has been obtained. The new global error point can evaluate with the help of current global and minimized local error. The weight modification process accomplishes two times, firstly for present local error and secondly for current global error. The proposed method indicates that, the determined weights by the minimization of the global error are more optimal then the conventional gradient decent approach.
1 illus, 17 ref
Bhasin R;Venkata P P K
015028 Bhasin R;Venkata P P K (NO, BITS Pilani, Goa Campus, Goa, Email: rajeshy_bhasin46@yahoo.com) : Development of a cross-platform framework to assist computational fluid dynamics (CFD) users. J Coop Univ Res Ind Enterpr 2009, 2(1), 38-48.
Describes a project which aims to design and develop a framework to assist researchers working in the field of Computational Fluid Dynamics (CFD). Huge datasets result from the simulation of CFD problems on computers. Thus it is necessary to have faster and efficient methods to analyze these large amounts of abstract mathematical data to assist decision makers who study CFD problems. The project aims to help them by developing an interface which presents this abstract data in the form of 3D images after converting it from the binary format which is stored in CFD General Notation System (CGNS) files. The interface is user friendly and does not require much programming expertise from the user. The paper also discusses various methods used to develop the interface keeping in mind various user requirements and design issues in interface development. Suitable examples and rationale for the design approach are given at appropriate places. Towards the end we discuss the different features of the Graphical User Interface (GUT) and provide a comparison with the existing software.
9 illus, 1 table, 9 ref
Tanbeer S K;Ahmed C F;Jeong B S
013911 Tanbeer S K;Ahmed C F;Jeong B S (Computer Engineering Dep, Kyung Hee Univ, 1 Seochun-dong, Kihung-gu, Youngin-si, Kyunggi-do, 446-701, Email: tanbeer@khu.ac.kr) : Parallel and distributed algorithms for frequent pattern mining in large databases. IETE Tech Rev 2009, 26(1), 55-66.
Mining frequent patterns (FP) from large-scale databases has emerged as an important problem in the data mining and knowledge discovery research community. A significant number of parallel and distributed FP mining algorithms have been proposed, when the database is large and/or distributed. Among them, paral-lelization of the FP-growth algorithm using the FP-tree has been proved to be more efficient, when compared to the Apriori-based approaches. However, the FP-tree based techniques suffer from two major limitations - multiple database scans requirement (i.e., high I/O cost) and huge communication overhead. Therefore, in this paper, we propose a novel tree structure, called PP-tree (Parallel Pattern tree) that significantly reduces the I/O cost by capturing the database contents with a single scan and facilitates efficient FP-growth mining on it. Our parallel algorithm works independently at each local site and merges the locally generated global frequent patterns at the final stage, thereby reducing inter-processor communication overhead and getting a high degree of parallelism. Extensive experimental study on datasets of different types reflects that parallel and distributed FP mining with our PP-tree is highly efficient on large databases.
6 illus, 1 table, 19 ref
Suri P;Sudesh Rani
013910 Suri P;Sudesh Rani (Computer Science & Applications Dep, Kurukshetra Univ, Kurukshetra, Email: dhillon_sudesh@rediffmail.com) : Parallelism in relational database management systems. Punjab Univ Res J Sci 2007, 57, 217-26.
Parallelism is the key to realizing high performance, scalable, fault tolerant database management systems. With the predicted future database sizes and complexity of queries, it has become necessary to exploit parallelism in query processing. In this paper we discuss the intraquery parallelism in a relational database management system (DBMS). Provides a broad framework for the study of the numerous issues that need to be addressed in supporting parallelism efficiently and flexibly. Classification analysis of architectures of parallel database systems is also discussed. Discusses the major issues in DB2 intraquery parallelism.
6 illus, 21 ref
Su-Cheng H;Lee C S
013909 Su-Cheng H;Lee C S (Faculty of Information Technology, Multimedia Univ, Malaysia Cyberjaya, Malaysia, Email: schaw@mmu.edu.my) : Efficient preprocesses for fast storage and query retrieval in native XML database. IETE Tech Rev 2009, 26(1), 28-40.
XML (extensible mark-up language) has emerged as one of the popular data representation standards for information storage and exchange. Proposes an extended INLAB architecture, INLAB2, focusing on preprocessing the XML document for fast native storage and accurate query retrieval. Firstly, we propose our xParse parser to check the well-formedness of an XML document. Next, we use a (self-end) labeling scheme to encode each element in the XML database, by its positional information, to establish parent-child (P-C) or ancestor-descendant (A-D) relationships between nodes. Subsequently, our TwigINLAB2 algorithm is used to optimize query retrieval. TwiglNLAB2 is a generalization of TwigStack, the stack-based algorithm for matching twig query. However, the TwigStack algorithm is efficient for A-D relationship queries only. Thus, in order to overcome this limitation, we enhance query retrieval by utilizing indices to speed up the matching and merging phases. Experimental results indicate that TwigINLAB2 can, on an average, process twig queries 23% better than the TwigStack algorithm and 10% better than TwigINLAB1, in terms of execution time.
11 illus, 5 tables, 27 ref
Singh H;Sood S;Rupinder Kaur;Ratti N
013908 Singh H;Sood S;Rupinder Kaur;Ratti N (Computer Science and Engineering Dep, G.N.D. Univ, Amritsar-143 005, Email: hardeep_gndu@rediffmail.com) : Metrics framework for reengineering process. Punjab Univ Res J Sci 2007, 57, 251-5.
Reengineering is the examination and alteration of an existing system to reconstitute it into a new form and the subsequent implementation of the new form. The objective is to maintain the existing functionality and prepare for functionality to be added later. Software Metrics measure certain properties of a software project by mapping them to numbers (or other symbols) according to well defined, objective measurement rules. Usually, measurements are made to provide a foundation of information upon which decisions about software engineering tasks can be both planned and performed better. In this paper, a metrics framework has been proposed, that can be used to calculate reengineering requirement cost (RRC) & reengineering requirement cost of module (RRCM). On the basis of the results obtained by this metric, a decision can be made regarding maintenance/reengineering/retirement need of the software/part of software.
2 illus, 3 tables, 14 ref
Ramesh Singh;Sharma H C;Ambrish Kumar
013907 Ramesh Singh;Sharma H C;Ambrish Kumar (NO, National Research Centre for Agroforestry, Jhansi, Uttar Pradesh) : Artificial neural network modelling of reference evapotranspiration for Garhwal region of Uttarakhand. Indian J Soil Conserv 2008, 36(3), 131-5.
Artificial Neural Network (ANN) technique was used for modelling of reference evapotranspiration for Nagini watershed located near Chamba-Ranichauri, on Rishikesh-Uttarkashi route, in Tehri Garhwal district of Uttarakhand, India. Using pattern matching capability of ANN, the network was trained to recognize the pattern of daily weather data and its corresponding ETo obtained by Penman-Monteith method. More than 250 architectures were tried and 3 - 6 - 4 - 1 was observed to be the best architecture. The developed model was validated by testing its performance using correlation coefficient, root mean square error and coefficient of variation of the residual error. On the basis of these performance indicators the developed ANN model performed satisfactory and it can be utilized effectively for prediction of ETo for the study area.
2 illus, 1 table, 26 ref
Manoj M;Jacob E
013906 Manoj M;Jacob E (Computational Modelling and Simulation Unit, National Institute of Interdisciplinary Sciences & Technology, Trivandrum-695 019, Email: www.manoj.m@gmail.com) : Information retrieval on internet using meta-search engines. J scient ind Res 2008, 67(10), 739-46.
Meta-search engines (MSEs) on Internet have improved continually with application of new methodologies. Understanding and utilisation of MSEs are valuable for computer scientists and researchers, for effective information retrieval. This paper reviews functionality, working principles, growth, evolution and popularity of various MSEs.
^iia3 illus, 72 ref
Kumar S;Singh R
013905 Kumar S;Singh R (Mechanical Engineering Dep, S V National Institute of Technology, Surat-395 007, Email: skbudhwar@med.svnit.ac.in) : Expert system for selection of piloting for sheet metal work on progressive die. J scient ind Res 2008, 67(10), 774-9.
An expert system (PLTSEL) has been developed for selection of piloting to assist die designers and process planners working in stamping industries. Production rule-based expert system approach, used to construct PLTSEL, is coded in AutoLISP language and designed to be loaded into prompt area of AutoCAD. Low implementation cost makes system affordable for small and medium size stamping industries.
3 illus, 1 table, 29 ref
Gupta A;Singla R K
013904 Gupta A;Singla R K (Computer Science & Applications Dep, Panjab Univ, Chandigarh-160 014, Email: anugupta, rksingla}@pu.ac.in) : Using defect repository to determine free/open source software quality. Punjab Univ Res J Sci 2007, 57, 239-49.
Invention of Internet can be termed as a major breakthrough in the field of Information and Communication Technology (ICT). The expansion of Internet in the 90's inspired a novel model of software development in form of Free/ Open Source Software (F/OSS). The F/OSS development practices are giving rise to a new view of how complex software systems can be constructed, deployed and evolved. The reason for the increasing proliferation of F/OSS is not only the low cost and easy access to the code, but also consistently high quality of software products. The quality and suitability is achieved through an elaborate peer-review process performed by a large community of users, who act as co-developers to identify and correct software defects and request or add features. These characteristics give a large boost to software development resulting in a more effective software production, with true field testing and fast defect fixing. Publicly accessible defect data of five F/OSS hosted on a collaborative development site is used to determine the impact of defect handling in F/OSS maintenance quality. The results indicate that the defect locating effectiveness and the defect resolution efficiency is gradually strengthening which is resulting in gradual improvement in F/OSS quality. The insights gained from the current study can be supplemented with additional data about the projects and their defects to build models, perform analyses and provide valuable feedback to further improve F/OSS quality.
7 illus, 32 ref
Dutta K
013903 Dutta K (NO, National Institute of Technology, Hamirpur) : Survey of web servers in educational domains. Punjab Univ Res J Sci 2007, 57, 227-37.
Investigation of the trend occurring between the type of organization whom the website belongs to and the operating system platform along with the web server for hosting website in the given Indian domains. The study employed recent configuration of websites. The trends reflect the preference to proprietary software's in government and commercial websites. Educational sites use open source software. The incremental change in operating system and web server platform is also higher among educational sites, which reflects better level of commitment and awareness about the current resources among the staff responsible for maintaining the availability of websites in the educational Institutions. The paper concludes that educational Institutions in India are already making effective use of the resources by deploying open source software. Government of India also need to reorient its approach toward open source software deployment in government related services and thereby save money allocated for Information Technology to be spent on better activities.
5 tables, 28 ref
Colak H;Demirbas S;Sefa I;Irmak E;Tolga Kahraman H
013902 Colak H;Demirbas S;Sefa I;Irmak E;Tolga Kahraman H (Gazi Electrical Machines and Energy Control Group, Electrical Educatio, Gazi Univ, 06500, Ankara, Turkey, Email: icolak@gazi.edu.tr) : Remote controlling and monitoring of HVAC system over internet. J scient ind Res 2008, 67(9), 680-4.
In the study, internet-controlled heating ventilating and air-conditioning (HVAC) system has been proposed with programmable sleeping time and automatic operation mode, three steps fan speed unit, adjustable fan angle, a remote control device with an LCD and a web based control unit. A low cost microcontroller (MC) to control HVAC system and a PC as an Internet server are used. The system has three different control units (remote control by a hand-device, keypad control mounted on HVAC, and web based control). Each control unit has same menu options for users. A data acquisition board provides communication between server and MC. A webcam is used to monitor HVAC parameters and room environment remotely over the Internet. The system uses HTTP protocol to control devices. Proposed web based control method remotely works for long distance. As HVAC systems have high voltage, driver circuits with opto-couplers have been used for safety of the system.
5 illus, 16 ref
Kumar S;Sharma B K;Sharma P;Shamshi M A
012899 Kumar S;Sharma B K;Sharma P;Shamshi M A (NO, Central Scientific Instrument Organization (CSIO), Chandigarh-160 030) : 24 Bit seismic processor for analyzing extra large dynamic range signals for early warnings. J scient ind Res 2009, 68(5), 372-8.
Modified design is presented of existing 24 bit seismic data recorder comprising PC -architecture using PCI bus, ISA bus, and PC 104 bus in a single module to develop a flexible measurement set up. Elaborates use of building blocks [Disk on chip (DoC), GPS based timing unit, signal-processing module, and efficient software packages] worked out in visual C++ to develop compact sized instrument for quick decision-making with minimum error detection of true events. Paper describes Ethernet connectivity use for data downloading in a laptop without interruption of event data acquisition. Software packages for conversion of recorded data into SUDS and SEISAN formats have been realized and incorporated.
Kanchandani K B;Hussain M A
012898 Kanchandani K B;Hussain M A (Electronics & Telecommunication Dep, SSGM College of Engineering, Shegaon-444 203) : Emotion recognition using multilayer perceptron and gener alized feed forward neural network. J scient ind Res 2009, 68(5), 367-71.
Explores performance analysis of multilayer perceptron neural network (MLPNN) and generalized feed forward neural network (GFFNN) for detection of 7 human emotions (neutral, anger, boredom, disgust, fear, happiness, sadness) using speech signals. Overall accuracy was found as follows: MLPNN, 93%; and GFFNN, 99%.
Dutta S;Mandal J K
012897 Dutta S;Mandal J K (NO, B C Roy Engineering College, Durgapur-713 206) : Ensuring information security through 123-bit recursive substitution of bits through prime-nonprime detection of sub-stream (RSBP). J scient ind Res 2009, 68(7), 584-91.
Presents a 123-bit private-key based block-cipher, RSBP, which is capable of encrypting files of up to 11 MB. It is formulated on the basis of base-10 value corresponding to a block of bits, which is to be checked if is prime or not. It results in an alteration of size for file being encrypted. Its executable performance is analyzed on the basis of execution time, graphical layout of frequency distribution of characters, and Chi square values for varying degrees of freedom. RSBP is found to be highly compatible with existing cryptosystems.
Demirdag O;Murat Y S
012896 Demirdag O;Murat Y S (Civil Engineering Dep, Engineering Faculty, Pamukkale University, 20070 Kinikli, Denizli, Turkey) : Free vibration analysis of elastically supported Timoshenko columns with attached masses using fuzzy neural network. J scient ind Res 2009, 68(4), 285-91.
Presents elastically supported Timoshenko column with attached masses for free vibration analysis using fuzzy neural network. Neuro fuzzy frequency estimation (NFEE) models were developed to compute vibration frequencies using fuzzy logic toolbox of software Matlab 7.0. Gaussian membership functions and adaptive neuro fuzzy inference system (ANFIS) were used in NFFE model. Hybrid learning rule was applied for quantifying output variables in NFFE model. Frequency values of column with 1, 5 and 10 attached masses were computed. Training sets for NFFE models used transfer matrix method (TMM). During testing of NFFE model, good agreement was observed with results obtained using TMM as reduction in computation effort.
Akramifar S A;Ghasem Sani G
012895 Akramifar S A;Ghasem Sani G (Computer Engineering Dep, Sharif University of Technology, P O Box 11365-9517, Tehran, Iran) : Heuristic search by guided enforced hill climbing in fast forward automated planning. J scient ind Res 2009, 68(4), 278-84.
Enforced hill climbing (EHC), a heuristicaa search method, has been frequently used in a number of AI planning systems. This paper presents a new form of EHC, guided enforced hill climbing (GEHC), to enhance EHC efficiency. Main feature in GEHC is an adaptive ordering function. GEHC has shown a significant improvement in EHC efficiency, especially when applied to larger problems.
Venkata Subbaiah K;Naidu M M;Subba Rao D V
011884 Venkata Subbaiah K;Naidu M M;Subba Rao D V (CSE Dep, K.S.R.M. College of Engineering, Kadapa, Email: kvsubbaiah2@yahoo.com) : Cluster-head fuzzy routing for mobile Ad-hoc networks. Karpagam J Comput Sci 2008, 2(2), 537-46.
Mobile ad-hoc networks operate in the absence of any supporting infrastructure. The absence of any fixed infrastructure in mobile ad-hoc networks makes it difficult to utilize the existing Techniques for network services, and poses number of various challenges in the area. The discovery and maintenance of energy aware routing is most flinty challenge. In recent years, many approaches and techniques have been explored for the optimization of energy usage in Mobile Ad-hoc networks. Routing is one of these areas in which attempts for efficient utilization of energy have been made. These attempts use fixed (crisp) metrics for making energy-aware routing decisions. In this paper, we present a novel generalized fuzzy, logic-based approach for energy aware routing in Mobile Ad hoc networks. This generalized approach is soft and tunable and hence it can accommodate ad hoc networks comprising of different types of Mobile nodes.
7 illus, 1 table, 25 ref