Welcome to Journal of Theoretical and Applied Information Technology

Title:

A COMPARISON OF RK-FOURTH ORDERS OF VARIETY OF MEANS AND EMBEDDED MEANS ON MULTILAYER RASTER CNN SIMULATION

Author:

R.Ponalagusamy, S.Senthilkumar

Source:

Journal of Theoretical and Applied Information Technology    07-14, 2007

Abstract

In this paper an adaptable algorithm for simulating CNN arrays is implemented using various RK-fourth order means such as Arithmetic Mean [AM], Centroidal Mean [CM], Harmonic Mean [HM], Contra Harmonic Mean [CoM], Heronian Mean [HeM], Geometric Mean [GM] and Root Mean Square [RMS] also, it is compared with RK-fourth order embedded means such as the RK-Embedded Heronian Mean, RK-Embedded Centroidal Mean, Harmonic Mean and Contra-Harmonic Mean. The role of the simulator is that it is capable of performing Raster Simulation for any kind as well as any size of input image. It is a powerful tool for researchers to investigate the potential applications of CNN. This article proposes an efficient pseudo code exploiting the latency properties of Cellular Neural Networks along with well known RK-Fourth Order numerical integration algorithms. Simulation results and comparison have also been presented to show the efficiency of the various means in Numerical integration Algorithms. It is observed that the RK-fourth order embedded means outperforms well in comparison with RK-fourth order means. In particular it is found that the RK-Embedded Heronian Mean outperforms well in comparison with the RK-Embedded Centroidal Mean, Harmonic Mean and Contra-Harmonic Mean.

Title:

Tracking Control of 3-DOF Robot Manipulator Using Genetic Algorithm Tuned Fuzzy PID Controller

Author:

Srinivasan Alavandar , M.J.Nigam

Source:

Journal of Theoretical and Applied Information Technology    15-24, 2007

Abstract

Control of an industrial robot includes nonlinearities, uncertainties and external perturbations that should be considered in the design of control laws. This paper presents the Genetic algorithm tuned Fuzzy PID controller (GAFPID) to trace the desired trajectory for a three degree of freedom (DOF) robot arm. Numerical simulation using the dynamic model of three DOF robot arm shows the effectiveness of the approach in trajectory tracking problems. Comparative evaluation with respect to PD, PID and Fuzzy PID controls are presented to validate the controller design. The results presented emphasize that a satisfactory tracking precision could be achieved using the proposed controller than conventional controller.

Full Text

Title:

Author:

P.M.Khilar, S.Mahapatra

Source:

Journal of Theoretical and Applied Information Technology    25-44, 2007

Abstract

This paper proposes a multi-cluster hierarchical distributed system-level diagnosis approach for large-scale self-diagnosable distributed systems, such as one formed by overlaying mobile adhoc networks (MANETs) over another class of adhoc network, known as wireless sensor networks (WSN). The proposed diagnosis strategy assumes multiple numbers of initiators that initiate the diagnosis process in contrast to a diagnosis strategy having one initiator node, which creates a bottleneck in the system. The diagnosis strategy also avoids a costly distributed diagnosis algorithm where every node is an initiator of the diagnosis process. The approach enables the diagnosis at the host level where some external operator can collect all the diagnostic information accessing any active node after issuing a command to the initiator nodes at the highest layer in the hieararchy. The sensor nodes in the lowest layer in the hierarchy are static tested only nodes and thus do not maintain any kind of diagnosis information. Key results of this paper include an adaptive heartbeat-comparison based testing mechanism and fault models and an efficient and scalable distributed diagnosis algorithm using clustering that provides every active node a global diagnostic view of all the nodes. The correctness proof of the algorithm has been given. The analysis of algorithm has shown that the diagnosis latency and message complexity of the algorithm are O(lc(Tx + Tf + max(Tout1,Tout2)) + Txcg ) and O(ncCs) respectively. The simulation results show that the diagnostic latency, message complexity and diagnosability of the proposed clustering approach is better than the equivalent non-clustering approach.

Title:

Soft computing-Neural networks Ensembles

Author:

K.Venu Gopala Rao, P.Prem Chand, M.V.Ramana Murthy

Source:

Journal of Theoretical and Applied Information Technology    45-50, 2007

Abstract

Neural Network ensemble is a learning paradigm where a collection of finite number of neural networks is trained for the same task. It is understood that the generalization ability of neural networks, i.e., training many neural networks and then combining their predictions. ANN ensemble techniques have become very popular amongst neural network practitioners in a variety of ANN application domains. There are many different ensemble techniques, but the most popular include some elaboration of bagging and boosting or stacking. When applied to Ann’s, ensemble techniques can produce dramatic improvements in generalization performance.

Since this technology behaves remarkably well, recently it has become a very hot topic in both neural networks and machine learning communities, and has already been applied to diversified areas such as face recognition, optical character recognition, etc. In general, a neural networks ensemble is constructed in two steps, i.e., training a number of component neural networks, then combining the component predictions.

Title:

GENERIC DYEING AND COLOR CORRECTION

Author:

Jyoti Tandukar

Source:

Journal of Theoretical and Applied Information Technology    51-60, 2007

Abstract

A novel approach for computer color matching is proposed. In contrast to the existing computer color matching processes, this approach does not need the database of colorants with their absorption and scattering coefficients in order to predict the dyeing recipe and correct it. Despite various factors of the color device, dyes, dyeing medium, and dyeing conditions, the color can be corrected using this algorithm based on available color information of the sample and its real time output. The basic aim has been to predict recipe for reproducing desired color by using only three primary subtractive colors: Cyan, Magenta, and Yellow, then be able to correct it dynamically for a reproduction close to the target. This generic process has other applications too, like displaying the same color across different monitors, or reproducing color from a sample to print.

Title:

APPLICATIONS OF DATA MINING TECHNIQUES IN PHARMACEUTICAL INDUSTRY

Author:

Jayanthi Ranjan

Source:

Journal of Theoretical and Applied Information Technology    61-67, 2007

Abstract

Almost two decades ago, the information flow in the pharmaceutical industry was relatively simple and the application of technology was limited. However, as we progress into a more integrated world where technology has become an integral part of the business processes, the process of transfer of information has become more complicated. Today increasingly technology is being used to help the pharmaceutical firms manage their inventories and to develop new product and services. The implications are such that by a simple process of merging the drug usage and cost of medicines (after completing the legal requirements) with the patient care records of doctors and hospitals helping firms to conduct nation wide trials for its new drugs. Other possible uses of information technology in the field of pharmaceuticals include pricing (two-tier pricing strategy) and exchange of information between vertically integrated drug companies for mutual benefit. Nevertheless, the challenge remains though data collection methods have improved data manipulation techniques are yet to keep pace with them.
Data mining fondly called patterns analysis on large sets of data uses tools like association, clustering, segmentation and classification for helping better manipulation of the data help the pharma firms compete on lower costs while improving the quality of drug discovery and delivery methods. A deep understanding of the knowledge hidden in the Pharma data is vital to a firm’s competitive position and organizational decision-making. The paper explains the role of data mining in pharmaceutical industry.
The paper presents how Data Mining discovers and extracts useful patterns from this large data to find observable patterns. The paper demonstrates the ability of Data Mining in improving the quality of decision making process in pharma industry.
 

Title:

GAUSSIAN NOISE ESTIMATION TECHNIQUE FOR GRAY SCALE IMAGES USING MEAN VALUE

Author:

Saravanan C, Ponalagusamy R

Source:

Journal of Theoretical and Applied Information Technology    68-73, 2007

Abstract

The usage of digital image becomes ubiquitous. Also, the digital images are processed using digital devices. There are many mathematical techniques available to estimate the Gaussian noise of reproduced digital image. Assessing quantity of the Gaussian noise in a digital image is a difficult task. There are few factors affecting the process of digitizing images. The electronic devices used for acquiring images are the cause of the Gaussian noise. In this paper, a mathematical technique is proposed to estimate the Gaussian noise in the reproduced digital image. The proposed technique estimates quantity of the Gaussian noise in the reproduced image in a better way. The proposed technique is compared with Root Mean Square Error (RMSE), Peak Signal to Noise Ratio (PSNR), and Structural Similarity Index value (SSIM). This experiment shows that the proposed technique is suitable for estimating the exact amount of Gaussian noise in the reproduced image than the other mentioned techniques.

Title:

ATM NETWORK PLANNING: A GENETIC ALGORITHM APPROACH

Author:

Susmi Routray, A.M.Sherry, B.V.R.Reddy

Source:

Journal of Theoretical and Applied Information Technology    74-79, 2007

Abstract

The Asynchronous Transfer Mode (ATM) network is expected to become a backbone network for high speed multimedia services as the demand for applications such as internet and video on demand increases. One of the major issues in ATM network is the design. The design of an optimal ATM network is a complex comprehensive task. ATM network based on Passive Optical Network (PON) is one such solution. The deployment of optical fiber in the local access network is an essential step towards the provision of advanced ATM network services to the end user. Considering the strategic and financial implications for communications providers, it is clearly very important that fiber networks are implemented in a cost-effective manner. This paper demonstrates an optimization based approach using Genetic Algorithm (GA) for network planning. The optimal backbone ATM network design is characterized by the requirement to minimize the cost of fiber ducts. The objective of the optimization is to install a minimum net present cost network that satisfies the customer demand criterion. In this paper GA has been used to optimize the ATM backbone network. In addition GA has been used to provide end user connectivity. From the results obtained it can be inferred that computer based technique using genetic algorithm is a powerful tool for reducing the complexity of the planning task and ATM network based on PON provides a cost effective solution to ATM design.

Title:

MULTI FOCUS IMAGE FUSION BASED ON THE INFORMATION LEVEL IN THE REGIONS OF THE IMAGES

Author:

R.Maruthi, Dr.K.Sankarasubramanian

Source:

Journal of Theoretical and Applied Information Technology    80-85, 2007

Abstract

An image fusion algorithm based on activity measures like Spatial frequency and Visibility for fusing multi focus images is presented in this paper. For each sub window in the source multi focus images, the spatial frequency and Visibility is calculated. The fusion procedure is performed by a selection mode according to the magnitude of the spatial frequency and Visibility. The fused images are then assessed using the same activity measures that is used for fusion. Experiments results shows that the proposed algorithm works well in multi focus image fusion

Title:

Outline of an Agent Based Approach for a Dynamically Distributed Service

Author:

Prof. A.Damodaram, I.Ravi Prakash Reddy

Source:

Journal of Theoretical and Applied Information Technology    86-96, 2007

Abstract

The concept of distributed computing implies a network / internet-work of independent nodes which are logically configured in such a manner as to be seen as one machine by an application. They have been implemented in many varying forms and configurations, for the optimal processing of data. Various benefits, e.g. speedup, scale up, enhanced reliability, resource sharing etc, are accrued on their optimal dynamic exploitation.

Agents and multi-agent systems are useful in modeling complex distributed processes. They focus on support for (the development of) large-scale, secure, and heterogeneous distributed systems. Research in this domain includes scalable and secure agent platforms, location services, directory services, and systems management. They are expected to abstract both hardware and software vis-à-vis distributed systems.

For optimizing the use of the tremendous increase in processing power, bandwidth, and memory that technology is placing in the hands of the designer, a Dynamically Distributed Service (to be positioned as a service to a network / internet-work) is proposed. This paper examines the rationale for, and features of, such a service. An agent approach with thread migration is recommended as the scheme of implementation. The basic mechanism underlying the scheme is discussed, along with the data structures that are migrated to implement the service

Title:

A Neural Network Approach in Medical decision systems

Author:

K.Venu Gopala Rao, P.Prem Chand, M.V.Ramana Murthy

Source:

Journal of Theoretical and Applied Information Technology    97-101, 2007

Abstract

Artificial Neural Networks are useful for pattern recognition and also popular as classification mechanisms in medical decision support systems despite the fact that they are unstable predictors An important application of Gene Expression Data is classification of biological samples or prediction of clinical and outcomes. In this paper a method is proposed that combines statistical technique and Artificial Neural Network(ANN) to identify the prostate cancer diseased genes from normal genes and classify them using metrics call values. The system has 5 steps: 1.Data Collection along with filtering 2. Preprocessing of data using the gene selection method 3.Dimension reduction using statistical method 4.Classification using neural networks. 5. Comparing the results of gene selection followed by ANN and dimension reduction followed by ANN with varying number of predictors chosen from the gene selection method. The subset of genes that contribute significantly to the success of the neural classifiers are identified.

Title:

GENETIC ALGORITHM BASED ROBOT MASSAGE

Author:

Srinivasan Alavandar, K Adhi vairava Sundaram, M.J.Nigam

Source:

Journal of Theoretical and Applied Information Technology    102-109, 2007

Abstract

In this paper, a new robot massage experimental setup for leg using genetic algorithm based camera calibration is presented. TeachMover, a five axis articulated robot is used to press the muscle from ankle to knee. The real leg massage problem is approximated by a frustum shaped model, which can be easily extended to real leg massage. Three different sensors that are encoders; mounted at each joint of the robot with six degrees of freedom, a calibrated camera and a grip switch; mounted at the wrist of the manipulator were used. Camera calibration is done with the help of an algorithm proposed by Qiang Ji et. al [1] to estimate internal and external camera parameters using seven control points. The distance between camera and the robot is assumed to be fixed. By estimating the position and orientation of the object, which is the frustum model, the linear trajectory is found which the robot follows. The result shows the feasibility of the use of above-mentioned approach. The algorithm works satisfactorily for wide range of varying parameters i.e. the position and orientation of the model.

Title:

DECISION TREE INDUCTION USING ROUGH SET THEORY – COMPARATIVE STUDY

Author:

Ramadevi Yellasiri, C.R.Rao, Vivekchan Reddy

Source:

Journal of Theoretical and Applied Information Technology    110-114, 2007

Abstract

Dimensional reduction has been a major problem in data mining problems. In many real time situations, e.g. database applications and bioinformatics, there are far too many attributes to be handled by learning schemes, majority of them being redundant. Taking predominant attributes reduces the dimensions of the data, which in turn reduces the size of the hypothesis space, allowing classification algorithm to operate faster and more efficiently. The Rough Set (RS) theory is one such approach for dimension reduction. RS offers a simplified search for predominant attributes in datasets. Rough Set based Decision Tree(RDT) is constructed based on the predominant attributes. The comparative analysis with the existing decision tree algorithms was made to show that the intent of RDT is to improve efficiency, simplicity and generalization capability.