Informatics Publishing Limited

S. J. Gnanasoundhari ¹, M. Balamurugan ¹

Affiliations
1 SCSEA, Bharathidasan University, Tiruchirappalli-620023, IN

Abstract

Data Mining is a process that extracts knowledge from a large amount of data. Data Mining has the capability for classification, prediction, estimation and pattern recognition. The Healthcare industry is generally rich in information but somewhat poor in knowledge. Data Mining plays a vital role in predicting the heart disease using the datasets. Many kinds of information are accessible in the prevision of heart disease. The Heart disease diagnosis is a complicated task which requires more experience and knowledge. The aim of this work is to create a MLPT, to predict Myocardial Infraction. After getting the patient information this MLPT, forecast that the patient is caused by heart attack or not which is performed by using three Data mining techniques: Naive Bayes, Decision tree and WAC (Weighted Associative Classifiers). Using the medical prognosis such as chest pain type, thalassic, slope etc., it can predict the probabilities of patients getting a heart disease in the future. The prediction is performed from extracting the patient's diachronic data or data storage. The research is mainly developed to recover the hidden information from the database. The system has been implemented in JSP and checked using the datasets that is been collected from UCI machine learning repository.

Keywords

Naive Bayes, Decision Tree, Weighted Associative Classifier (WAC).

Full Text

     

S. Vijaykumar ¹, S. G. Saravanakumar ², M. Balamurugan ³

Affiliations
1 School of Computer Science, Engineering and Applications, Bharathidasan University, Trichy – 23, Tamil Nadu, IN
2 6th SENSE, An Advanced Research and Scientific Experiment Foundation, Kumbakonam, Tamil Nadu, IN
3 School of Computer Science, Engineering and Applications, Bharathidasan University, Trichy – 23, Tamil Nadu, India, IN

Abstract

Today, The Computing architectures are one of the most complex constrained developing area in the research field. Which delivers solution for different domains computation problem from its stack above. The architectural integration constrains makes difficulties to customize and modify the system for dynamic industrial and business needs. This model is the initiation towards the solution for findings of Industry 4.0 and Bigdata needs. This “Unique sense” smart computing implementation model for Industry 4.0 holds the innovative Smart computing prototype is a part of “UNIQUE SENSE” computing architecture which can delivers alternate solution for today’s computing architecture to satisfy the future generation needs of diversified technologies and techniques, which brings extended support to the ubiquitous environment. Primitively the industrial 4.0 having a lots of chained interlinked process which also holds valuable information. So it is especially designed for fault tolerance data processing integrated system. This implementation model constructed in the way that smart control and self-accessible system for next generation cyber physical machine and automation controlling system. Also that focusing towards dynamic customization, reusability, eco friendliness for next generation controlling and computation power

Keywords

Bigdata, IOT, HPC, Industry 4.0, Industrial Model, Prototype, Smart Computing, Unique Sense

Full Text

   

Sarat Kumar Sahoo ¹, Dhaval Patel ¹, M. Balamurugan ¹, Ashwin Kumar Sahoo ²

Affiliations
1 School of Electrical Engineering, VIT University, Vellore - 632014, Tamil Nadu, IN
2 Department of EEE, SSN College of Engineering, Chennai - 603110, Tamil Nadu, IN

Abstract

Objectives This paper proposes a control methodology of bidirectional Converter for grid connected systems. Methods/ Statistical Analysis: The bidirectional converter is used for converting the DC-AC power by using decoupled current control strategy in stationary reference frame. The simulations of the proposed system are carried out in PSIM software. An experimental prototype of Bidirectional converter has been built in the laboratory. A TMS320F28335 digital signal processor (DSP) is used for generating the control pulses for the Converter. Findings: The control of grid converter along with battery backup would provide support to the grid. Thus the proposed control strategy could independently control active and reactive power and also able to avoid grid failures. Application/Improvements: Reduction of operating switching frequency, improved dynamics as well as robustness against grid impedance variation are some of the improvements of proposed system.

Keywords

Battery, Bidirectional Converter, DSP, Grid, PSIM.

Full Text

   

M. Balamurugan ¹, S. Kannan ¹

Affiliations
1 Department of Computer Applications, Madurai Kamaraj University, IN

Abstract

In Ensemble classifiers, the Combination of multiple prediction models of classifiers is important for making progress in a variety of difficult prediction problems. Ensemble of classifiers proved potential in getting higher accuracy compared to single classifier. Even though by the usage ensemble classifiers, still there is in-need to improve its performance. There are many possible ways available to increase the performance of ensemble classifiers. One of the ways is sampling, which plays a major role for improving the quality of ensemble classifier. Since, it helps in reducing the bias in input data set of ensemble. Sampling is the process of extracting the subset of samples from the original data set. In this research work, analysis is done on sampling techniques for ensemble classifiers. In ensemble classifier, specifically one of the probability based sampling techniques is being always used. Samples are gathered in a process which gives all the individuals in the population of equal chances, such that, sampling bias is removed. In this paper, analyse the performance of ensemble classifiers by using various sampling techniques and list out their drawbacks.

Keywords

Ensemble of Classifiers, Sampling, Random Forest, Boosting.

Full Text

     

M. Balamurugan ¹, S. Vijaykumar ², S.G. Saravanakumar ³

Affiliations
1 School of Computer Science, Engineering and Applications, Bharathidasan University, Trichy - 620024, Tamil Nadu, IN
2 School of Computer Science, Engineering and Applications, Bharathidasan University, Trichy - 620024, Tamil Nadu, IN
3 6th SENSE, An Advanced Research and Scientific Experiment Foundation, kumbakonam - 612001 Tamil Nadu, IN

Abstract

This study explores existing design principles of the processor architecture and identifies future design approach that will help to solve existing business problems that are operable on the scalable environment. We considered the two broader classifications of the instruction sets- RISC and CISC and analyzed the ways to improve the performance of the existing processor design approach. Findings show that all the design principles have been made for different engineering level points to work on the different kinds of task that are specific to the respective fields and SIMD data can be handled well in vector processing environments. Improvements can be made while choosing the design environment based on the business problem and significant design improvement can make to overcome the existing performance-related issues.

Keywords

CISC, MIMD, MISD, RISC, SIMD, SISD, Vector Processing, VLIW.

Full Text

   

Vijaykumar Selvam ¹, M. Balamurugan ¹

Affiliations
1 Department of Computer Science and Engineering, Bharathidasan University, Trichy (T.N.), IN

Abstract

Green revolution initiatives occurred between the 1930s and the late 1960s around the world, notably by Norman Ernest Borlaug, Cresco, Lowa, US. born.Later transformed to India by Mankombu Sambasivan Swaminathan, who was born in Kumbakonam, Tamil Nadu, IN. He was one who known as the father of the green revolution, also the back bone for green research findings of high-yield varieties of wheat and rice in India. Also, administrative contributions towards various green initiations. Today in the era of computing will deliver growth in various fields, including industries called Industry 4.0 in, similar to that Information and Communication Technology obtain development in various fields, which directly and indirectly related to our agriculture. Even Agriculture needs to be getting various up-gradation by which not simply by hardware and software. The modern scientific societies state that the era of developing software applications is not enough for the next generation growth, it needs technology growth for today’s findings. As same Identifying and incorporating, those innovative technologies based scientific findings for delivering the solution for next generation green revolution is the most challenging issue today.

Keywords

Agriculture, Green Revolution, Smart Computing, Big Data Analytics, Technology.

Full Text