Informatics Publishing Limited

J. Abuthahir ¹, S. Vaaheedha Kfatheen ², M. Nazreen Banu ³

Affiliations
1 Department of Computer Science, Jamal Mohamed College, Tiruchirappalli-20, Tamil Nadu, IN
2 Department of Computer Science, Jamal Mohamed College, Tiruchirappalli, TN, IN
3 Department of MCA, M.A.M. College of Engg, Trichirappalli, Tamil Nadu, IN

Abstract

The purpose of grid computing is to produce a virtual supercomputer by using free resources available through widespread networks such as the Internet. Resource management and scheduling plays a crucial role in achieving high utilization of resources in grid computing environments. Due to heterogeneity of resources, scheduling an application is significantly complicated and challenging task in grid system. Most of the researches in this area are mainly focused to improve the performance of the grid system. In this paper, a new task scheduling algorithm called Resource Allocation Algorithm (RAA), is proposed to the resource allocation model with multiple load originating processors as an economic model. RAA first estimates the neighbour node details and most recently used resources to schedule a task. Linear programming approach is used to execute the Task Monitor which it avoids delay in the execution of large tasks and supports concurrency in the execution of large and small tasks. Experimental results proved that the proposed Algorithm (RAA) on scheduling independent tasks within grid environments achieves comparatively lower makespan and better solution in terms of cost and time.

Keywords

Gird Workflows, Workload Distribution, Economic Model, Cost Optimization, RAA Algorithm.

Full Text

     

K. Fathima Bibi ¹, M. Nazreen Banu ²

Affiliations
1 Dept. of Comp. Sc., Bharathiyar University, Coimbatore, TN, IN
2 Dept. of Comp. Sc. & Engg., MAM College of Engineering, Tiruchirappalli, TN, IN

Abstract

Feature Selection plays an important role in data mining. Dealing with excessive number of features has become a computational burden on learning algorithms. Removing irrelevant and redundant features makes data mining task more efficient and improves its accuracy. In this review, different feature selection approaches, relation between them and the various learning algorithms are discussed. Applications that support the use of feature selection technique are also included. We conclude this work by reviewing the contribution of the various feature selection approaches.

Keywords

Feature Selection, Classification, Clustering, Supervised, Unsupervised, Semi-Supervised.

Full Text

     

J. Hajiram Beevi ¹, M. Nazreen Banu ²

Affiliations
1 Department of Computer Science and Information Technology, Jamal Mohamed College, Trichy, Tamilnadu, IN
2 Department of Computer Science and Engineering, M.A.M. College of Engineering, Trichy, Tamilnadu, IN

Abstract

This paper makes a broad survey on the various web 2.0 tools that are used to create learning resources in E-learning 2.0 environments. Interestingly, these tools provide the capacity for supporting modern and independent lifelong learners and they can be applicable for all the stages of e-learning system development. They shift the roles of a teacher into a mentor and also make the learners as active participants. This paper shows the influential contribution of Web 2.0 tools and their applications on higher education. Also, it provides some real time examples for creating learning resources using Web 2.0 tools. Further, it proves that Web 2.0 tools have a strong positive impact on e-learning which shapes the students' social attitudes. In addition, it presents the characteristics of various web 2.0 tools and analyzes each of them separately.

Keywords

Web 1.0, Web 2.0, E-Learning 1.0, RSS, E-Learning 2.0, Podcasting.

Full Text

     

K. Fathima Bibi ¹, M. Nazreen Banu ²

Affiliations
1 Department of Computer Science, Bharathiar University, Coimbatore – 641046, Tamil Nadu, IN
2 Department of MCA, MAM College of Engineering, Tiruchirappalli – 621105, Tamil Nadu, IN

Abstract

In data mining feature subset selection is a preprocessing step in classification that lessens dimensionality, eliminates unrelated data, increases accuracy and improves unambiguousness. The next step in classification is to produce enormous amount of rules from the reduced feature set from which high class rules are chosen to build effectual classifier. In this paper, Information Gain (IG) has been used to rank the features. Multi-Layer Perception (MLP) with back-propagation reduces features to achieve higher accuracy in classification. Artificial Neural Networks (ANN) classifier is used for classification. We handle the discretization of continuous valued features by dividing the series of values into a limited number of subsections. Wine Recognition data set taken from the UCI machine learning repository is used for testing. Original 13 features are drawn in classification. The thirteen features are reduced to five features. Experimental results show that the accuracy in training dataset is 98.62% and in the validation dataset is 96.06%. The accuracy difference between 13 features and 5 features in the training data is 5.54% and in validation data is 2.00%. We then build a Decision Tree and concentrate on discovering significant rules from the reduced data set that provide better classification.

Keywords

Back-Propagation, Classification, Decision Tree, Feature Subset Selection, Multi-Layer Perception

Full Text