Informatics Publishing Limited

M. Sivajothi ¹, E. R. Naganathan ²

Affiliations
1 Departmaent of Computer Science in Sri Parasakthi College for Women, IN
2 School of Computer Science and Head of the Department of Computer Science and Engineering, Alagappa University, IN

Abstract

The major problem in Ad hoc Networks is to find a route between the communication end points. The topology of the network changes constantly due to the mobility of the nodes and paths which were initially competent and can quickly become incompetent or even infeasible. Ant Colony Optimization (ACO) is a biological inspiration simulating the ability of real ant colony of finding the shortest path between the nest and food source. Ant colony algorithms are motivated by the observation of real ant colonies. ACO is one of the successful applications of Swarm Intelligence (SI) which is the field of Artificial Intelligence (AI) that studies the intelligent behavior of ants. In this paper, Multiple Ant Colony Optimization based routing protocol is used to support multimedia communication in Ad Hoc Network. This approach avoids the stagnation problem on ants. This approach increases effectiveness and adaptiveness. Moreover, tabu search is used in this approach which avoids the blind alley problem of ants.

 

Keywords

Swarm Intelligence (SI), Ant Colony Optimization (ACO), Stagnation, Tabu Search.

Full Text

     

P. Sasikala ¹, E. R. Naganathan ²

Affiliations
1 Department of Computer Science, Sriram College of Arts & Science, Perumalpattu, IN
2 Department of CSE, Hindustan University, Chennai, IN

Abstract

This paper presents implementation of Neuro-Fuzzy methods for improving web personalization. The objective of this work is to predict the next useful page for the user based on his previous visits in the website. Due to increase in the number of users on a particular web page, there is continued research going to cater to the users next expected web page. This increases the business for the website launcher. Now a days lots of online transactions are performed in the form of purchase of new products, selling of second hands products. Personalization helps inexperienced users to make transactions quickly with less difficulty. Many methods have evolved over the period to improve personalization of web usage. Existing artificial neural network (ANN) algorithms are combined with Fuzzy logic to form Neuro-Fuzzy logic methods for improving the performance of personalization of web usage.

Keywords

Artificial Neural Network, Web Server Log, Web Usage Mining, Data Mining, E-Learning, User Access Patterns.

Full Text

     

E. R. Naganathan ¹, S. Narayanan ², K. Ramesh Kumar ²

Affiliations
1 Department of Computer Applications, Velammal College of Management and Computer Studies, Ambattur-Redhills Road, Chennai – 600066 Tamil Nadu, IN
2 Dept. of Computer Sci. & Engg., Alagappa University, Karaikudi-630003, Tamil Nadu, IN

Abstract

Frequent itemset discovery algorithms have been used to solve various interesting problems over the year.  As data mining techniques are being introduced and widely applied to non-traditional itemsets, existing approaches for finding frequent itemsets cannot be used as they cannot model the requirement of these domains.  An alternate way of modeling the objects in these data sets, is to use a graph to model the database objects.  Within that model, the problem of finding frequent patterns becomes that of finding subgraphs that occur frequently over the entire set of graphs.   Modeling objects using graphs allows us to represne tarbitrary relations among entities.  In this paper we present a computationally efficient algorithm for finding the ranking of such frequent subgraphs.  The subgraph finding method may follow any one of the existing algorithm.  In order to find out the ranking of subgraphs we present a new method called “graphgain”.  A graphgain is the normalization technique applied at each position for a chosen value of Discounted Cumulative Gain (DCG) of a subgraph.  The DCG alone cannot be used to achieve the performance from one query to the next in the search engine’s algorithm.  To obtain the graphgain an ideal ordering of DCG (IDCG) at each position is to be found out.  For this, a Modified Dicounted Cumulative Gain using “lift” is introduced here and IDCG is also evaluated.  Then the graphgain is evaluated.  Finally, the graphgain for all rules can be averaged to obtain a measure of the average performance of a search engine’s ranking algorithm.  And also the ordering of graphgain will provide the order of evaluation of rules which gives in turn the efficient ranking of subgraph process.

 

Keywords

Graphgain, Lift, Discounted Cumulative Gain.

Full Text

     

V. Saravana Kumar ¹, E. R. Naganathan ²

Affiliations
1 Centre for Bioinformatics, Pondicherry University, IN
2 Department of Computer Science and Engineering, Hindustan University, IN

Abstract

Hyperspectral image analysis is a complicated and challenging task due to the inherent nature of the image. The main aim of this work is to segment the object in hyperspectral scene using image processing technique. This paper address a novel approach entitled as Segmentation of hyperspectral image using JSEG based on unsupervised cluster methods. In the preprocessing part, single band is picked out from the hyperspectral image and then converts into false color image. The JSEG algorithm is segregate the false color image properly without manual parameter adjustment. The segmentation has carried in two major stages. To begin with, colors in the image are quantized to represent several classes which can be used to differentiate regions in the image. Besides, hit rate regions with cognate color regions merging algorithm is used. In region merging part, K-means, Fuzzy C-Means (FCM) and Fast K-Means weighted option (FWKM) algorithm are used to segregate the image in accordance with the color for each cluster and its neighborhoods. Experiment results of above clustering method could be analyzed in terms of mean, standard deviation, number of cluster, number of pixels, time taken, number of objects occur in the resultant image. FWKM algorithm results yields good performance than its counterparts.

Keywords

Cluster, Region Growing, Hit Ratio Region, Class-Map, Quantize.

Full Text

     

N. Vivekananthamoorthy ¹, E. R. Naganathan ², R. Rajkumar ³

Affiliations
1 Information Technology, SRM University, Kattankulathur, Chennai - 603203, Tamil Nadu, IN
2 Computer Science and Engineering, Hindustan University, Padur, Chennai - 602203, Tamil Nadu, IN
3 Computer Science and Engineering, SRM University, Kattankulathur, Chennai -603203, Tamil Nadu, IN

Abstract

Background/Objectives: In recent times, there is phenomenal increase in usage of Social Networking Sites like Facebook, LinkedIn etc. by college students and young professionals. This study focuses on identifying key factors that influence LinkedIn usage and the role of eWOM communication in enhancing social connectivity and engagement of students in meaningful activities to improve their social and academic standings. A theoretical model on social networking by students is proposed and the results and recommendations of this study will be brought to practical use towards student empowerment. Methods/Statistical Analysis: A preliminary survey was conducted to understand how young university students use the Social Networking Site LinkedIn and the responses were used to frame a questionnaire. A second level survey was conducted among the same set of participants by collecting their responses in five point Likert Scale. Exploratory Factor Analysis was conducted using the LinkedIn Survey responses to identify the hidden factors associated with the indicator items in the data set. Subsequently, a theoretical model was constructed using Structural Equation Modeling principles, depicting the interrelationships between the latent constructs and indicator items constituting a measurement model and a structural model. Four Hypotheses were framed such that Social Media Usage and eWOM communication have significant positive effect on Student Empowerment. Finally, Confirmatory factor analysis was done to prove the hypotheses and to analyze how well the model fits into the theory. The software IBM SPSS, and AMOS 23 were used to perform multivariate statistical analysis on the LinkedIn Survey response items. Findings: The exploratory study on LinkedIn Usage Survey responses revealed three latent factors that accounted for 69.462 percent of the total variance. The three key factors explaining the eWOM behavior of students in LinkedIn usage were Expert Opinion Seeking, Networking with Professionals and Notification of Profile Changes. The latent factors and associated relationships were used to frame a theoretical model based on SEM techniques. Based on Confirmatory factor analysis done on this model using the data set revealed that the model supported the hypotheses H1, H2, H3 and H4 and all indicators in the model significantly loaded to their respective factors and the predicting variables had a significant positive effect on the predicted variable. The factor loadings were fair to excellent ranging from .634 to .853 and the test for model fitness showed good fitness result based on value of various fitness indices which were within accepted limits. Based on CFA, the important fitness indices and their values arrived at were: CMIN/df = 2.022, NFI = 0.824, TLI = 0.887, RMSEA = 0.098 and CFI = 0.901. Improvements/Applications: The accuracy of the predicting ability of the proposed theoretical model can be improved by augmenting this research study and statistical analysis to be extended to a larger target group belonging to different institutions to achieve good model fit as well as for testing the scalability of the model. As a future work, this model can be integrated with online learning systems also with the aim of improving student engagement in the current online learning scenarios.

Keywords

Confirmatory Factor Analysis, eWOM Communication, Exploratory factor Analysis, Multivariate Analysis, Online Social Networking Sites (OSN), Principal Component Analysis, Social Networking Sites, Structural Equation Modeling (SEM), Student Empowerment, Uses and Gratification Theory (U&G)

Full Text

   

S. Anantha Sivaprakasam ¹, E. R. Naganathan ²

Affiliations
1 Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, IN
2 Department of Computer Science and Engineering, Hindustan University, IN

Abstract

Cervical cancer that is a disease, in which malignant (cancer) cells form in the tissues of the cervix, is one of the fourth leading causes of cancer death in female community worldwide. The cervical cancer can be prevented and/or cured if it is diagnosed in the pre-cancerous lesion stage or earlier. A common physical examination technique widely used in the screening is called Papanicolaou test or Pap test which is used to detect the abnormality of the cell. Due to intricacy of the cell nature, automating of this procedure is still a herculean task for the pathologist. This paper addresses solution for the challenges in terms of a simple and novel method to segment and classify the cervical cell automatically. The primary step of this procedure is pre-processing in which de-nosing, de-correlation operation and segregation of colour components are carried out, Then, two new techniques called Morphological and Statistical Edge based segmentation and Morphological and Statistical Region Based segmentation Techniques- put forward in this paper, and that are applied on the each component of image to segment the nuclei from cervical image. Finally, all segmented colour components are combined together to make a final segmentation result. After extracting the nuclei, the morphological features are extracted from the nuclei. The performance of two techniques mentioned above outperformed than standard segmentation techniques. Besides, Morphological and Statistical Edge based segmentation is outperformed than Morphological and Statistical Region based Segmentation. Finally, the nuclei are classified based on the morphological value. The segmentation accuracy is echoed in classification accuracy. The overall segmentation accuracy is 97%.

Keywords

Cervical Cancer Cell, PAP Smear Test, Segmentation, Classification, Morphological And Statistical Edge Based Segmentation, Morphological and Statistical Region Based Segmentation.

Full Text