Informatics Publishing Limited

A. Nisha Jebaseeli ¹, E. Kirubakaran ²

Affiliations
1 Department of Computer Science, Bharathidasan University Constituent College, Lalgudi, Trichy, IN
2 Bharat Heavy Electrical Ltd.(BHEL), Trichy, IN

Abstract

The development of communication technology has resulted in easy information access via internet. The rapid increase in the use of mobile devices has popularized pedagogical methods like learning through mobile devices, PDAs, etc. Various Mobile Learning (M-Learning) systems are available and the user opinions about them are aired in social blogs or review websites. This research paper investigates Opinion mining classifications particularly of M-Learning system based not only on words but also on the corpus from the reviewed documents. A preprocessing methodology is proposed in this paper to enhance classifications in the dataset under study. The corpus is ranked using SVD through which, the data is prepared for Opinion mining. The classification accuracy is evaluated through Naïve Bayes, Random Forest, k Nearest Neighbor (kNN) data mining algorithms and Learning Vector Quantization (LVQ), Elman Neural Network, Feed Forward Neural Network (FFNN) algorithms with the preprocessed dataset.

Keywords

Classification Accuracy, Machine Learning, M-Learning, Opinion Mining, Preprocessing.

Full Text

     

A. Adhiselvam ¹, E. Kirubakaran ², R. Sukumar ³

Affiliations
1 Department of MCA, S.T.E.T. Women’s College, Mannargudi, Tiruvarur - 614 016, Tamil Nadu, IN
2 STTP (System), Bharat Heavy Electrical Limited, Trichy - 620014, Tamil Nadu, IN
3 Department of CSE, Sethu Institute of Technology, Kariapatti, Virudhunagar - 626115, Tamil Nadu, IN

Abstract

Background/Objectives: Software bugs are generally the faults or errors that occur in the code that may leads to incorrect results. It is therefore necessary to find software bugs to increase the quality of software and for making the software to meet the user requirements. Methods/Statistical Analysis: The testing enables assessment of software which ensures the system whether it meets the system requirements. Graph mining is an approach to find software bugs and furthermore testing involves more computational complexity and cost therefore tools are developed for testing the software. The challenging task is to locate and fix bugs automatically. Bug localization using high performance map reducing process is very successful in the recent research. Findings: This proposed technique called Map Reduce Technique Based Apriori (MRTBA) Algorithm based on Graph mining in where edge weights are used to takes the program as the input and computes method calls of the program. Here each node represents method and each edge represents a method call. The algorithm aims for detection of faulty nodes. It uses Hash Map for reducing edge weights. JUnit test cases are performed at last for detecting bugs by giving different inputs. The approach mines all nodes not only at same level but also the nodes present at different levels. Conclusion/Improvements: In this work MRTBA, a dynamic control flow centered approach for bug localization has been presented. In future this technique can be extended with the proposal of steps involves in connecting same nodes at different levels. It also further enhances to find out indirect recursions for both the trivial and non-trivial map reduction in software bug localization.

Keywords

Apriori, Call Graph, Graph Mining, Map Reduce, Software Bug Localization, Subgraph

Full Text

   

S. Padmapriya ¹, E. Kirubakaran ¹, N. M. Elango ²

Affiliations
1 Department of Information Technology, AVCCE, IN
2 Madanapalle Institute of Technology and Science, IN

Abstract

Background/Objective: To create a Computer Aided Diagnosis system to detect the abnormalities in the human tissue images by extending the attributes. Methods/Statistical Analysis: An efficient and hybrid classifier using "K-Ratio Super Item set Finding-Nearest Neighborhood Classifier (KRSIF-NNC) Algorithm" is proposed. It classifies the tumor cells in an effective manner by adopting extended attributes from small datasets. The glioblastoma and lung cancer tissue image samples are keyed in to the algorithm which classifies them into four grades. Findings: From the histopathology (tissue) images the pathologists will be able to diagnose the abnormalities in the tissues. Examination and judgments are based on the pathologist's personal experience. The problem is during the manual diagnosis, there is a chance of missing some cancerous cells in the tissue images. This is solved by adopting the proposed classifier which automatically do the diagnostic process and classify it into proper grade. Thus the proposed classifier improves the classification process. Applications/Improvements: Improved classification is required to detect the cancer grades. This hybrid approach has better classification accuracy than other approaches with 4% improvement which is very essential.

Keywords

Extending Attributes, Hybrid Classifier, KRSIF-NNC, Tissue Images Naïve Bayes Classifier

Full Text

   

B. Lakshmi ¹, E. Kirubakaran ², T. N. Prabakar ³

Affiliations
1 Jayaram College of Engineering and Technology, Tiruchirappalli, IN
2 Bharat Heavy Electricals Ltd, Tiruchirappalli, IN
3 Saranathan College of Engineering, Tiruchirappalli, IN

Abstract

In this paper, a novel encryption scheme with time based key technique on an FPGA is presented. Time based key technique ensures right key to be entered at right time and hence, vulnerability of encryption through brute force attack is eliminated. Presently available encryption systems, suffer from Brute force attack and in such a case, the time taken for breaking a code depends on the system used for cryptanalysis. The proposed scheme provides an effective method in which the time is taken as the second dimension of the key so that the same system can defend against brute force attack more vigorously. In the proposed scheme, the key is rotated continuously and four bits are drawn from the key with their concatenated value representing the delay the system has to wait. This forms the time based key concept. Also the key based function selection from a pool of functions enhances the confusion and diffusion to defend against linear and differential attacks while the time factor inclusion makes the brute force attack nearly impossible. In the proposed scheme, the key scheduler is implemented on FPGA that generates the right key at right time intervals which is then connected to a NIOS - II processor (a virtual microcontroller which is brought out from Altera FPGA) that communicates with the keys to the personal computer through JTAG (Joint Test Action Group) communication and the computer is used to perform encryption (or decryption). In this case the FPGA serves as hardware key (dongle) for data encryption (or decryption).

Keywords

Encryption, Decryption, Real Time Systems, Time Based Key, Brute Force Attack, Cryptanalysis, FPGA.

Full Text

     

C. Beulah Christalin Latha ¹, Sujni Paul ¹, E. Kirubakaran ², Sathianarayanan ³

Affiliations
1 Department of Computer Applications, Karunya University, IN
2 Bharat Heavy Electricals Limited, Tiruchirapalli, IN
3 NIIT, Coimbatore, IN

Abstract

Weather forecasting is a crucial application in meteorology. Weather is a continuous, data-intensive, multidimensional, dynamic process that makes weather forecasting a formidable challenge. This paper proposes a novel method to develop a service oriented architecture for a weather information system and forecast weather using data mining techniques. Web services provide a way of using software as a service. Developing service oriented architecture enables this software to be used anywhere at anytime and on any device. Data mining is an interesting field of computer science that can be used for various applications. This proposal aims at developing a weather information system as a web service that can be used by any type of application and uses the prediction techniques of data mining for weather forecasting.

Keywords

Data Mining, Service Oriented Architecture, Weather Forecasting, Web Services.

Full Text