Service Oriented Architecture Quality Model for Software Security[Full-Text ]
J.Avesh Gopal, DR.P.G.V Suresh Kumar, Mohammed Kemal, M Sajeeva Reddy, Nune Sreenivas

The paper presents an approach to locating security aspects in the Service Lifecycle and Service Oriented Architecture (SOA) quality model. The first part of the paper focuses on the quality of SOA and security measures and investigates some functional and non-functional requirements for security measurement. The general discussion about SOA quality and security measures have been summarized by the proposition of the multi-agent architecture for SOA systems security level evaluation in the second part of the paper.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Rule Induction using Ant-Miner Algorithm[Full-Text ]
NimmyCleetus, Dhanya K.A

In this paper, we extracted robust rules for identifying different forms of network attacks. Initially, we mined set of rules using the data mining rule such as WEKA using conjunction, JRIP, NNge, OneR, part rules. It observed that the rule created with WEKA was not optimal as indicated with classification accuracy. We synthesized prominent rules by eliminating important ones. Swarm intelligence (SI) is a technique where the rules are discovered through the study of collective behaviour in decentralized, self-organized systems such as ants. Ant-miner is a rule induction algorithm that uses SI techniques to form rules. Using our customized rule synthesis parse, Later these synthesized rules where optimal using Ant-Miner. Ant-miner results in better synthesized WEKA generated rules.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Multiclass Emotion Extraction from Sentences[Full-Text ]
Bincy Thomas, Vinod P, Dhanya K A

This paper aims to investigate the extraction of different classes of emotion from sentences using supervised machine learning technique, Multinomial Naïve Bayes (MNB). Here a bag of word approach is used to capture the emotions. The unigrams are mainly used for this and the bigrams and trigrams are used to capture lower order dependencies. The work is done on the ISEAR dataset [14]. The experiments with different feature sets selected using Weighted log-likelihood score (WLLS) [12] shows that the MNB classifier provides good results when the unigram feature set size is 450 which provides an average accuracy of 76.96% across all emotion classes.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

A Study on Applications of Fuzzy Set Theory in Datamining[Full-Text ]
Santhosh Kottam, Dr. Varghese Paul

Data Mining Researchers are working for scalable and efficient data analysis algorithms apt for data mining functionalities. Data mining is an interdisciplinary area and it does include database systems, statistics, machine learning, visualization and information science. Data uncertainty and incomplete information raise challenges in Data mining. New mathematical approaches- Fuzzy Set, Rough Set and Soft Set-have wide applications in dealing uncertain data. This paper is a study on applications of the Fuzzy set theories in the domain of Data mining. We first discuss an introduction to Crisp set, Fuzzy set and Uncertain Data mining. In addition to these topics our paper is covering applications of Fuzzy sets in Association Analysis and Clustering. Finally this paper gives a brief on advantages of Fuzzy set theory.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

New Searching Techniques for Ontology Based Search Engines – A Conceptual Study[Full-Text ]
Sareena Rose, Asst.Professor, Department of Computer Science, Vimala College, Thrissur

Search engines play a vital role in finding the relevant information across the web and make it available on the finger tips. But they seemed to be less efficient to understand the relationship between the keywords which had an adverse effect on the results it produced. Semantic search engines – only solution to this; is still an unrealized dream due to various underlying issues. The hindrances faced by semantic search engines such as annotating the documents spread across network, quality of annotation, incompatibility between the query languages stumbles the growth of semantic search engines. Reducing the time taken to search the semantic annotated documents is a highly demanded solution for these search engines. This paper focuses on a study and improvisation of searching techniques used in semantic search engines keeping time complexity as the major factor.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Cluster Computing of Nucleotide Sequence by Fractal Analysis[Full-Text ]
Deepa Mary Mathews, Dr.K.S.M.Panicker

The nucleotide sequence has capacity to represent information. Biological DNA represents the information which directs the functions of a living thing. Sequences can be read from the biological raw material through DNA sequencing methods. Cluster computing for nucleotide sequence by fractal analysis can be useful for identifying certain diseases. Very long sequences of nucleotide were checked with parallel search methods of cluster system. This paper uses box counting algorithm for fractal analysis. The implementation of fractal analysis in cluster system is done by parallel computing load balancing method. Finally performance of fractal analysis is tested. This paper, proposing new enhancement that can be included for nucleotide sequence analysis application programme by implementing MPI communication on clusters.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------