IADIS-INTERNATIONAL JOURNAL ON COMPUTER SCIENCE AND INFORMATION SYSTEMS

1-13

Authors: Ali Ahmad Alawneh and Rashad Aouf

The undertaking of managing the Information systems projects has become a critical challenge for many information technology companies. This is due to the fact that these projects are complex with many diverse phases and activities; knowledge-intensive that depends heavily on the people's knowledge; and collaborative that involve many people working hand-by-hand for long periods of time. Therefore, there is a need for best utilizing the knowledge of all members of the project during all phases in order to get it done as needed. Actually it is the role of knowledge and its management as valuable resources for guaranteeing the success of these projects and avoid repeating past mistakes. This paper attempts to present a new paradigm that combines the knowledge management processes (discovery, capture, sharing, and application) with the phases of information systems project management (initiation, planning, execution, and closing-down). It is expected that the application of the new paradigm by IT companies might improve the success rate of IS projects, through enhancing the activities of combination, exchange, socialization and transfer of knowledge and experience among project members. This might also lead to developing an organizational memory of a knowledge repository that shall serve as a reference for best practices and lessons learned that might in turn support the decision making process for further projects.

26-44

Authors: Gunupudi Rajesh Kumar, N. Mangathayaru and Gugulothu Narsimha

This work discusses the approach for intrusion detection and classification by devising a membership function, inspired from Yung, Jung, & Shie-Jue (2014) and used in this work to carry the dimensionality reduction of processes present in the training set in evolutionary approach. The reduced process representation may then be used to perform classification and prediction for detecting intrusion. It is seen that the reduced representation of processes retains the system call distribution of the initial process. Experiment results show the proposed approach is better compared to existing approaches and helps in effective identification of U2R and R2L attacks.

61-74

Authors: Yelipe UshaRani, Anurag Yepuri and Porika Sammulal

Disease prediction and classification using medical record datasets is a challenging data mining research problem that also requires attribute value imputation to be carried implicitly. Medical datasets that are available in public databases are not free from missing values and this is also true when data is collected and sampled through various clinical trials. In this context, there is always a need to turn up with new approaches and methods for accurate and efficient analysis of medical records. Several imputation strategies are proposed in the literature and each of them have reported accuracies achieved on benchmark datasets. However, a better imputation approach always helps in improving classification accuracies and this in turn helps to more accurate disease prediction. An approach for imputing medical records is proposed in this paper. We name the approach as Class-Based-Clustering-Imputation (CBC-IM). Experiments are carried out on several benchmark datasets. Results achieved using our imputation approach is compared to existing imputation approaches using classifiers such as KNN, SVM and C4.5. The results show improved performance on most of the datasets and are almost nearer to remaining approaches discussed in this paper.

88-108

Authors: T. V. Rajinikanth and G. Suresh Reddy

Feature dimensionality has always been one of the key challenges in text mining as it increases complexity when mining documents with high dimensionality. High dimensionality introduces sparseness, noise, and boosts the computational and space complexities. Dimensionality reduction is usually addressed by implementing either feature reduction or feature selection techniques. In this work, the problem of dimensionality reduction is addressed using singular value decomposition and the results are compared to information gain approach through retaining top-k features. High dimensional clustering is carried by using k-means algorithm with gaussian function. The proposed dimensionality reduction and clustering approaches are compared to conventional approaches and results prove the importance of our approach.

126-142

Authors: Vangipuram Radhakrishna, P. V. Kumar, V. Janaki and Aravind Cheruvu

This research address the design of a new dissimilarity measure and applying it to find all valid similarity profiled patterns in a temporal database defined over finite number of time slots. The proposed dissimilarity measure is a function of the reference sequence, threshold and standard deviation. Given, a reference time sequence and allowable dissimilarity limit, unearthing all eccentric (similar) temporal association patterns requires a similarity or correlation measure that can estimate similar association patterns accurately, efficiently, and is computationally optimal. This research also proposes a method to estimate temporal pattern support bounds. The experiment result shows the advantage of our proposed measure and bound estimation approach and also proves that our method is computationally efficient when compared to naïve, sequential and Spamine approaches.