Pit Pichappan

Improving arabic text categorization using decision trees

This paper presents the results of classifying Arabic text documents using a decision tree algorithm. Experiments are performed over two self collected data corpus and the results show that the suggested hybrid approach of Document Frequency Thresholding using an embedded information gain criterion of the decision tree algorithm is the preferable feature selection criterion. The study concluded that the effectiveness of the improved classifier is very good and gives generalization accuracy about 0.93 for the scientific corpus and 0.91 for the literary corpus and we also conclude that the effectiveness of the decision tree classifier was increased as we increase the training size, and the nature of the corpus has such a influence on the classifier performance.

Debunking the designs of contemporary nature-inspired computing algorithms: from moving particles to roaming elephants

Nature-inspired computing algorithms (NICs in short) inherit a certain length of history tracing back to Genetic Algorithm and Evolutionary Computing in the 50s. Since February 2008 by the birth of Firefly Algorithm, NICs started to receive lots of attentions from researchers around the global. Variants and even new species of NIC algorithms boomed like sprouts after rain. While it may be disputable to question the necessity of creations of further new species of such algorithms, it is observed that these algorithms are fundamentally made up of several core components. By exposing these components, the underlying design of a collection of so-called modern NIC optimization algorithms is revealed. In this paper, these core components of constructs from some popular NIC algorithms are reviewed, thereby debunking the myths of novelty and perhaps the excitement of inventing something really New by simply branding a NIC search method by merely a name of another living creature. This work also serves as a general guideline and reference for any algorithm architect who wants to create a new NIC algorithm in the future.

Development and investigation of a new compression technique using Boolean minimizations

This paper suggests a new algorithm for data compression based on Boolean minimization of binary data. On the compressor side, the input bit-stream is divided into blocks of N-bits each, and a ℌsum of products” function is found for each block using Quine-McCluskey algorithm.

Reshaping email relationships

This paper suggests splitting the existing email handlers into more than one group. The existing group can be considered as a group that has one-to-one relationship. However, this paper suggests that the email handlers should give the user the choice to select the relation that he would like to implement in the email handler. Three types of relationships can be identified. They are: one-to-one, one-to-many, and many-to-many relationship. Enforcing a certain type of relationship saves the time of the emailer. The paper will describe the prototype for each relationship with the advantages of each one.

Welcome message from the editors

Welcome to the Seventh International Conference on Digital Information Management (ICDIM 2012). This year at Macau, after the previous editions in Bangalore (2006-India), Lyon (2007-France), London (2008-UK), Michigan (2009- USA), Lakehead (2010-Canada) and Melbourne (2011- Australia), we are pleased to host here at University of Macau that, in an impressive display of the international diversity of this conference, brings together participants from 4 different continents: America, Asia, Africa and Europe, and over 28 countries.

Generalized confidence interval for R = P(X > Y) of Pareto distribution and its application in Web performance

In this paper we present a method useful for the system engineer to improve the service performance of a Web server through session-based Web workload, the best indicator of the users perception of the Web quality. Bytes transferred per session is one of the characteristics of intra-session which collectively describe session-based Web workload. This characteristic exhibits heavy-tailed behavior and its distribution match well with the Pareto Type I distribution [Goseva-Popstojanova et al. (2006)]. So for the performance study, we estimate the probability, R = P(X > Y), when X and Y are two independent but not identically distributed random variables following Pareto Type I distribution, using the maximum likelihood estimator. Extensive simulation studies are carried out to study the performance of the estimator. A generalized two-sided confidence interval for R of the Pareto type I distribution is constructed. The derived confidence interval suits both small samples and large samples. The average width and the coverage probability of this confidence interval is compared with the usual asymptotic confidence interval through simulations. Using real data, we illustrate how R and generalized confidence interval of R can be used for improving the service performance of a Web server.

Transaction clustering of web log data files using genetic algorithm

Increasingly web applications found to impact on numerous environments. The web log data offer more promises and particularly application of the genetic algorithms is significant as it represents the relations between different data components. We have used simple genetic algorithms to log files and we found that the preliminary results are more promising there by open more avenues for future research.