IJMSC Vol. 5, No. 4, Nov. 2019
Cover page and Table of Contents: PDF (size: 244KB)
The advent of different social networking sites has enabled people to easily connect all over the world and share their interests. However, Social Networking Sites are providing opportunities for cyber bullying activities that poses significant threat to physical and mental health of the victims. Social media platforms like Facebook, Twitter, Instagram etc. are vulnerable to cyber bullying and incidents like these are very common now-a-days. A large number of victims may be saved from the impacts of cyber bullying if it can be detected and the criminals are identified. In this work, a machine learning based approach is proposed to detect cyber bullying activities from social network data. Multinomial Naïve Bayes classifier is used to classify the type of bullying. With training, the algorithm classifies cyber bullying as- Shaming, Sexual harassment and Racism. Experimental results show that the accuracy of the classifier for considered data set is 88.76%. Fuzzy rule sets are designed as well to specify the strength of different types of bullying.[...] Read more.
We choose a better pseudo-random number generator from a list of eight pseudo-random number generators derived from the library function rand() in C/C++, including rand(); i.e. a random number generator which is more random than all the others in the list. rand() is a repeatable pseudo-random number generator. It is called pseudo because it uses a specific formulae to generate random numbers, i.e. to speak the numbers generated are not truly random in strict literal sense. There are available several tests of randomness, some are easy to pass and others are difficult to pass. However we do not subject the eight set of pseudo random numbers we generate in this work to any known tests of randomness available in literature. We use statistical technique to compare these eight set of random numbers. The statistical technique used is correlation coefficient.[...] Read more.
As with any ANOVA, a repeated measure ANOVA tests the equality of means. However, a repeated measure ANOVA is used when all members of a random sample are measured under a number of different conditions. As the sample is exposed to each condition in turn, the measurement of the dependent variable is repeated. Using a standard ANOVA in this case is not appropriate because it fails to model the correlation between the repeated measures: the data violate the ANOVA assumption of independence. Some ANOVA designs combine repeated measures factors and independent group factors. These types of designs are called mixed-model ANOVA and they have a split plot structure since they involve a mixture of one between-groups factor and one within-subjects factor.
The work present an application of the mixed model factorial ANOVA, using scores obtained by 120 secondary school students in mathematics. The between group factor is the different categories of students (science, commercial humanities) with three levels while the within group factor is the three years spent in senior secondary School.[...] Read more.
Numerical integration compromises a broad family of algorithm for calculating the numerical value of a definite integral. Since some of the integration cannot be solved analytically, numerical integration is the most popular way to obtain the solution. Many different methods are applied and used in an attempt to solve numerical integration for unequal data space. Trapezoidal and Simpson’s rule are widely used to solve numerical integration problems. Our paper mainly concentrates on identifying the method which provides more accurate result. In order to accomplish the exactness we use some numerical examples and find their solutions. Then we compare them with the analytical result and calculate their corresponding error. The minimum error represents the best method. The numerical solutions are in good agreement with the exact result and get a higher accuracy in the solutions.[...] Read more.
The aim of this article is to propose a novel and simple technique for solving bi-matrix games with rough intervals payoffs. Since the payoffs of the rough bi-matrix games are rough intervals, then its value is also a rough interval. In this technique, we derived four bilinear programming problems, which are used to obtain the upper lower bound, lower lower bound, lower upper bound and upper upper bound of the rough interval values of the players in rough bi-matrix games which we called in this article as 'solution space'. Moreover, the expected value operator and trust measure of rough interval have been used to find the α-trust equilibrium strategies and the expected equilibrium strategies of rough interval bi-matrix games. Finally, numerical example of tourism planning management model is presented to illustrate the methodologies adopted and solution procedure.[...] Read more.