Tan Tse Guan

A comparison on the performance of crossover techniques in video game

This paper describes the performance of four crossover operators used in evolving the required controllers in a video game. The crossover operators used in this research are the two-point crossover, the uniform crossover, the N-point crossover, and the single-point crossover. The performance of these crossover methods were tested using Infinite Mario Bros game. This video game was chosen due to the dynamicity and complexity of the game. This paper also presents a newly designed nondeterministic based Finite State Machine (FSM) method. The Mario character uses the proposed FSM as its strategy in the game. The proposed FSM is then optimized using a modified Genetic Algorithm (GA). The results showed that the required controllers were generated successfully using the proposed method. The results also showed that the N-point crossover performed well compared to the uniform crossover, the two-point crossover and the single-point crossover methods.

Self-synthesized controllers for tower defense game using genetic programming

In this paper, we describe the results of implementing Genetic Programming (GP) using two different Artificial Neural Networks (ANN) topologies in a customized Tower Defense (TD) games. The ANNs used are (1) Feed-forward Neural Network (FFNN) and (2) Elman-Recurrent Neural Network (ERNN). TD game is one of the strategy game genres. Players are required to build towers in order to prevent the creeps from reaching their bases. Lives will be deducted if any creeps manage to reach the base. In this research, a map will be designed. The AI method used will self-synthesize and analyze the level of difficulty of the designed map. The GP acts as a tuner of the weights in ANNs. The ANNs will act as players to block the creeps from reaching the base. The map will then be evaluated by the ANNs in the testing phase. Our findings showed that GP works well with ERNN compared to GP with FFNN.

Cloud-Based Mobile-Retail Application for Textile Cyberpreneurs: Task-Technology Fit Perspective Analysis

Cloud computing and mobile computing paradigms have enhanced the usage of current technology among people from various sectors to perform certain required tasks. The combination of these two paradigms have also created the existence of cloud-based mobile applications that are designed to be useful in business environments, such as to be used by textile cyberpreneurs for m-retail transaction. As cloud adoption was previously low among Malaysian entrepreneurs, the usage intention factors of related cloud technology and services should be determined for better clarification and future technology enhancements. Besides, studies on m-retail or m-shopping adoption in Malaysia were merely focused on customers perspectives rather than retailers perspectives. In measuring the suitability of Cloud-based M-Retail (CBMR) application with the task requirements of textile cyberpreneurs, Task-Technology Fit (TTF) model is used as the basis of this research. Results from a pilot study, through a survey of selected group of textile cyberpreneurs shows instruments reliability and positive feedbacks to support the intention to use the technology. The future direction of the study in which to apply the instruments to a larger group of respondents is also discussed along with its potential contribution.

A Comparison of BPNN, RBF, and ENN in Number Plate Recognition

In this paper, we discuss a research project that related to autonomous recognition of Malaysia car plates using neural network approaches. This research aims to compare the proposed conventional Backpropagation Feed Forward Neural Network (BPNN), Radial Basis Function Network (RBF), and Ensemble Neural Network (ENN). There are numerous research articles discussed the performances of BPNN and RFB in various applications. Interestingly, there is lack of discussion and application of ENN approach as the idea of ENN is still very young. Furthermore, this paper also discusses a novel technique used to localize car plate automatically without labelling them or matching their positions with template. The proposed method could solve most of the localization challenges. The experimental results show the proposed technique could automatically localize most of the car plate. The testing results show that the proposed ENN performed better than the BPNN and RBF. Furthermore, the proposed RBF performed better than BPNN.