omar al jadaan

Genetic algorithms are becoming increasingly valuable in solving large-scale, realistic, difficult problems, and new customer personalization is one of these problems. In this paper, a method combining GA based clustering algorithm with Collaborative Filtering CF-based Recommender system is proposed named Information Gain Clustering using Genetic Algorithm (IGCGA), which alleviates the problem of being trapped in local clustering centroids using k-mean. Simulation results show that the proposed IGCGA, in most of the cases, is able to find much accurate personalization of new users compared to IGCN other Collaborative Filtering based Recommender system. Much better performance of IGCGA is observed.

Information Gain Clustering through Roulette Wheel Genetic Algorithm (IGCRWGA) is a novel heuristic used in Recommender System (RS) for solving personalization problems. In a bid to generate information on the behavior and effects of Roulette Wheel Genetic Algorithm (RWGA) in Recommender System (RS) used in personalization of cold start problem, IGCRWGA is developed and experimented upon in this work / paper. A comparison with other heuristics for personalization of cold start problem - such as Information Gain Clustering Neighbor through Bisecting K-Mean Algorithm (IGCN), Information Gain Clustering through Genetic Algorithm (GCEGA), among others -- showed that IGCRWGA produced the best recommendation for large recommendation size (i.e. greater than 30 items) since it is associated with the least Mean Absolute Error (MAE), the evaluation metric used in this work.

Computational grids have become attractive and promising platforms for solving large-scale high-performance applications of multi-institutional interest. However, the management of resources and computational tasks is a critical and complex undertaking as these resources and tasks are geographically distributed and are heterogeneous in nature. This paper presents a new stochastic approach for scheduling independent tasks in the grid environment by minimizing Makespan. The novel algorithm, Mutation Based Simulated Annealing Algorithm (MSA), speeds up convergence better than the previous algorithms by using the selection of Simulated Annealing, single change Mutation and a new Random Minimum Completion Time (Random-MCT) heuristic. In order to make the algorithm MSA working fast, it maintains two solutions at a time. The experiments on the algorithm MSA provide a reduction in Makespan equals to eighteen (18) when it is compared with algorithm Min-Min, and equals to three (3) when it is compared with previous genetic algorithm. The simulation results display that the assumed algorithm has better performance than previous genetic algorithm and Min-Min algorithm in terms of quality of solution and Load Balancing, as well as Resource Utilization. However, in this work the gain in average time consumed by algorithm is about 93%, which makes MSA algorithm very high QoS and more preferable for realistic scheduling in Grid environment.

IGCEGA, an acronym for Information Gain Clustering through Elitizt Genetic Algorithm, is a novel heuristic used in Recommender System (RS) for solving personalization problems. In comparison with IGCGA (Information Gain Clustering through Genetic Algorithm), IGCEGA is not associated with the inherent problem of increasing the possibility of losing good solution during the crossover phase, which translates into increasing the guarantee of converging to a global minima and consequently, enhancing the accuracy of the recommendation. Besides, IGCEGA using the technique of global minima still resolves the problem associated with IGCN (Information Gain through Clustered Neighbor), which traps the algorithm in local clustering centroids. Although this problem was alleviated by IGCGA, IGCEGA solves the problem even better because IGCEGA assumes the lowest Mean Absolute Error (MAE), the evaluation matrix used in this work. Results of the experimentation of the various heuristics / techniques in RS used in personalization for cold start problems -- for instance Popularity, Entropy, IGCN, IGCGA - showed that IGCEGA is associated with the lowest MAE, therefore, best clustering, which in turn results into best recommendation.

A criticism of evolutionary algorithms might be the lack of efficient and robust generic methods to handle constraints. The most widespread approach for constrained search problems is to use penalty methods, because of their simplicity and ease of implementation. Nonetheless, the most difficult aspect of the penalty function approach is to find an appropriate penalty parameters. In this paper, a method combining the new non-dominated ranked genetic algorithm (NRGA), with a parameterless penalty approach are exploited to devise the search to find Pareto optimal set of solutions. The new parameterless penalty and the nondominated ranked genetic algorithm (PP-NRGA) continuously find better Pareto optimal set of solutions. This new algorithm have been evaluated by solving four test problems, reported in the multi-objective evolutionary algorithm (MOEA) literature. Performance comparisons based on quantitative metrics for accuracy, coverage, and spread are presented.

A criticism of evolutionary algorithms might be the lack of efficient and robust generic methods to handle constraints. The most widespread approach for constrained search problems is to use penalty methods, because of their simplicity and ease of implementation. The penalty function approach is generic and applicable to any type of constraint (linear or nonlinear). Nonetheless, the most difficult aspect of the penalty function approach is to find an appropriate penalty parameters needed to guide the search towards the constrained optimum. In this paper, GA's population-based approach and Ranks are exploited to devise a penalty function approach that does not require any penalty parameter called Adaptive GA-RRWS. Adaptive penalty parameters assignment among feasible and infeasible solutions are made with a view to provide a search direction towards the feasible region. rank-based roulette wheel selection operator (RRWS) is used. The new adaptive penalty and rank-based roulette wheel selection operator allow GA's to continuously find better feasible solutions, gradually leading the search near the true optimum solution. GAs with this constraint handling approach have been tested on five problems commonly used in the literature. In all cases, the proposed approach has been able to repeatedly find solutions closer to the true optimum solution than that reported earlier.