Title:Military Dog Based Optimizer and its Application to Fake Review

Abstract:Over the last three decades more then sixty meta-heuristic algorithms have been proposed by the various authors. Such algorithms are inspired from physical phenomena, animal behavior or evolutionary concepts. These algorithms have been widely used for solving the various real world optimization problems. Researchers are continuously working to improve the existing algorithms and also proposing new algorithms that are giving competitive results as compared to the existing algorithms present in the literature. In this paper a novel meta heuristic algorithm based on military dogs squad is introduced. The proposed algorithm mimics the searching capability of the trained military dogs. Military dogs have strong smell senses by which they are able to search the suspicious objects like bombs, wildlife scats, currency, or blood as well as they can communicate with each other by their barking. The performance of the proposed algorithm is tested on 17 benchmark functions and compared with five other meta-heuristics namely particle swarm optimization (PSO), multiverse optimizer (MVO), genetic algorithm (GA), probability based learning (PBIL) and evolutionary strategy (ES). The results are validated in terms of mean and standard deviation of the fitness value. The convergence behavior and consistency of the results have been also validated by plotting convergence graphs and BoxPlots. Further the, proposed algorithm is successfully utilized to solve the real world fake review detection problem. The experimental results demonstrate that the proposed algorithm outperforms the other considered algorithms on the majority of performance parameters.