Design And Implementation Of Online Examination System Based On Improved Fish Swarm Algorithm

With the continuous development of computer,Internet and multimedia technology,internet plus has been widely used in different fields.In the field of education,the integration of computer technology and education provides a new direction for the current development of education.Especially after the "COVID" in 2020,the new mode of online classroom teaching has been widely developed in domestic universities and there is a trend of normalization.Online examination is an important means of online teaching activities detection.More and more researchers apply artificial fish swarm algorithm to online examination system to improve the efficiency of test paper generation,but there is still room for further improvement.In this paper,three improved basic artificial fish swarm algorithms are proposed,and the improved fish swarm algorithm is applied to the paper generation module of the online examination system,the main work contents are as follows:(1)An improved fish swarm algorithm with adaptive execution of behavior segmentation is proposed.The four behaviors in the basic fish swarm algorithm are tested one by one,amend to only two of them are tested at the same time,so as to eliminate the trial process of meaningless behavior and improve the convergence effect of fish swarm.Through the simulation study of the improved strategy and the comparison with the previous research results,it is verified that the improved algorithm not only reduces the convergence time,but also improves the convergence accuracy.(2)An improved fish swarm algorithm with segment reset is proposed.In the later stage of the basic artificial fish swarm algorithm,the reset idea was added,and the artificial fish was reset when the optimization was stagnant.Experiments show that the improved fish swarm algorithm with segment reset improves the convergence speed and convergence accuracy,and the convergence performance is better than other improved algorithms when the global extreme value is hidden and the local extreme value is prominent.The improved algorithm can jump out of the local extreme value and find the global optimal value in a short time.(3)An improved fish swarm algorithm based on particle swarm optimization is proposed.The improved algorithm integrates the speed and direction idea of particle swarm optimization algorithm into fish swarm algorithm.Each movement of artificial fish is determined by two parameters:individual extreme value and global extreme value,so as to avoid blind and invalid movement,so as to improve the convergence speed and convergence accuracy of the algorithm.(4)An online examination system based on improved fish swarm algorithm is designed and implemented.The system is used to solve the problems existing in the function of generating test paper,such as the quality of test paper does not meet the requirements of synthetic test paper,the time of generating test paper is too long,the success rate of generating test paper is low,and there is no good detection standard.After applying the improved fish swarm algorithm studied in this paper to the test paper generation module of the system,the test shows that the system can group more test papers that meet the requirements of synthetic test paper in a short time,and improve the quality of test paper generation.Among them,an improved fish swarm algorithm with adaptive execution of behavior segmentation is suitable for generating papers when the test question parameters in the question bank are evenly distributed,while an improved fish swarm algorithm with segmented reset has better effect when the test question parameters in the question bank are extremely distributed.The main functions of the system are:the teacher uploads the test questions and forms the test paper quickly and accurately,so as to effectively assess the examinee’s mastery of knowledge points;Students complete the online examination and view the examination results;The administrator manages users,subjects,question bank and grades.