Improved AFSA for Solving Intelligent Test Problem

Test paper generation is a multi-objective parameter optimization problem under multi-constraints. Aiming at the defects of test paper quality and speed in intelligent test paper generation, the genetic algorithm is improved on the basis of the original genetic algorithm. From the coding method of the test questions, the establishment of weighted objective function to optimize the fitness function, the selection of crossover operator, the setting of mutation, and the combination of optimization strategy and roulette to select the operation, the optimization design is carried out. It achieves good adaptability and can generate test paper quickly and successfully, which improves the efficiency of generating test paper. The simulation results show that, under the same constraints, the algorithm can generate test papers successfully, and improve the success rate and efficiency of generating test papers, and effectively optimize the problems in automatic generating test papers.

With the wide application of online examinations in colleges and universities, intelligent test paper generation technology has become a hot topic. In this paper, we propose an improved genetic simulated annealing algorithm (GSA) and apply it to the test paper generation process to improve the quality and speed of intelligent test paper generation. Firstly, we optimize the genetic operators in real number coding and introduce an adaptive adjustment strategy on the premise of maintaining individual diversity. Secondly, we apply simulated annealing to individuals of each generation. Finally, we conduct abundant experiments. The results show that the improved algorithm can promote the success rate of test paper generation and effectively enhance the quality of test paper.

In order to solve the problems such as blindfold search, slower convergence, and sometimes unsuccessfully search in the present genetic algorithms used for intelligent test paper generation, this paper introduces an improved genetic algorithm for intelligent test paper generation. This algorithm generates optimized initial chromosome group and controlling crossing and variation recurring to test paper parameters and randomization. Consequently, it can avoid the problems of blind search and slower convergence resulted by complete randomization controlling adopted by the present genetic algorithms. At last, this paper proves the improved genetic algorithm can better solve the intelligent test paper generation problem compared to present genetic algorithms mentioned in algorithm analysis section of this paper.

An improved quantum artificial fish swarm algorithm is proposed in this paper. Based on that quantum compu- ting have exponential acceleration for heuristic algorithm, by examining eight most recent patents and some literatures in the area of artificial fish swarm algorithm and quantum computing. The new algorithm uses qubits to code artificial fish and quantum revolving gate, preying behavior, following behavior and variation of quantum artificial fish to update the artificial fish for searching for optimal value. Then, apply the new algorithm, the basic artificial fish swarm algorithm and the global edition artificial fish swarm algorithm to the simulation experiment high dimensional and complex non-convex programming respectively. The simulation results show that the improved algorithm can escape from the local extremum effectively, and has higher convergence speed and precision.

In order to deal with the problems of inefficient searching and low accuracy of Artificial Fish Swarm Algorithm(AFSA) for multimodal function optimization,an improved AFSA for multimodal function optimization was proposed.In the algorithm,the strategy of the survival of the fitter suppression was adopted,eliminating artificial fish which was situated in food with low concentration of similar artificial fish to select elite artificial fish swarm.Optimization for swarming behavior and following behavior contributed to artificial fish careful search in a new optimization trajectory to enhance its local search capacity.Modifying for preying behavior,artificial fish was avoided sinking flat position.In combination with Pattern Search Method(PSM),its local accuracy search capacity was enhanced.The simulation results indicate that the proposed algorithm has stronger global optimization and local optimization capabilities,and the search for each optimal solution accuracy has reached the ideal value,and it is able to be used for complex multimodal function optimization.

Intelligent test paper generation belongs to an important research topic in the field of computer-aided education. Among them, Genetic Algorithm (GA), as an algorithm with high efficiency and performance, is widely used in the intelligent test paper generating system. But generally speaking, traditional GA always has some problems, such as local optimal solution and prematurity. The traditional genetic algorithm is improved in the aspects of the initial population, fitness function and some genetic operators to make it more in line with the requirements of intelligent paper generation, improving the efficiency and success ratio.

Based on a newly improved artificial fish swarm algorithm– local neighborhood artificial fish swarm algorithm (LNAFSA), we studied its application for adaptive polarization mode dispersion (PMD) compensation. In the adopted local neighborhood structure, each artificial fish only communicated with other five neighboring artificial fishes. Under a simulated 40 Gb/s adaptive second-order PMD compensation for non-return- to-zero and carrier-suppressed return-to-zero formats, the study demonstrated the effectiveness of LNAFSA. In a defined searching process, the eye diagrams showed that the algorithm resulted in an effective compensation effect and produced the maximum consuming time within 300 ms. Much better performance than simplex algorithm and genetic algorithm was also observed. In tracking process, LNAFSA achieved the goal of compensation in about 14 ms.

In order to comprehensively measure these two indicators and make reasonable portfolio investment decisions, the author proposes using swarm intelligence optimization algorithm - artificial fish swarm algorithm to solve multi-objective investment portfolio problems, and has achieved good results. In order to verify the effectiveness and superiority of the artificial fish school algorithm, the author used MATLAB programming to conduct simulation experiments using AFSA algorithm and genetic algorithm (GA), and compared the results obtained. The results show that compared to the GA algorithm, the artificial fish school algorithm can obtain better investment portfolio decision-making solutions for investing in five types of assets, making investment returns as large as possible while minimizing risks, indicating the efficiency and superiority of the algorithm in solving multi-objective investment portfolio problems.

Taking the radial basis function as a kernel function, a prediction model is developed based on the support vector regression machine (SVR). The optimization of the model parameters, including penalty factor and kernel function variance, is realized by the artificial fish swarm algorithm. The model is used to predict nine foreign exchange rate data with updating and rolling. At the same time, simulating by the cross validation, genetic algorithm, particle swarm optimization algorithm and then evaluating the results from the total error (TE), relative error (RE), absolute root mean square error (ARMSE) and correct trend rate (CTR) comprehensively, the comparison shows that the errors of the model based on SVR optimized by artificial fish swarm algorithm are all minimum and CTR are maximum. In the end, in order to improve the convergence speed and precision further, the self-adaption artificial fish swarm algorithm is presented which is joined the attenuation factor and based on the average distance visual. The result is ideal. Therefore, SVR optimized by the improved artificial fish swarm algorithm can be effectively used in forex prediction.

The discovery of high utility itemsets (HUIs) is an attractive topic in data mining. Because of its high computational cost, using heuristic methods is a promising approach to rapidly discovering sufficient HUIs. The artificial fish swarm algorithm is a heuristic method with many applications. Except the current position, artificial fish do not record additional previous information, as other related methods do. This is consistent with the HUI mining problem: that the results are not always distributed around a few extreme points. Thus, we study HUI mining from the perspective of the artificial fish swarm algorithm, and propose an HUI mining algorithm called HUIM-AF. We model the HUI mining problem with three behaviors of artificial fish: follow, swarm, and prey. We explain the HUIM-AF algorithm and compare it with two related algorithms on four publicly available datasets. The experimental results show that HUIM-AF can discover more HUIs than the existing algorithms, with comparable efficiency.

One of the most difficult combinatorial optimization problems in recent studies is job shop scheduling. Job shop scheduling which also holds the key to the company's profitability is a crucial problem faced by many manufacturing companies. Well-structured scheduling has the potential to reduce operating costs and increase profits. Artificial Fish Swarm Algorithm (AFSA) is one of optimization algorithms to solve combinatorial problems. This paper talks about the implementation of AFSA in job shop scheduling cases to produce an optimal solution, containing a minimum completion total time (makespan) of the entire job. The results showed that the AFSA which is designed for job shop scheduling problem optimization is able to provide solutions with the best efficiency value ever achieved was 75%. This figure is still considered unsatisfactory based on the makespan resulted. Nevertheless, the AFSA ability in the searching process for solutions is quite good considering that level of efficiency is achieved by only 10000 artificial fishes around 100 generations within 3,72e+41 solution spaces.

Geomagnetic matching (GM) is a new developmental technology in recent years. For resolving the cumulate errors of position, velocity, attitude in an inertial navigation system (INS), utilizing artificial fish swarm algorithm (AFSA) to carry on GM is proposed, and then the attained matching position regarded as measurement of filter to achieve the emendation of INS. Firstly, affine transformation model between inertial and real trajectories is displayed and the principle of INS/GM is explained. Secondly, the state of artificial fish (AF), distance and food consistence are defined afresh, and then the flow chart based on AFSA is given. Lastly, simulation analysis is performed in actual geomagnetic reference map (GRM). The results show that the algorithm actualizes the GM when the inertial system exist position, velocity and attitude error, that is the algorithm got the optimized global solution, with that, the data fusing is finished with the results that not only reduce error evidently but also verify the validity and feasible of the algorithm.

Conventional mathematics methods are extensively adopted in the most test paper generation systems, but these methods can't deal with multi-objective optimization question of more than one constrains effectively. This paper proposes an improved parallel GA(Genetic Algorithm) with good adaptability and universal applicability, which is used in test paper generation. The experiment results indicate that the algorithm is more adaptively and faster than conventional parallel GA, and improve the intelligent level of the test paper generation system effectively.

An improved artificial fish swarm algorithm (AFSA) for solving a combinatorial optimization problem— a berth allocation problem (BAP), which was formulated. Its objective is to minimize the turnaround time of vessels at container terminals so as to improve operation efficiency customer satisfaction. An adaptive artificial fish swarm algorithm was proposed to solve it. Firstly, the basic principle and the algorithm design of the AFSA were introduced. Then, for a test case, computational experiments explored the effect of algorithm parameters on the convergence of the algorithm. Experimental results verified the validity and feasibility of the proposed algorithm with rational parameters, and show that the algorithm has better convergence performance than genetic algorithm (GA) and ant colony optimization (ACO).

The scalability of traditional clustering algorithm is not strong. Its capacity of processing isolated points is also weak. Artificial fish swarm algorithm (AFSA) is an algorithm for global optimization based on animal behavior. It is used in web users clustering; imitating fish feeding, clustering, pileup and random acts to construct artificial fish. Throu gh the local optimization of each individual fish, wefind the global optimal value, and thusget reasonable clustering for web access users. The actual results verify that the algorithm is effective.