Research On Automatic Generation Of Test Data Based On Improved Flower Pollination Algorithm

The test data is the core factor in the software test,and the efficiency of the test data directly affects the effect of the software test.The main research object in this paper is the automatic generation method of the test data.As a new intelligent optimization algorithm with good searching ability,flower pollination algorithm has been successfully applied to various multiobjective optimization problems because of its simple parameters and easy implementation.This paper applies this algorithm to the automatic generation of test data.First proposed a series of improvement measures for its defects,and then through a large number of experiments to verify the superiority of the improved algorithm in the automatic generation of test data.The main contents and innovations of this paper include the following aspects:(1)For basic flower pollinate algorithm search slower,optimization accuracy is not high,and in the later period the defects of easy to fall into local extreme,from two direction of adjust algorithm parameters and introducing other intelligent algorithm for mixed to improve the basic flower pollinate algorithm.In this paper,an adaptive hybrid flower pollination algorithm is proposed.Firstly,take advantages of the high precision and fast convergence of particle swarm optimization in the early stages to obtain a set of better quality solutions as the initial solution of flower pollination algorithm.Secondly,a guard function is presented to reflect the discrepancy of the population.Finally,an adaptive mechanism is proposed to update the solution.The adaptive mechanism consists of two parts: adaptive Cauchy mutation and adaptive step size factor.Adaptive search based on the discrete degree of the current population and the position state of the solution,so as to improve the search ability.(2)Based on the analysis of the basic flower pollination algorithm applied to the generation of test data,this paper research how to apply the adaptive hybrid flower pollination algorithm to the automatic generation of test data,a test data generation model based on adaptive hybrid flower pollination algorithm is established.Simultaneously,an improved fitness function construction method is proposed,each branch is assigned a corresponding weight parameter value by a different degree of difficulty in covering the branch to more accurately reflect the coverage of the branch,so as to further improve the test data generation efficiency.(3)Finally,the feasibility and efficiency of the adaptive hybrid flower pollination algorithm proposed in this paper are validated in the automatic generation of test data.Four typical testprograms with different complexity are selected,and the test data is automatically generated by MATLAB platform.Compared with the other two intelligent algorithms used in the automatic generation of test data,the average time-consuming,the average number of iterations and the average branch coverage rate are compared and analyzed.