Automatic test paper generation method for test questions in question bank through genetic algorithm

Abstract

The invention relates to a method for automatically forming test papers in a test bank by using a genetic algorithm. It adopts real number codes for the test questions in the test bank, which can overcome the shortcomings of using binary codes in the past that the search space is too large and the code length is too long. The decoding time of the test papers to be assembled improves the speed of obtaining the final test papers. In addition, the present invention generates the initial population of the test papers to be grouped by randomly generating the initial population according to the total number of questions, the proportion of questions, and the total score of the test papers to be grouped, which can speed up the convergence of the genetic algorithm and reduce the number of iterations, thereby reducing Formation time of the test paper. Therefore, the present invention can be widely used in the field of automatic paper formation.

Description

technical field [0001] The invention relates to a method for forming papers, in particular to a method for automatically forming papers for test questions in a question bank by using a genetic algorithm. Background technique [0002] At present, the paperwork of various written test papers is basically done manually. Usually, when a formal examination is released, the corresponding examination organization invites experts in various fields to conduct analysis based on the examination population and examination objectives, and to formulate questions in a closed and concentrated manner. Then, according to the actual needs, select the chapters, knowledge points, and question types of the required test papers, and manually complete the whole process of test paper composition, resulting in a lot of resource investment, high cost, heavy workload, incomplete coverage of knowledge points, and failure to produce test papers. The volume takes a long time and so on. Therefore, the au...

AI Technical Summary

Problems solved by technology

[0003] Existing examination methods generally use traditional optimization algorithms and traditional search algorithms, and traditional optimization algorithms iteratively seek the optimal solution from a single initial value, so it is easy to enter the local optimal solution by mistake; and the traditional search algorithms are single-point search Algorithm, it is easy to fall into the local optimal solution, and the local optimal may be the most comprehensive examination of a certain knowledge point compared with the whole test paper, resulting in the unreasonable situation of the whole test paper

Images