A Fast and Efficient Algorithm for Intelligent Test Paper Generating

Abstract

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 generating, this paper introduces an intelligent test paper generating algorithm based on chapter ratio score and release strategy. The algorithm transforms test paper generating problem into the question of finding a suitable solution of multivariate equations group with multiple conditions. It initially determines the numbers of those pending small questions of various question types in each chapter according to its ratio score. If the number determined by the proportion can’t meet the requirements, then firstly adjust the numbers of various types of questions to be selected into paper within the chapter while maintaining the same chapter ratio score. If the number can’t be adjusted within the chapter, then make other adjustable chapters release its part share of scarce type of questions for the chapter. Eventually it will find scheme for generating test paper to meet ratio score of each chapter and strictly to meet the types of questions and the number of small questions of various types. Theoretical analysis and experimental results show that the algorithm can quickly and efficiently find a reasonable solution for generating test paper compared with the current genetic algorithms.