Research And Application On The System Of Generating Test Paper Intelligently On The AGA

The rapid development of network, long-distance network education in modern education has become an important embranchment with lots of benefits such as new method, time flexibility, and wide range of features. Examination is an important component of education and a key to test student that can accept the knowledge .Test also make students understand their own merits, learning target; Simultaneity school can review the quality of teachers teaching, and the degree of the teaching difficulty. In long-distance education, examination is also an important component of the indivisible part of the education. So the problem of generating test paper intelligently as the key issues was widespread concerned.The problem of generating test paper intelligently is actually a multi-objective optimization problem. The traditional method such as the random selection and back tracking in the generating paper has some problems that the time is too long and the success rate is too low, there are obvious shortcomings, it is difficult to find an effective algorithm. How to find a useful method to solve the problem that can aim at the actual characteristics of generating test paper intelligently in the basis of the study has become a hot spot of concern to everyone. Many studies have shown that Genetic Algorithm for solving the generating test paper intelligently has great potential.This paper aim at the problem of generating test paper intelligently in the long-distance education , bring forward a generating test paper strategy based on the SGA and the new strategy is a advance AGA with the integer subsection coding ,focus on improving the GA's selection operation. In choosing areas of operation, a selection competition mechanism, and competition under adaptive selection mechanism has been improved and will contest the original selection mechanism in the race fixed size, and change the operation of the algorithm can be adaptive to change the size of the competition size, and further enhance efficiency of the algorithmic. Accordance with the actual characteristics of generating test paper intelligently to improve the mutation operation, and make it have a certain direction.This algorithm is applied to the actual online examination system, and compare with the original algorithms, validate the algorithm is effective and superior.