Auto-generating Test Paper Research Articles

A Quantum-behaved Particle Swarm Algorithm on Autogenerating Test Paper

This paper puts forward an improved particle swarm algorithm in order to improve the efficiency of autogenerating test paper. The paper conducts mathematically modeling based on item response theory. In the new algorithm, the inertia weight is expressed as functions of particle evolution velocity and particle aggregation by defining particle evolution velocity and particle aggregation so that the inertia weight has adaptability. At the same time, slowly varying function is introduced to the traditional location updating formula so that the local optimal solution can be effectively overcome. Simulation results show that compared with the quantum-behaved particle swarm algorithm, the proposed algorithm has better performance in success rate and composing efficiency.

Intelligent Generating Examination Paper from Item Bank Based on Improved Genetic Algorithm

In this paper, we analyze the disadvantage of common generating test paper algorithm. An improved genetic algorithm (IGA) is proposed and used in auto-generating examination paper algorithm. We design the mathematical model of auto-generating test paper algorithm and improved the traditional GA fitness evaluation form. A computational study is carried out to verify the algorithm. Simulation results demonstrate that the performance of IGA can work efficiently than traditional ones.

The Research of Auto-Composing Test Paper Technology Based on Genetic Algorithm

In order to meet the demand of test paper in the network test system, the paper establishes test paper mathematical model, uses the integer coding strategy, and applies the improved genetic algorithm to the auto-generating test paper module. The results of the test show that the algorithm can complete the intelligent test paper well, and improve it effectively. Key words: auto-generating test paper; genetic algorithm; fitness function