基于遗传算法的人力资源组合测评模型

Hybrid evaluating model for human resource based on genetic algorithm

提出一种组合测评模型,通过构造权向量表示不同测评模型对人才评价的测评性能,把多测评模型的组合问题转化为对权重的优化问题.在测评准确度的基础上设计适应度函数,采用遗传算法对权向量进行优化,并设计了个体选择策略,避免陷入局部优化.实验表明,所构造的模型能够有效降低测评结果的平均误差.

At the basis of genetic algorithm,a hybrid recommender model is proposed to integrate outputs produced by every recommender.Through constructing weight vectors that represent different forecasting performances of each recommender,the hybrid problem is translated into optimizing problem about weight vectors.Fitness function is designed based on forecasting accuracy,and a selecting strategy is constructed to avoid premature of GA.Experiments express the proposed model has higher capability than traditional method.