摘要
利用投影寻踪和遗传算法,建立基于遗传投影寻踪的期刊质量评价模型。该模型把期刊质量多维评价指标综合成一维投影指标,根据该投影指标值的大小对样本集进行统一评价,从而解决了各单项期刊质量评价指标评价结果的不相容问题,提高了期刊质量综合评价问题各层次的分辩力。以水利工程类核心期刊(2008年版)为例进行期刊质量评价分析,并将评价结果与主成分分析法和灰色关联度法进行对比,验证了该方法的科学性、合理性和有效性。
Base on projection pursuit and genetic algorithm,the assessment model is presented for assessing the quality of sci-tech journals,integrating several dimension of evaluation indicator into one dimension as well as concluding composite assessment of samples with the evaluation of such projection.It solves the problem of incompatible assessment indexes among different evaluation indicator of sci-tech journals,in order to improve the resolution of each-level assessment on comprehensive evaluation.As an sample,Hydraulic core periodicals(2008)by Assessment Model of the Quality of Journals Based on Genetic Projection Pursuit(AMQJGPP).Compared assessment with the principle component analysis and Gray correlation degree,this model is proved to be scientifical,valid and rational.
出处
《情报科学》
CSSCI
北大核心
2013年第2期58-62,共5页
Information Science
关键词
期刊
质量
评价
投影寻踪
遗传算法
水利工程
journal
quality
assessment
projection pursuit
genetic algorithm
water conservancy and hydropower project