A new evaluation approach of science and technology to objective weighting: Weight giving based on data independence and data fluctuation (DIDF)

摘要

In this paper, we establish a evaluation approach system of science and technology, and analyze five evaluation approaches of linear objective weighting. A new evaluation approach of science and technology to objective weighting is then put forward on these bases— Weight Giving Based on Data Independence and Data Fluctuation. It is also tested and verified by data from papers in agricultural journals of Institute of Scientific and Technical Information of China. The method is to calculate the deviation ratio of evaluation index, and the index's independent information rate by the improved multiple correlation coefficient, then multiply the standardized deviation ratio by the standardized index's independent information rate and normalize the product to get weight. It is a synthesis of advantages of all approaches of linear objective weighting by now and it has enriched objective weighting theory. The applicable conditions for objective weighting are analyzed and discussed as well. The results show that the entropy weight method and discrete coefficient method have similarities in computing the weight of indices.