Research On Author Impact Assessment In Academic Network

Academic network is a kind of network structure developed in the context of the rapid growth of academic papers.With a large number publication of academic papers,the scale of academic networks is becoming increasingly large.It's an important and difficult task at present that how to efficiently find valuable academic information from the complex academic networks.The academic influence assessment algorithm is an evaluation algorithm developed to obtain the most valuable academic information in academic networks.Academic impact assessment refers to the academic network of various academic subjects,such as authors,papers,journals and affiliations to evaluate.As the most important subject of research,papers and authors link various academic subjects in academic networks,and the evaluation of academic influence mainly focuses on papers and authors.This paper focuses on papers rankings and author's influence evaluation algorithm,the main work is as follows:(1)Aiming at the existing problems in the current paper ranking algorithms,through introducing the time effective of the publication and reference to calculate the attenuation of the papers' value with time.At the same time,considering the influence of the impact factors of the journal where the paper is located on the paper ranking algorithm are improved.PageRank algorithm has a multi-feature paper ranking algorithm that combines multiple features.The improved multi-feature paper ranking algorithm is used to test the data set in ACM Information System Proceedings and Microsoft Academic Search Data Set Software Testing respectively,and the performance of the algorithm is evaluated.Compared with the PageRank algorithm,the multi-feature paper ranking algorithm considers the publication time of the paper and the time when the paper is referenced,improves the ranking algorithm of the paper value based on the time effective of the paper and the impact factor of the journal in which it is used,and improves the performance of the algorithm.(2)Based on the multi-feature paper ranking algorithm,h-index algorithm is improved based on the author's value.The multi-feature author influence assessment method is designed and the value of author is based on the signature position of authors in their papers and the level of author's affiliations to reflect.Experiments on multi-feature author's influence assessment method in the field of software testing of Microsoft Academic Search data set prove that the accuracy of the algorithm is better than h-index algorithm,and obtained the better result obtained in the data set of Software Testing.