StarIn:An Approach to Predict the Popularity of GitHub Repository
作者机构:School of SoftwareDalian University of TechnologyDalian 116620LiaoningChina
出 版 物:《国际计算机前沿大会会议论文集》 (International Conference of Pioneering Computer Scientists, Engineers and Educators(ICPCSEE))
年 卷 期:2020年第2期
页 面:258-273页
学科分类:08[工学] 0812[工学-计算机科学与技术(可授工学、理学学位)]
基 金:the Natural Science Foundation of 453China grant 61502069, 61672128 by the Fundamental Research Funds 454 for the CentralUniversities grant DUT18JC39, DUT18GF108
主 题:Stargazer influence Popularity prediction Network relationship GitHub
摘 要:The popularity of repository in GitHub is an important indicator to evaluate its *** the trend of popularity is a crucial guideline to study its development ***,StarIn,a stargazer-influence based approach is proposed to predict the popularity of GitHub *** the followers in GitHub as a basic dataset,stargazer-following based network was *** indicator,stargazer influence,was measured from three aspects of basic influence,network dynamic influence and network static *** was conducted,and the correlation of StarIn was analyzed with the popularity of repository from the perspective of six *** experimental evaluation provides an interesting approach to predict the popularity of repositories in GitHub from a new *** achieves an excellent performance of predicting the popularity of repositories with a high accurate rate under two different classifiers.