论文部分内容阅读
网络科技论文影响力的评价效果取决于评价指标变量的选择。将网络科技论文影响力评价与论文排名相关联,以Web of Science数据库中数学类论文为样本,从6个不同的排名等级组,即排名前0.01%、0.01%-0.1%、0.1%-1%、1%-10%、10%-20%、20%-50%,分别抽取论文数十篇,用文献信息方法对单篇论文从内容、论文载体和论文作者三个层面初选28个特征变量,以324篇论文的排名等级与28个学术链接指标样本建立为“序回归”模型的研究问题。基于Lasso方法对28个学术链接指标进行变量选择和参数估计,获得9个学术链接特征指标作为评价网络科技论文学术影响力的基本特征指标;以418篇OA论文的排名等级对23个网络影响计量及其衍生变量进行变量选择,获得5个论文网络传播与利用影响力的评价指标。最终共获得14个网络科技论文学术影响力的评价指标。
The effectiveness of the evaluation of online scientific papers depends on the choice of variables of evaluation index. The dissertation evaluates the influence of web-based scientific papers on the dissertation rankings. Based on the math papers in the Web of Science database, six disparate ranking rankings, ie, top 0.01%, 0.01% -0.1%, 0.1% -1 %, 1% -10%, 10% -20% and 20% -50%, respectively, and dozens of papers were extracted respectively, and 28 papers were selected from the three aspects of content, essay and author The eigenvariables, with the rankings of 324 papers and 28 samples of academic link indicators, were established as the research questions of “Regression” model. Based on the Lasso method, 28 academic link indexes were selected and their parameters were estimated. Nine academic link characteristics indexes were obtained as the basic characteristic indexes to evaluate the academic influence of the network scientific papers. Based on the rankings of 418 papers, 23 network influence measures And its derivative variables to choose variables to obtain the evaluation index of the influence of 5 papers on network communication and utilization. In the end, a total of 14 network science and technology thesis academic influence evaluation index.