融合论文引用与学者信息的交叉领域学术影响力分析

The Calculation of Academic Influence in Cross-field by Integrating Paper Citations and Scholar Information

[目的/意义]交叉学科内的学术影响力评价是科学计量领域的重要研究内容,其对探究跨学科知识融合趋势及科研人员影响力的测度与评价等具有十分重要的意义。当前的影响力度量方法普遍存在着论文引用的载体单一化、传播单向化,并且没有考虑学者领域交叉程度等问题。[方法/过程]通过提取学者自身的领域信息特征,把影响力双向传播引入论文引用网络中,提出了一种融合论文引用传播与学者信息传播的交叉领域学术影响力识别方法(IPCSI),该方法首先通过量化t时刻节点文献的学科交叉度,构建学者—论文交叉度矩阵,在此基础上将引证文献对节点文献的引用传播作为影响力连边,计算t+1时刻引文因子向量,利用图上排序算法实现学术影响力的后向引文传播。其次,从学者间合著领域相关性视角,通过增加以学者自身为传播载体的学者影响力连边,计算t-1时刻学者因子向量,模拟该时刻下学术影响力的前向信息传播。最后,通过同时建模论文引文传播与学者信息传播,客观再现学术影响力的前序与后序传播过程,并采用调和计量方法分配学术贡献度,最终完成学者影响力双向传播的量化分析。[结果/结论]在真实数据集上验证了IPCSI方法的有效性与可行性,实验结果表明通过结合论文引用粒度与学者信息粒度,能够丰富交叉领域内学术影响力的传播路径,提升学术影响力评价的准确性。

[Purpose/significance]The evaluation of academic influence is an important research content in the field of scientometrics.It is of great significance in exploring the trend of interdisciplinary knowledge fusion and evaluating the influence of researchers.However,there are many defects in the current influence measurement methods,such as the single carrier of paper citations and the one-way communication.Meanwhile,they do not take into account the degree of interdisciplinary of scholars.[Method/process]This paper proposes an integrating paper citations and scholar information method(IPCSI),which combines citations dissemination and scholar information dissemination by extracting scholar domain information characteristics.Specifically,first of all,a cross-domain crossover matrix is constructed based on quantifying the interdisciplinary crossover of the t-time node literature.Afterwards,the t+1 time citation factor vector is calculated,where the citation propagation of the cited literature to the node literature is taken as the influence linkage.Based on this,the backward citation propagation of academic influence can be realized by using the graph ranking algorithm.Secondly,the linkage of scholar’influence,which takes scholars themselves as the communication carrier,is added to the citation network from the perspective of domain correlation of the coauthor ship.On this condition,the scholar factor vector at t-1 time is calculated to simulate the forward information propagation of academic influence.Finally,the process of pre-order and post-order propagation of academic influence is reproduced objectively by modeling both citation propagation and scholar information propagation.In addition,the harmonic measurement method to allocate academic contribution has reached the goal of completing the quantitative analysis of scholar influence dissemination.[Result/conclusion]The validity and feasibility of IPCSI is verified on real data sets.The experimental results show that by combining the granularity of paper citations and the granularity of scholar information,the propagation paths of academic influence in cross-fields can be enriched and the accuracy of academic influence evaluation can be improved.