中外供应链科研合作网络社团成员演化对比分析

Comparative Analysis on the Evolution of Community Members in Chinese and Foreign Supply Chain Scientific Research Cooperation Networks

考虑社区集体成员行为演化特征对理解网络演化有很大的作用。文章搜集了中外2010~2020年供应链相关文献,分为11个时间片,构建学者合作网络,用Louvain社团识别算法识别每个时间片上的社团。首先分析每个时间上的新老成员的演化特征,其次提出了一个新指标来表征新节点加入新社团的趋势,最后加入基于断边重连的零模型与实证结果进行对比。结果发现,国外新成员增长速度和占比总体上高于国内,说明国外学者供应链领域的研究更吸引新学者的加入;在新指标计算结果下,虽然新学者自身形成社团的比例都比较大,但是国内的趋势比国外的趋势更加稳定,比重稳定在80%上下;在零模型对比中,结果与实证差距较大,发现这样的行为并不是随机化的结果,而是有选择、有目的的。该工作对探究合作网络提供了一个新的视角,进一步加深对供应链方面学术合作的了解。

Considering the behavior evolution characteristics of community collective members plays a great role in our understanding of network evolution. This paper collected the Chinese and foreign relevant literature of supply chain from 2010 to 2020,divided it into 11 time slices, constructed the scholar cooperation network, and identified the communities on each time slice with Louvain community recognition algorithm. First analyzes the evolution characteristics of new and old members in each time, then proposes a new index to characterize the trend of new nodes joining new communities, and finally introduces the reshuffle model based on broken edge reconnection to compare with the empirical results. The results show that the growth rate and proportion of foreign new members are generally higher than those in China, indicating that foreign scholars’ research in the field of supply chain attracts more new scholars to join. According to the calculation results of the new indicators, although the proportion of new scholars forming associations is relatively large, the trend in China is more stable than that in foreign countries, and the proportion is stable at about 80%;in the comparison of reshuffle model, there is a large gap between the results and empirical results.It is found that such behavior is not the result of randomization, but selective and purposeful. This work provides a new perspective for us to explore the cooperation network and further deepen our understanding of academic cooperation in supply chain.