依著作及共同作者來評斷研究者之學術貢獻

Metrics for Judging an Author's Academic Contribution from his Publications and Co-Authors

連子淳(Zi-Chun Lian)

指導教授 ： 雷欽隆

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2010年)

https://doi.org/10.6342/NTU.2010.01516

摘要

在眾多評判研究者的研究成果的方法中，我們可以將之分為兩大群，一個是單一作者評分法，另一個則是多重作者評分法。在單一作者評分法中，並不加以考慮一篇出版物有多少位共同作者。但事實上，若是一篇出版物有較多的作者，比起較少作者的出版物，會有較高的機會被其他出版物引用。因此，在多重作者評分法中，會提出各種方法，使得在計算某一研究者之成果時，能排除其他共同作者的貢獻。此一問題被稱之為「多重作者之問題」。
這篇論文中，我們提出了一個新的方法來計算某一研究者的成果，稱之為hex-index。我們提出的方法對於現今的單一作者評分法和多重作者評分法，皆有較佳之處。在單一作者評分法中，比起現今最常用的評分法h-index，有著較高的解析度。在多重作者評分法中，我們提出了較合理的方法來評斷各個共同作者對一篇出版物的貢獻。除此之外，我們的評分方法不只可以計算某一研究者排除所有共同作者之貢獻後的研究成果，還可計算某一研究者排除部份共同作者之貢獻後的研究成果。
實驗的部份，我們從DBLP和Google Scholar網站，收集了各個研究者的著作、共同作者和引用紀錄等資訊。從收集到的資訊中，我們隨機挑選兩百位研究者做為實驗對像。我們發現，對於擁有較高h-index之研究者，hex-index之值多數會略大於著名的多重作者評分法，hm-index之值。

關鍵字

作者排名 ； 評判研究成果 ； 多重作者 ； 共同作者 ； 出版物之引用

並列摘要

There are several well-known methodologies for determining the ranking of authors. These methodologies could be classified into two major groups; one is single-authored index, in this group, it does not take into consideration the number of co-authors in one single publication; the other is multi-authored index. Since papers with multi-authors tend to get more citations, indexes in this group propose some solutions to evaluate a single author’s index alone, without other co-authors. This is the so called multi-authored problem.
In this paper, we proposed a novel methodology for author ranking; we called it hex-index. We made improvements both in single-authored index and multi-authored index. In single-authored index, we have more resolution than the h-index, the most popular index for author ranking around the academic world. And in multi-authored index, we provide more reasonable measurements for the contribution of each author in one publication. In addition, our method not only could calculate an author’s index excluding all co-authors, but also could measure his or her index when excluding a portion of the co-authors.
We collected data from DBLP and Google Scholar, including each author’s publications, co-authors and citation records. From this dataset, we sampled two hundred authors, and we discovered the fact that for authors with higher h-index, their hex-index would slightly bigger than hm-index, the famous multi-authored index.

並列關鍵字

author ranking ； author index ； citation index ； multi-authored index ； citation network

參考文獻

[2] A. Sidiropoulos, D. Katsaros, Y. Manolopoulos, “Generalized Hirsch h-index for disclosing latent facts in citation networks,” Scientometrics, 2007.

[3] A. Borodin, G. O. Roberts, J. S. Rosenthal, P. Tsaparas, “Link Analysis Ranking Algorithms Theory and Experiments,” ACM Transactions on Internet Technology, 2005.

[4] A. Schubert, W. Glänzel, B. Thijs, “The weight of author self-citations. A fractional approach to self-citation counting,” Scientometrics, 2006.

[5] A. Sidiropoulos, Y. Manolopoulos, “Generalized comparison of graph-based ranking algorithms for publications and authors,” Journal for Systems and Software, 2006.

[7] B. H. Jin, L. M. Liang, R. Rousseau, L. Egghe, “The R- and AR-indices: Complementing the h-index,” Chinese Science Bulletin, 2007.