本文選題：作者學術影響力 + 加權引用��； 參考：《情報學報》2017年08期

【摘要】：在闡釋作者加權引用內(nèi)涵與類型基礎上,從非迭代加權引用和迭代加權引用兩個視角構建作者學術影響力評價模型。以國際信息計量學領域學者為評價對象,比較和檢驗了這兩種模型之間以及與其他普遍流行的指標模型——被引次數(shù)、h指數(shù)、傳統(tǒng)加權PageRank、加權PageRank_D之間的聯(lián)系與區(qū)別。結果表明非迭代加權引用模型與迭代加權引用模型之間有很顯著的較強相關性,但是兩者之間仍然存在一定差異。迭代加權引用模型的檢驗效果最好,非迭代加權引用模型的檢驗效果居中,但AUC檢驗值也達到了0.9以上,檢驗效果亦優(yōu)良。非迭代加權引用模型簡單易懂、便于操作,能夠較好地甄別出"領域活躍者";而迭代加權引用模型算法精準,能夠較好地甄別出"領域先行者"。在實際中,可以根據(jù)自身需求選擇最合適的評價模型。
[Abstract]:On the basis of explaining the connotation and type of author's weighted citation, the author's academic influence evaluation model is constructed from the perspectives of non-iterative weighted citation and iterative weighted citation. Taking international researchers in the field of information metrology as the evaluation object, this paper compares and tests the relationship and difference between the two models and other popular index models, such as the index of cited times, the traditional weighted Page rank, and the weighted PageRank_D. The results show that there is a strong correlation between the non-iterative weighted reference model and the iterative weighted reference model, but there are still some differences between the two models. The test effect of the iterative weighted citation model is the best, the non-iterative weighted citation model is the middle, but the AUC test value is more than 0. 9, and the test effect is also good. The non-iterative weighted reference model is easy to understand, easy to operate, and can be used to identify the "domain active", while the iterative weighted reference model algorithm is accurate and can be used to identify the "domain pioneer". In practice, the most suitable evaluation model can be selected according to its own needs.
【作者單位】： 山西大學管理與決策研究所山西大學科學評價研究中心;
【基金】：國家社會科學基金青年項目“作者引用網(wǎng)絡模式與功效研究”(12CTQ026) 山西省高等學校優(yōu)秀青年學術帶頭人項目“基于作者學術關系的學科知識圖譜研究”(2014052006)
【分類號】：G353.1
，

本文編號：1845541