Exploiting citation networks for large-scale author name disambiguation

Abstract

We present a novel algorithm and validation method for disambiguating author names in very large bibliographic data sets and apply it to the full Web of Science (WoS) citation index. Our algorithm relies only upon the author and citation graphs available for the whole period covered by the WoS. A pair-wise publication similarity metric, which is based on common co-authors, self-citations, shared references and citations, is established to perform a two-step agglomerative clustering that first connects individual papers and then merges similar clusters. This parameterized model is optimized using an h-index based recall measure, favoring the correct assignment of well-cited publications, and a name-initials-based precision using WoS metadata and cross-referenced Google Scholar profiles. Despite the use of limited metadata, we reach a recall of 87% and a precision of 88% with a preference for researchers with high h-index values. 47 million articles of WoS can be disambiguated on a single machine in less than a day. We develop an h-index distribution model, confirming that the prediction is in excellent agreement with the empirical data, and yielding insight into the utility of the h-index in real academic ranking scenarios.

The Global Burden of Journal Peer Review in the Biomedical Literature: Strong Imbalance in the Collective Enterprise

“…Conversely, a given individual researcher could appear as several researchers because of misspellings. We acknowledge that we did not use algorithmic author name disambiguation [13]. The first type of error would lead to underestimating the number of potential reviewers and the second to overestimating the number of potential reviewers.…”

“…We did not use any methods for author name disambiguation for researchers indexed under the same “LastName”, “ForeName” and “Initials”. [13, 14] We set N s to be equal to the number of publications for which we identified at least one author.…”

Evaluating author name disambiguation for digital libraries: a case of DBLP

“…Specifically, some scholars assume that if a name in paper X appears in paper Y that cites X, those two name instances appearing both in paper X and Y refer to the same author (Levin et al, 2012;Schulz, Mazloumian, Petersen, Penner, & Helbing, 2014;. Email address has also been used for the automatic construction of labeled data: two names associated with the same email address are assumed to represent the same author (e.g., Schulz et al, 2014). As shown in the example studies in Table 1, scholars have evaluated AND for digital libraries using various types of labeled data and evaluation metrics.…”

“…The performance of DBLP's disambiguation was tested on automatically labeled data (SelfCite hereafter) consisting of pairs of name instances in self-citation relation. These labeled data can be used for measuring recall defined as the ratio of the number of pairs matched by DBLP's disambiguation over the number of pairs in the labeled data (Liu et al, 2014;Schulz et al, 2014;. As this recall metric considers only name pairs in self-citation relation, it is different from the pairwise-F recall which considers all possible pairs among name instances referring to the same author.…”

A Graph Combination With Edge Pruning‐Based Approach for Author Name Disambiguation

“…Name homonymy causes a merge citation problem, whereas name synonymy leads to a split citation (Lee, On, Kang, & Park, ) problem. Such ambiguities deteriorate the performance of document retrieval, web search, and various bibliometric analysis like coauthorship link prediction (Wang & Sukthankar, ), collaboration network analysis (Viana, Amancio, & Costa, ), and citation network analysis (Amancio, Oliveira Jr, & da Fontoura Costa, ; Schulz, Mazloumian, Petersen, Penner, & Helbing, ).…”