On Graph-Based Name Disambiguation

Abstract

Name ambiguity stems from the fact that many people or objects share identical names in the real world. Such name ambiguity decreases the performance of document retrieval, Web search, information integration, and may cause confusion in other applications. Due to the same name spellings and lack of information, it is a nontrivial task to distinguish them accurately. In this article, we focus on investigating the problem in digital libraries to distinguish publications written by authors with identical names. We present an effective framework named GHOST (abbreviation for GrapHical framewOrk for name diSambiguaTion), to solve the problem systematically. We devise a novel similarity metric, and utilize only one type of attribute (i.e., coauthorship) in GHOST. Given the similarity matrix, intermediate results are grouped into clusters with a recently introduced powerful clustering algorithm called Affinity Propagation . In addition, as a complementary technique, user feedback can be used to enhance the performance. We evaluated the framework on the real DBLP and PubMed datasets, and the experimental results show that GHOST can achieve both high precision and recall .

Foundations of Digital Universities

“…For example, the institutional repository of the University of Trento does not provide any mechanism to disambiguate locations (of conferences and of editors). Examples of useful heuristics, though restricted to author names only, can be found in [12].…”

Network based Framework for Author Name Disambiguation Applications

“…In [2] the authors propose a general object distinction methodology called DISTINCT. DISTINCT assumes a relational structure between neighbor tuples then computes the linkage strengths using two measures -set resemblance and random walks.Fan,et al,proposed another kind of network based name disambiguation method called GHOST [4]. By using only one type of the publication attributes (coautorship) GHOST is claimed to get better performance than DISTINCT.…”

“…As an example, we take the one proposed in [4] which is described as follows (the title of each publication is replaced with its venue label like "SIGMOD'12" to save space).…”

“…For example, there are 950 papers with author "Wei Wang" in DBLP which actually corresponds to over 50 people with the name. Author name ambiguity problem widely exists in the digital libraries such as DBLP 1 , CiteSeer 2 , ACM Digital Library 3 , PubMed 4 etc. Nevertheless, it remains challenging to solve the name ambiguity problem.…”

Author Profile Enrichment for Cross-Linking Digital Libraries

“…The data are represented as vector space model where the distance between vectors represents the similarity. Such algorithms include the Cosine Similarity (CS) with TF-IDF, Jaccard Similarity, Jaro Winkler, and Levenshtein algorithms [7,9,[12][13][14].…”