Author Name Disambiguation by Using Deep Neural Network

Tran, Hung; Huynh, Tin Van; Do, Tien Van

doi:10.1007/978-3-319-05476-6_13

Cited by 49 publications

(51 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used this probability as the similarity measure for finding that either of two sets belong to the same author. Vishnyakova presented a journal descriptor indexing (JDI) tool to resolve ambiguous authors [19]. It utilized titles, abstracts, and MeSH terms as an input to the JDI and it returned the journal descriptors and semantic types as its output.…”

Section: Related Workmentioning

confidence: 99%

Resolving namesakes using the author's social network

Hussain¹,

Asghar²

2018

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Author name ambiguity may occur when multiple authors share the same name or different name variations of a single author exist. This degrades search results and correct attributions in bibliographic databases. Existing solutions require either the actual number of ambiguous authors or extra information that is collected from the Web. However, in many scenarios, obtaining such auxiliary information is not possible or requires much extra effort. An effective and scalable method, ASONET, is proposed that uses graph community detection algorithms and graph operations to disambiguate namesakes. The citation dataset is preprocessed and ambiguous author blocks are formed. A graph structural clustering, gSkeletonClu, is applied to identify hubs, outliers, and clusters of nodes in a coauthor's graph.Namesakes are resolved by splitting these clusters across the hub if their feature vector similarity is less than a predefined threshold. ASONET utilizes only coauthors and titles that are surely available in all bibliographic databases. To validate the ASONET performance, experiments are performed on two real-world datasets of Arnetminer and DBLP. The results confirm that ASONET is scalable and outperforms baselines.

show abstract

Section: Related Workmentioning

confidence: 99%

Resolving namesakes using the author's social network

Hussain¹,

Asghar²

2018

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…Also, there have been several proposed methods based on a Siamese network for face verification [6] [16] [11]. Tran, Huynh, and Do [17] applied a DNN as a classifier to the author matching problem. Since they represented feature vectors representing pairs of objects by concatenating fixed similarities or distance metrics, this approach cannot utilize a DNN's ability to obtain representations automatically.…”

Section: Related Workmentioning

confidence: 99%

“…While there have been many studies of applying a DNN to datum-wise classification, there have been few studies of applying a DNN to pairwise classification. Tran, Huynh, and Do [17] used a DNN as a classifier for the author matching problem. Since they represented feature vectors representing pairs of objects by concatenating fixed similarities and distance metrics, their approach does not take advantage of a DNN's ability to obtain feature representations automatically.…”

Section: Introductionmentioning

confidence: 99%

A Deep Neural Network for Pairwise Classification: Enabling Feature Conjunctions and Ensuring Symmetry

Atarashi

Oyama

Kurihara

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Pairwise classification is a computational problem to determine whether a given ordered pair of objects satisfies a binary relation R which is specified implicitly by a set of training data used for 'learning' R. It is an important component for entity resolution, network link prediction, protein-protein interaction prediction, and so on. Although deep neural networks (DNNs) outperform other methods in many tasks and have thus attracted the attention of machine learning researchers, there have been few studies of applying a DNN to pairwise classification. Important properties of pairwise classification include using feature conjunctions across examples. Also, it is known that making the classifier invariant to the data order is an important property in applications with a symmetric relation R, including those applications mentioned above. We first show that a simple DNN with fully connected layers cannot satisfy these properties and then present a pairwise DNN satisfying these properties. As an example of pairwise classification, we use the author matching problem, which is the problem of determining whether two author names in different bibliographic data sources refer to the same person. We show that the method using our model outperforms methods using a support vector machine and simple DNNs.

show abstract

“…These data usually contain noises, especially author name ambiguity [9]. Many approaches to solve this problem have been proposed recently [9], [10].…”

Section: B Bibliographic Datamentioning

confidence: 99%

A Potential Approach to Overcome Data Limitation in Scientific Publication Recommendation

Tran

Huynh

Hoang

2015

2015 Seventh International Conference on Knowledge and Systems Engineering (KSE)

Self Cite

View full text Add to dashboard Cite

Abstract-Data are essential for the experiments of relevant scientific publication recommendation methods but it is difficult to build ground truth data. A naturally promising solution is using publications that are referenced by researchers to build their ground truth data. Unfortunately, this approach has not been explored in the literature, so its applicability is still a gap in our knowledge. In this research, we systematically study this approach by theoretical and empirical analyses. In general, the results show that this approach is reasonable and has many advantages. However, the empirical analysis shows both positive and negative results. We conclude that, in some situations, this is a useful alternative approach toward overcoming data limitation. Based on this approach, we build and publish a dataset in computer science domain to help advancing other researches.

show abstract

Author Name Disambiguation by Using Deep Neural Network

Cited by 49 publications

References 9 publications

Resolving namesakes using the author's social network

Resolving namesakes using the author's social network

A Deep Neural Network for Pairwise Classification: Enabling Feature Conjunctions and Ensuring Symmetry

A Potential Approach to Overcome Data Limitation in Scientific Publication Recommendation

Contact Info

Product

Resources

About