BibSonomy :: publication ::

tag user group author concept BibTeX key search:all search:pitman

The blue social bookmark and publication sharing system.

(0)

This publication has not been reviewed yet.

rating distribution

average user rating

The average rating is computed over all reviews. However, some of them may be invisible to you due to the visibility setting chosen by the reviewers.

(0.0 of 5.0 based on 0 reviews)

Efficient topic-based unsupervised name disambiguation

by: Yang Song, Jian Huang, Isaac G. Councill, Jia Li, and C. Lee Giles

In: JCDL '07: Proceedings of the 2007 conference on Digital libraries New York, NY, USA: ACM (2007) , p. 342--351.

Resources (URL, PDF, PS...)

URL:	http://clgiles.ist.psu.edu/papers/JCDL2007-topic_based_name_disambiguation.pdf
DOI:	10.1145/1255175.1255243
Additional URLs:	ACM Guide URL:text:

Name ambiguity is a special case of identity uncertainty where one person can be referenced by multiple name variations in different situations or even share the same name with other people. In this paper, we focus on the problem of disambiguating person names within web pages and scientific documents. We present an efficient and effective two-stage approach to disambiguate names. In the first stage, two novel topic-based models are proposed by extending two hierarchical Bayesian text models, namely Probabilistic Latent Semantic Analysis PLSA and Latent Dirichlet Allocation LDA. Our models explicitly introduce a new variable for persons and learn the distribution of topics with regard to persons and words. After learning an initial model, the topic distributions are treated as feature sets and names are disambiguated by leveraging a hierarchical agglomerative clustering method. Experiments on web data and scientific documents from CiteSeer indicate that our approach consistently outperforms other unsupervised learning methods such as spectral clustering and DBSCAN clustering and could be extended to other research fields. We empirically addressed the issue of scalability by disambiguating authors in over 750,000 papers from the entire CiteSeer dataset.

Description

Efficient topic-based unsupervised name disambiguation

BibTeX record

@inproceedings{1255243,
  abstract = {Name ambiguity is a special case of identity uncertainty where one person can be referenced by multiple name variations in different situations or even share the same name with other people. In this paper, we focus on the problem of disambiguating person names within web pages and scientific documents. We present an efficient and effective two-stage approach to disambiguate names. In the first stage, two novel topic-based models are proposed by extending two hierarchical Bayesian text models, namely Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA). Our models explicitly introduce a new variable for persons and learn the distribution of topics with regard to persons and words. After learning an initial model, the topic distributions are treated as feature sets and names are disambiguated by leveraging a hierarchical agglomerative clustering method. Experiments on web data and scientific documents from CiteSeer indicate that our approach consistently outperforms other unsupervised learning methods such as spectral clustering and DBSCAN clustering and could be extended to other research fields. We empirically addressed the issue of scalability by disambiguating authors in over 750,000 papers from the entire CiteSeer dataset.

},
  acm = {http://portal.acm.org/citation.cfm?id=1255243},
  added-at = {2008-01-09T01:08:56.000+0100},
  address = {New York, NY, USA},
  author = {Song, Yang and Huang, Jian and Councill, Isaac G. and Li, Jia and Giles, C. Lee},
  biburl = {http://www.bibsonomy.org/bibtex/2624d2a29008c475245aa0bca052e7c0f/pitman},
  booktitle = {JCDL '07: Proceedings of the 2007 conference on Digital libraries},
  description = {Efficient topic-based unsupervised name disambiguation},
  doi = {http://doi.acm.org/10.1145/1255175.1255243},
  interhash = {e82d8ac4a42c2199149348551bbad51d},
  intrahash = {624d2a29008c475245aa0bca052e7c0f},
  isbn = {978-1-59593-644-8},
  keywords = {ml},
  location = {Vancouver, BC, Canada},
  pages = {342--351},
  publisher = {ACM},
  timestamp = {2008-01-09T01:08:56.000+0100},
  title = {Efficient topic-based unsupervised name disambiguation},
  url = {http://clgiles.ist.psu.edu/papers/JCDL2007-topic_based_name_disambiguation.pdf },
  year = 2007
}

Endnote record

%0 Conference Paper
%1 1255243
%A Song, Yang
%A Huang, Jian
%A Councill, Isaac G.
%A Li, Jia
%A Giles, C. Lee
%B JCDL '07: Proceedings of the 2007 conference on Digital libraries
%@ 978-1-59593-644-8
%C New York, NY, USA
%D 2007
%I ACM
%K 
%P 342--351
%T Efficient topic-based unsupervised name disambiguation
%R http://doi.acm.org/10.1145/1255175.1255243
%U http://clgiles.ist.psu.edu/papers/JCDL2007-topic_based_name_disambiguation.pdf 
%X Name ambiguity is a special case of identity uncertainty where one person can be referenced by multiple name variations in different situations or even share the same name with other people. In this paper, we focus on the problem of disambiguating person names within web pages and scientific documents. We present an efficient and effective two-stage approach to disambiguate names. In the first stage, two novel topic-based models are proposed by extending two hierarchical Bayesian text models, namely Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA). Our models explicitly introduce a new variable for persons and learn the distribution of topics with regard to persons and words. After learning an initial model, the topic distributions are treated as feature sets and names are disambiguated by leveraging a hierarchical agglomerative clustering method. Experiments on web data and scientific documents from CiteSeer indicate that our approach consistently outperforms other unsupervised learning methods such as spectral clustering and DBSCAN clustering and could be extended to other research fields. We empirically addressed the issue of scalability by disambiguating authors in over 750,000 papers from the entire CiteSeer dataset.

BibSonomy :: publication ::

Efficient topic-based unsupervised name disambiguation

Resources (URL, PDF, PS...)

Abstract

Description

BibTeX record

Endnote record