Bi-directional Joint Inference for Entity Resolution and Segmentation Using Imperatively-Defined Factor Graphs

Singh, Sameer; Schultz, Karl; McCallum, Andrew

doi:10.1007/978-3-642-04174-7_27

Cited by 19 publications

(17 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Every pair of consecutive samples in the MCMC chain is ranked according to the model and the ground truth, and the parameters are updated when the rankings disagree. This allows the learner to acquire more supervision per instance, and has led to efficient training for models in which inference is expensive and generally intractable [23].…”

Section: Rank-based Learning and Distant Supervisionmentioning

confidence: 99%

Modeling Relations and Their Mentions without Labeled Text

Riedel

Yao

McCallum

2010

Lecture Notes in Computer Science

Self Cite

1,008

952

View full text Add to dashboard Cite

Abstract. Several recent works on relation extraction have been applying the distant supervision paradigm: instead of relying on annotated text to learn how to predict relations, they employ existing knowledge bases (KBs) as source of supervision. Crucially, these approaches are trained based on the assumption that each sentence which mentions the two related entities is an expression of the given relation. Here we argue that this leads to noisy patterns that hurt precision, in particular if the knowledge base is not directly related to the text we are working with. We present a novel approach to distant supervision that can alleviate this problem based on the following two ideas: First, we use a factor graph to explicitly model the decision whether two entities are related, and the decision whether this relation is mentioned in a given sentence; second, we apply constraint-driven semi-supervision to train this model without any knowledge about which sentences express the relations in our training KB. We apply our approach to extract relations from the New York Times corpus and use Freebase as knowledge base. When compared to a state-of-the-art approach for relation extraction under distant supervision, we achieve 31% error reduction.

show abstract

Section: Rank-based Learning and Distant Supervisionmentioning

confidence: 99%

Modeling Relations and Their Mentions without Labeled Text

Riedel

Yao

McCallum

2010

Lecture Notes in Computer Science

Self Cite

1,008

952

View full text Add to dashboard Cite

show abstract

“…does not contain rules referring to specific strings occurring in the data, which achieves an AURPC of .971 [40]. Note that the MLN based approach in [40] -as well as more recent approaches achieving still higher accuracy [33,39] -perform collective classification, and therefore can exploit the fact that the binary relation on bibliographic records that one predicts is an equivalence relation. The two classification models we have used both perform independent predictions for each pair of bibliographic records, and therefore cannot be expected to achieve results that are competitive with state-of-the-art collective approaches.…”

Section: Coramentioning

confidence: 99%

“…The two classification models we have used both perform independent predictions for each pair of bibliographic records, and therefore cannot be expected to achieve results that are competitive with state-of-the-art collective approaches. It should be emphasized, though, that in [33,39] the MLN structure (i.e. the set of logical formulae) was carefully designed by hand, while in our experiments the TET structure is learned from data.…”

Section: Coramentioning

confidence: 99%

Type Extension Trees for feature construction and learning in relational domains

Jaeger

Lippi²,

Passerini

2013

Artificial Intelligence

View full text Add to dashboard Cite

Please cite this article in press as: M. Jaeger et al., Type extension trees for feature construction and learning in relational domains, Artificial Intelligence (2013), http://dx.doi.org/10. 1016/j.artint.2013.08.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Type Extension Trees for Feature Construction and Learning in Relational DomainsManfred Jaeger AbstractType Extension Trees are a powerful representation language for "count-ofcount" features characterizing the combinatorial structure of neighborhoods of entities in relational domains. In this paper we present a learning algorithm for Type Extension Trees (TET) that discovers informative count-of-count features in the supervised learning setting. Experiments on bibliographic data show that TET-learning is able to discover the count-of-count feature underlying the definition of the h-index, and the inverse document frequency feature commonly used in information retrieval. We also introduce a metric on TET feature values. This metric is defined as a recursive application of the Wasserstein-Kantorovich metric. Experiments with a k-NN classifier show that exploiting the recursive count-of-count statistics encoded in TET values improves classification accuracy over alternative methods based on

show abstract

“…The dataset References origins in a domain that is very popular for the evaluation of novel IE techniques (cf. [1,11,12,13]), whereas the dataset Curricula Vitae belongs to classical IE problems of template extraction.…”

Section: Datasetsmentioning

confidence: 99%

“…The work of Peng et al [11] provides a deep analysis of different settings and established linear-chain CRFs as the state-of-the-art for the segmentation of references. Approaches for joint inference [12,13] combine different tasks within a model. Here, the accuracy of the labeling can be increased when entity resolution and segmentation are jointly performed.…”

Section: Related Workmentioning

confidence: 99%

Collective Information Extraction with Context-Specific Consistencies

Kluegl

Toepfer

Lemmerich

et al. 2012

Machine Learning and Knowledge Discovery in Databases

View full text Add to dashboard Cite

Abstract. Conditional Random Fields (CRFs) have been widely used for information extraction from free texts as well as from semi-structured documents. Interesting entities in semi-structured domains are often consistently structured within a certain context or document. However, their actual compositions vary and are possibly inconsistent among different contexts. We present two collective information extraction approaches based on CRFs for exploiting these context-specific consistencies. The first approach extends linear-chain CRFs by additional factors specified by a classifier, which learns such consistencies during inference. In a second extended approach, we propose a variant of skip-chain CRFs, which enables the model to transfer long-range evidence about the consistency of the entities. The practical relevance of the presented work for real-world information extraction systems is highlighted in an empirical study. Both approaches achieve a considerable error reduction.

show abstract

Bi-directional Joint Inference for Entity Resolution and Segmentation Using Imperatively-Defined Factor Graphs

Cited by 19 publications

References 12 publications

Modeling Relations and Their Mentions without Labeled Text

Modeling Relations and Their Mentions without Labeled Text

Type Extension Trees for feature construction and learning in relational domains

Collective Information Extraction with Context-Specific Consistencies

Contact Info

Product

Resources

About