Relation extraction using label propagation based semi-supervised learning

Chen, Jinxiu; Ji, Donghong; Tan, Chew Lim; Niu, Zheng-Yu

doi:10.3115/1220175.1220192

Cited by 66 publications

(59 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Similarly as Chen et al (2006), who in turn followed Charniak (2000) and Zhang (2004), we experimented with additional constituency parse features describing grammatical functions and chunk tag information for all five contexts, and IOB-chains of the heads of the two entities. However, as preliminary experiments showed that these additional features do not provide any performance gains, we decided not to include them in our final models, intending to evaluate these results in future work.…”

Section: Svm Modelmentioning

confidence: 99%

TakeLab at SemEval-2018 Task 7: Combining Sparse and Dense Features for Relation Classification in Scientific Texts

Gluhak¹,

Buono²,

Akkasi³

et al. 2018

Proceedings of the 12th International Workshop on Semantic Evaluation

View full text Add to dashboard Cite

We describe two systems for semantic relation classification with which we participated in the SemEval 2018 Task 7, subtask 1 on semantic relation classification: an SVM model and a CNN model. Both models combine dense pretrained word2vec features and hancrafted sparse features. For training the models, we combine the two datasets provided for the subtasks in order to balance the under-represented classes. The SVM model performed better than CNN, achieving an F1-macro score of 69.98% on subtask 1.1 and 75.69% on subtask 1.2. The system ranked 7th among 28 submissions on subtask 1.1 and 7th among 20 submissions on subtask 1.2.

show abstract

Section: Svm Modelmentioning

confidence: 99%

TakeLab at SemEval-2018 Task 7: Combining Sparse and Dense Features for Relation Classification in Scientific Texts

Gluhak¹,

Buono²,

Akkasi³

et al. 2018

Proceedings of the 12th International Workshop on Semantic Evaluation

View full text Add to dashboard Cite

show abstract

“…Work conducted in [19] [20] employed semi-supervised learning algorithms and achieved good performance using only a small amount of labeled examples. They performed multi-class classification in which all the relation types are already defined [21].…”

Section: Related Workmentioning

confidence: 99%

“…Work conducted in [19] [20] performed multi-class relation classification [21] based on a hierarchical classification of relation types. However, relations to be extracted from Wikipedia are more fine-grained and diverse so that no such relation type classification is available in Wikipedia.…”

Section: Introductionmentioning

confidence: 99%

PORE: Positive-Only Relation Extraction from Wikipedia Text

Wang

Zhu

2007

The Semantic Web

View full text Add to dashboard Cite

Abstract. Extracting semantic relations is of great importance for the creation of the Semantic Web content. It is of great benefit to semi-automatically extract relations from the free text of Wikipedia using the structured content readily available in it. Pattern matching methods that employ information redundancy cannot work well since there is not much redundancy information in Wikipedia, compared to the Web. Multi-class classification methods are not reasonable since no classification of relation types is available in Wikipedia. In this paper, we propose PORE (Positive-Only Relation Extraction), for relation extraction from Wikipedia text. The core algorithm B-POL extends a state-of-the-art positive-only learning algorithm using bootstrapping, strong negative identification, and transductive inference to work with fewer positive training examples. We conducted experiments on several relations with different amount of training data. The experimental results show that B-POL can work effectively given only a small amount of positive training examples and it significantly outperforms the original positive learning approaches and a multi-class SVM. Furthermore, although PORE is applied in the context of Wikipedia, the core algorithm B-POL is a general approach for Ontology Population and can be adapted to other domains.

show abstract

“…Further, due to the vast coverage of Wikipedia the extensions of the relations can be assumed to be relatively complete. Most of the above described datasets have been obtained from Wikipedia by automatically resolving category membership with the help of the CatScan 4 Tool by Daniel Kinzler. CatScan was applied iteratively to also obtain members of sub-categories.…”

Section: Datasetsmentioning

confidence: 99%

“…The methods include kernel-based methods [18,6] and graph-labeling techniques [4]. The advantage of such methods is that abstraction and partial matches are inherent features of the learning algorithm.…”

Section: Related Workmentioning

confidence: 99%

Using the Web to Reduce Data Sparseness in Pattern-Based Information Extraction

Blohm

Cimiano

Knowledge Discovery in Databases: PKDD 2007

View full text Add to dashboard Cite

Abstract. Textual patterns have been used effectively to extract information from large text collections. However they rely heavily on textual redundancy in the sense that facts have to be mentioned in a similar manner in order to be generalized to a textual pattern. Data sparseness thus becomes a problem when trying to extract information from hardly redundant sources like corporate intranets, encyclopedic works or scientific databases.We present results on applying a weakly supervised pattern induction algorithm to Wikipedia to extract instances of arbitrary relations. In particular, we apply different configurations of a basic algorithm for pattern induction on seven different datasets. We show that the lack of redundancy leads to the need of a large amount of training data but that integrating Web extraction into the process leads to a significant reduction of required training data while maintaining the accuracy of Wikipedia. In particular we show that, though the use of the Web can have similar effects as produced by increasing the number of seeds, it leads overall to better results. Our approach thus allows to combine advantages of two sources: The high reliability of a closed corpus and the high redundancy of the Web.

show abstract

Relation extraction using label propagation based semi-supervised learning

Cited by 66 publications

References 14 publications

TakeLab at SemEval-2018 Task 7: Combining Sparse and Dense Features for Relation Classification in Scientific Texts

TakeLab at SemEval-2018 Task 7: Combining Sparse and Dense Features for Relation Classification in Scientific Texts

PORE: Positive-Only Relation Extraction from Wikipedia Text

Using the Web to Reduce Data Sparseness in Pattern-Based Information Extraction

Contact Info

Product

Resources

About