Dependency-based semantic role labeling using sequence labeling with a structural SVM

Lim, Soojong; Lee, Changki; Ra, Dong-Yul

doi:10.1016/j.patrec.2013.01.022

Cited by 12 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A structural SVM was found to be suitable for SRL [7]. In this paper, we adopt the prior-model approach to facilitate domain adaptation in developing a structural SVM-based SRL system.…”

Section: Related Researchmentioning

confidence: 99%

“…In our SRL model we adopt a structural SVM, which was developed to build an SRL system and found to be effective for performing SRL [7]. In this section, we provide explanation for theoretical aspects of the structural SVM described in that work for completeness and readability of this paper.…”

Section: Structural Learning Model For Srlmentioning

confidence: 99%

“…The structural SVM for SRL explained in section III, subsection 3, is used to build our basic source-domain SRL component. For training the structural SVM, we follow the method that was introduced in [7].…”

Section: Basic Srl System For Source Domainmentioning

confidence: 99%

“…In building our system, SRL is carried out based on a structural learning model, actually a structural SVM, because it has been shown that the model is instrumental in building an SRL system with state-ofthe-art performance [7]. Out of several domain-adaptation methodologies, we choose an approach originally introduced in [8].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Domain-Adaptation Technique for Semantic Role Labeling with Structural Learning

Lim

Lee

Ryu

et al. 2014

ETRI J

Self Cite

View full text Add to dashboard Cite

Semantic role labeling (SRL) is a task in natural-language processing with the aim of detecting predicates in the text, choosing their correct senses, identifying their associated arguments, and predicting the semantic roles of the arguments. Developing a high-performance SRL system for a domain requires manually annotated training data of large size in the same domain. However, such SRL training data of sufficient size is available only for a few domains. Constructing SRL training data for a new domain is very expensive. Therefore, domain adaptation in SRL can be regarded as an important problem. In this paper, we show that domain adaptation for SRL systems can achieve state-of-the-art performance when based on structural learning and exploiting a prior model approach. We provide experimental results with three different target domains showing that our method is effective even if training data of small size is available for the target domains. According to experimentations, our proposed method outperforms those of other research works by about 2% to 5% in F-score.Keywords: Domain adaptation, semantic role labeling, natural language, semantic analysis, structured learning, prior model. I. IntroductionBig data explosion has led to an exponential growth in the amount of valuable textual data in many fields. Thus, automatic information retrieval (IR) and information extraction (IE) methods have become more important in helping researchers and analysts to keep track of the latest developments in their fields. Current IR is still mostly limited to keyword search and unable to infer relationships between entities in a text. A system that is able to understand how words in a sentence are related semantically can greatly improve the quality of IE and would allow IR to handle more complex user queries.Semantic role labeling (SRL) is a task for semantic processing of natural-language text, wherein the semantic role labels of the arguments associated with the predicates in a sentence are predicted. Recently, SRL has become increasingly popular as natural-language processing technology advances. The purpose of SRL is to find "who does what to whom, when, and where" in natural-language text by recognizing the semantic roles of the arguments of the predicates.As a result of performing SRL on a given sentence and its predicate, each word in the sentence is assigned a semantic role label. By combining the labels for the words, the output of SRL can be viewed as a sequence of semantic role labels. The sequence is generated for each predicate. For example, as in Fig. 1, the semantic role A0 represents the "agent" of "wants" and the semantic role A1 denotes the thing "being wanted." The information produced as a result of an SRL task is valuable for IE and other natural-language understanding tasks such as question answering [1] and online advertising services [2].In previous research, most works on SRL focused on

show abstract

“…A structural SVM was found to be suitable for SRL [7]. In this paper, we adopt the prior-model approach to facilitate domain adaptation in developing a structural SVM-based SRL system.…”

Section: Related Researchmentioning

confidence: 99%

Section: Structural Learning Model For Srlmentioning

confidence: 99%

Section: Basic Srl System For Source Domainmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Domain-Adaptation Technique for Semantic Role Labeling with Structural Learning

Lim

Lee

Ryu

et al. 2014

ETRI J

Self Cite

View full text Add to dashboard Cite

show abstract

“…Shalev-Shwartz et al (2011) proposed the Pegasos algorithm which takes a sub-gradient step with a predetermined step size and which can work in the mini-batch variant by choosing a set of examples and performing a sub-gradient step on it. Its structured 918 D. Mančev and B. Todorović version was successfully applied to various problems: dependency parsing (Martins et al, 2011), semantic role labeling (Lim et al, 2013), part-of-speech tagging (Ni et al, 2010), optical character recognition (Jaggi et al, 2012), and named entity recognition (Lee et al, 2011). The empirical performance indicated fast convergence with the results comparable with those of other structured algorithms, while Ratliff et al (2006) show that the cumulative prediction loss for the structured sub-gradient method grows only sublinearly in time.…”

Section: Introductionmentioning

confidence: 99%

A primal sub-gradient method for structured classification with the averaged sum loss

Mančev

Todorović

2014

International Journal of Applied Mathematics and Computer Science

View full text Add to dashboard Cite

We present a primal sub-gradient method for structured SVM optimization defined with the averaged sum of hinge losses inside each example. Compared with the mini-batch version of the Pegasos algorithm for the structured case, which deals with a single structure from each of multiple examples, our algorithm considers multiple structures from a single example in one update. This approach should increase the amount of information learned from the example. We show that the proposed version with the averaged sum loss has at least the same guarantees in terms of the prediction loss as the stochastic version. Experiments are conducted on two sequence labeling problems, shallow parsing and part-of-speech tagging, and also include a comparison with other popular sequential structured learning algorithms.

show abstract