Fast Training of Structured SVM Using Fixed-Threshold Sequential Minimal Optimization

Lee, Changki; Jang, Myung-Gil

doi:10.4218/etrij.apr2009.rp080276

Cited by 4 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We will call this model "Blog.Adapt Model". The training was done using fixed-threshold sequential minimal optimization method [16]. We also trained a model for tweets using the news model as the basis.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…His work, however, was applied to ACE 2003 RDC data, which uses 24 relations types which are too general and ambiguous for our task. While maximum entropy is a very popular classifier, we employ structured SVM, which shows better performance in various academic fields [16].…”

Section: Related Workmentioning

confidence: 99%

“…Structural SVM tries to solve the following quadratic program: 1) is the gradient of the hyperplane that maximizes the margin between the output classes. w can be trained by using cutting-plane algorithm [15] or its variants [16]- [18].…”

Section: Structural Svmmentioning

confidence: 99%

See 2 more Smart Citations

Extracting Events from Web Documents for Social Media Monitoring Using Structured SVM

Choi¹,

Ryu²,

Kim³

et al. 2013

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYEvent extraction is vital to social media monitoring and social event prediction. In this paper, we propose a method for social event extraction from web documents by identifying binary relations between named entities. There have been many studies on relation extraction, but their aims were mostly academic. For practical application, we try to identify 130 relation types that comprise 31 predefined event types, which address business and public issues. We use structured Support Vector Machine, the state of the art classifier to capture relations. We apply our method on news, blogs and tweets collected from the Internet and discuss the results.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Extracting Events from Web Documents for Social Media Monitoring Using Structured SVM

Choi¹,

Ryu²,

Kim³

et al. 2013

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…We modified the original cutting plane algorithm [15] and solved the dual form of the quadratic problem of line 11 in Alg. 1 through the Sequential Minimal Optimization (SMO) approach, inspired by the work of Lee and Jang [19]. Let L P be the Lagrangian of the problem in Eq.…”

Section: Shape-aware Loss Functionmentioning

confidence: 99%

Social Groups Detection in Crowd through Shape-Augmented Structured Learning

Solera

Calderara

2013

Image Analysis and Processing – ICIAP 2013

View full text Add to dashboard Cite

Abstract. Most of the behaviors people exhibit while being part of a crowd are social processes that tend to emerge among groups and as a consequence, detecting groups in crowds is becoming an important issue in modern behavior analysis. We propose a supervised correlation clustering technique that employs Structural SVM and a proxemic based feature to learn how to partition people trajectories in groups, by injecting in the model socially plausible shape configurations. By taking into account social groups patterns, the system is able to outperform state of the art methods on two publicly available benchmark sets of videos.

show abstract

“…In this figure, the precision indicates the ratio of correct candidate documents from candidate documents identified as duplicates by the proposed model; while the recall indicates the ratio of correct candidate documents from all of 924 duplicated documents. F-measure indicates the harmonic mean of the precision and the recall [12].…”

Section: Performance Of Proposed Modelmentioning

confidence: 99%

Detecting Partial and Near Duplication in the Blogosphere

Yoon

Jang

Kim

et al. 2012

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYIn this paper, we propose a duplicate document detection model recognizing both partial duplicates and near duplicates. The proposed model can detect partial duplicates as well as exact duplicates by splitting a large document into many small sentence fingerprints. Furthermore, the proposed model can detect even near duplicates, the result of trivial revisions, by filtering the common words and reordering the word sequence. key words: duplicate detection, sentence fingerprint, information retrieval, blogs

show abstract

Fast Training of Structured SVM Using Fixed-Threshold Sequential Minimal Optimization

Cited by 4 publications

References 12 publications

Extracting Events from Web Documents for Social Media Monitoring Using Structured SVM

Extracting Events from Web Documents for Social Media Monitoring Using Structured SVM

Social Groups Detection in Crowd through Shape-Augmented Structured Learning

Detecting Partial and Near Duplication in the Blogosphere

Contact Info

Product

Resources

About