2005
DOI: 10.1007/11562382_71
|View full text |Cite
|
Sign up to set email alerts
|

An Iterative Approach for Web Catalog Integration with Support Vector Machines

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
5
0

Year Published

2006
2006
2022
2022

Publication Types

Select...
3
1
1

Relationship

2
3

Authors

Journals

citations
Cited by 5 publications
(6 citation statements)
references
References 6 publications
1
5
0
Order By: Relevance
“…As previously explained, the analysis of the data was done by applying the taxonomy from Chen et al, (2005). The taxonomy is beneficial to describe the common gratitude strategies as well for the explanation of the common gratitude strategy realized by Indonesian female and male EFL learners.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…As previously explained, the analysis of the data was done by applying the taxonomy from Chen et al, (2005). The taxonomy is beneficial to describe the common gratitude strategies as well for the explanation of the common gratitude strategy realized by Indonesian female and male EFL learners.…”
Section: Resultsmentioning
confidence: 99%
“…Second, the differences of saying gratitude strategies in English were explained based on Chen et al, (2005) taxonomy between Indonesian females and males EFL learners in academic situations made by the researcher. Last, the conclusion then can be drawn based on the findings and the result of the discussion.…”
Section: Methodsmentioning
confidence: 99%
“…Since SVM has presented superior performance in classification problems [Dumais et al 1998;Joachims 1998;Yang and Liu 1999;Rennie and Rifkin 2001], many related studies have also adopted the SVM classifiers with different strategies to extract the implicit information and improve the integration accuracy. These SVM-based integration approaches include a cross-training technique for SVM classifiers (SVM-CT) [Sarawagi et al 2003], a topic restriction strategy (SVM-TR) [Tsay et al 2003], a cluster shrinkage approach (CS-TSVM) [Zhang and Lee 2004a], and an iterative approach with pseudo-relevance feedback (SVM-IA) [Chen et al 2005]. Most of these approaches employing the SVM classifiers were found to have higher accuracy then ENB.…”
Section: Related Workmentioning
confidence: 99%
“…The most important approach, called ENB, enhances the Naive Bayes classifiers with implicit source information. Other state-of-the-art approaches, including Support Vector Machines (SVMs) [Sarawagi et al 2003;Tsay et al 2003;Zhang and Lee 2004a;Chen et al 2005;Chen et al 2006;Ho et al 2006] and the Maximum Entropy model [Wu et al 2005], have been also presented to elevate the performance of Web catalog integration, and they further outperform the ENB approach.…”
Section: Introductionmentioning
confidence: 99%
“…The enhanced Naïve Bayes classifier (ENB) is shown to have more than 14% accuracy improvement on average. The work in Chen et al (2005) also has the similar concept in its iterative pseudo relevance feedback approach. As reported in Chen et al (2005), the enhanced SVM classifiers consistently achieve improvement.…”
Section: Integration Techniquesmentioning
confidence: 99%