An Evaluation of the Robustness of MTS for Imbalanced Data

Su, Chao‐Ton; Hsiao, Yu-Hsiang

doi:10.1109/tkde.2007.190623

Cited by 124 publications

(78 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The crucial screened characteristic variables can be used to construct reduced class prediction models with high total accuracy while maintaining the model's ability to identify each class [44,45]. Therefore, the MTS is appropriate for class prediction applications in various fields.…”

Section: Discussionmentioning

confidence: 99%

Applying the Mahalanobis–Taguchi System to Improve Tablet PC Production Processes

Peng

Tsai

et al. 2017

Sustainability

View full text Add to dashboard Cite

Product testing is a critical step in tablet PC manufacturing processes. Purchases of testing equipment and on-site testing personnel increase overall manufacturing costs. In addition, to improve manufacturing capabilities, manufacturers must also produce products with higher quality and at a lower cost than their competitors if they are to attract consumers and gain a competitive edge in their industry. The Mahalanobis-Taguchi System (MTS) is a novel technique proposed by Genichi Taguchi for performing diagnoses and forecasting with multivariate data. The MTS can be used to select important factors and has been applied in numerous engineering fields to improve product and process quality. In the present study, the MTS, logistic regression, and a neural network were used to improve the tablet PC product testing process. The results indicated that the MTS attained 98% predictive power after insignificant test items were eliminated. The MTS performance was superior to those of the conventional logistic regression and neural network, which attained 93.3% and 94.7% predictive power, respectively. After the testing process was improved using the MTS, the number of test items in the tablet PC product testing process was reduced from 56 to 14. This facilitated the development of more stable test site configurations and effectively reduced the testing time, number of testers required, and equipment costs.

show abstract

Section: Discussionmentioning

confidence: 99%

Applying the Mahalanobis–Taguchi System to Improve Tablet PC Production Processes

Peng

Tsai

et al. 2017

Sustainability

View full text Add to dashboard Cite

show abstract

“…To further illustrate the classification performance, we report also relative sensitivity (RS), a measure proposed by Su and Hsiao [18] to evaluate the relative accuracy on positive and negative examples; the relative sensitivity is defined as…”

Section: Cross-validation and Evaluation Criteriamentioning

confidence: 99%

Distributed Road Surface Condition Monitoring Using Mobile Phones

Perttunen

Mazhelis

Cong

et al. 2011

Lecture Notes in Computer Science

136

View full text Add to dashboard Cite

Abstract. The objective of this research is to improve traffic safety through collecting and distributing up-to-date road surface condition information using mobile phones. Road surface condition information is seen useful for both travellers and for the road network maintenance. The problem we consider is to detect road surface anomalies that, when left unreported, can cause wear of vehicles, lesser driving comfort and vehicle controllability, or an accident. In this work we developed a pattern recognition system for detecting road condition from accelerometer and GPS readings. We present experimental results from real urban driving data that demonstrate the usefulness of the system. Our contributions are: 1) Performing a throughout spectral analysis of tri-axis acceleration signals in order to get reliable road surface anomaly labels. 2) Comprehensive preprocessing of GPS and acceleration signals. 3) Proposing a speed dependence removal approach for feature extraction and demonstrating its positive effect in multiple feature sets for the road surface anomaly detection task. 4) A framework for visually analyzing the classifier predictions over the validation data and labels.

show abstract

“…A thorough review, analysis and criticism against MT strategy can be found in Woodall et al (2003) along with responses and discussions. A study (Su and Hsiao 2007) on imbalanced data indicates the superiority of the method with respect to robustness. More recent research extending MT strategy and combining it with other data mining methods can be found in Huang et al (2009), Su and Hsiao (2009) and Pal and Maiti (2010).…”

Section: Mahalanobis-taguchi Strategymentioning

confidence: 99%

“…The Taguchi methods have been extensively applied in industry, despite the criticisms and discussions raised by several statisticians (Woodall et al 2003;Su and Hsiao 2007;Huang et al 2009). …”

Section: Introductionmentioning

confidence: 99%

Applying the Mahalanobis-Taguchi strategy for software defect diagnosis

2011

View full text Add to dashboard Cite

The Mahalanobis-Taguchi (MT) strategy combines mathematical and statistical concepts like Mahalanobis distance, Gram-Schmidt orthogonalization and experimental designs to support diagnosis and decision-making based on multivariate data. The primary purpose is to develop a scale to measure the degree of abnormality of cases, compared to "normal" or "healthy" cases, i.e. a continuous scale from a set of binary classified cases. An optimal subset of variables for measuring abnormality is then selected and rules for future diagnosis are defined based on them and the measurement scale. This maps well to problems in software defect prediction based on a multivariate set of software metrics and attributes. In this paper, the MT strategy combined with a cluster analysis technique for determining the most appropriate training set, is described and applied to well-known datasets in order to evaluate the fault-proneness of software modules. The measurement scale resulting from the MT strategy is evaluated using ROC curves and shows that it is a promising technique for software defect diagnosis. It compares favorably to previously evaluated methods on a number of publically available data sets. The special characteristic of the Autom Softw Eng (2012) 19:141-165 MT strategy that it quantifies the level of abnormality can also stimulate and inform discussions with engineers and managers in different defect prediction situations.

show abstract

An Evaluation of the Robustness of MTS for Imbalanced Data

Cited by 124 publications

References 19 publications

Applying the Mahalanobis–Taguchi System to Improve Tablet PC Production Processes

Applying the Mahalanobis–Taguchi System to Improve Tablet PC Production Processes

Distributed Road Surface Condition Monitoring Using Mobile Phones

Applying the Mahalanobis-Taguchi strategy for software defect diagnosis

Contact Info

Product

Resources

About