Applying the Mahalanobis–Taguchi System to. Improve Tablet PC Production Processes. Chi-Feng Peng 2,†, Li-Hsing Ho 3,†, Sang-Bing Tsai. The purpose of this paper is to present and analyze the current literature related to developing and improving the Mahalanobis-Taguchi system (MTS) and to. ABSTRACT. The Mahalanobis-Taguchi System is a diagnosis and predictive method for analyzing patterns in multivariate cases. The goal of this study is to.
|Published (Last):||5 June 2004|
|PDF File Size:||4.61 Mb|
|ePub File Size:||10.22 Mb|
|Price:||Free* [*Free Regsitration Required]|
Classification performance results for the modified Naive Bayes classifiers.
MMTS outperforms the benchmarked algorithms especially when the imbalance ratio is greater than Assume there are two classes: The ratio between negative to positive observations left to right columns in Table 1 is representation for the class distribution i.
It has been shown in [ mahalznobis ] that PTM classifier performance outperformed MTS classifier performance; therefore, it has been selected to be benchmarked with the proposed classifier.
The most common used metrics for the systeem of the imbalance data classification performance are andwhere the last one uses weighted importance of the recall and precision controlled bythe default value of is 1which results in better assessment than accuracy metric, but still biased to one class [ 10 ].
Unfortunately, the PTM method is based on previously assumed parameters, and the accuracy of the classification results was less than the benchmarked classifiers this is one of the findings in this research, which will be discussed in Results.
Table of Contents Alerts. Bayes theorem is the center of Naive Bayesian classifier NB in which class conditional independence is assumed. The adaptation of decision tree classifier to suit the imbalance data can be accomplished by adjusting the probabilistic estimate of the tree leaf or developing new trimming approaches [ 14 ].
Unfortunately, the MTS suffers from the lack of a systematic rigorous method for determining the threshold to discriminate between the two classes.
The step of determining the optimal threshold is a critical one for effective MTS classier. Algorithmic level approach solutions are based upon creating a biased algorithm towards positive class.
Del Jesus, and F. This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
A set of data is sampled from both classes. In order to determine if there is a significant difference among the classifiers performances i.
The contribution of this paper mainly is in the area of establishing a reliable and systematic threshold for classification.
Modified Mahalanobis Taguchi System for Imbalance Data Classification
Unfortunately one of the bit falls for using this approach is that it can be computationally expensive [ 30 ]. The problem of treating the applications that have imbalance data with the common classifiers leads to bias in the classification accuracy i.
Literature Review In this section, an overview of the imbalance classification approaches, the Mahalanobis Taguchi System concept, its different areas of applications, weakness points, and its variants is presented.
The criterion for selecting the appropriate features is determined by selecting the features that possess high MD values for the positive observations. Unfortunately, imbalance ratio is not the only reason that causes degradation in classifier performance. It can be seen clearly that the MMTS outperforms the other classifiers.
To achieve an acceptable weld quality, nondestructive weld assessment should be performed.
Based on the above equation, the feature mean gain can be calculated by where is an index that represents the feature,and is the total number of features.
On the other hand, Sun et al.
The most appropriate hyperplane means the one with the largest width of the margin parallel to the hyperplane with no interior points. The border that separates balance from imbalance data is vague; for example, imbalance ratio, which is the ratio between the major to minor class observations, is reported from small values of to 1 to The curve drawn in the figure represents the MTS classifier performance for different threshold values.
If the data distribution of one class is different from distributions of others, then the data is considered imbalance. Table 2 contains a description of the selected datasets properties. In the case of highly imbalanced data, one-class learning showed good classification results [ 28 ].
Computational Intelligence and Neuroscience
Each weld has 28 features, which represents the dynamic resistance value in the 28 half cycles or welding time. If the stopping criteria i. MTS was used previously in predicting weld quality [ 3 ], exploring the mahalanobiz of chemicals constitution on hot rolling manufactured products [ 34 ], and selecting the significant features in automotive handling [ mahalnobis ].
While the problem reported [ 4 ] using the algorithmic approach is that it needs a deep understanding about the classier used itself and the application area i. The experimental setup, the materials used, and all the other related information can be found in the same reference. Mathematically, where is a variable vector of size and is the class. Using 1, the inverse of the correlation matrix, the mean, and the sample standard deviation of taguhci featurefor the negative data, respectively, the MD of the positive observations can be calculated.
The essential classifier performance can be explained by examining the confusion matrix Table 1.