PCA-Based Feature Selection Scheme for Machine Defect Classification

Malhi, A.; Gao, Robert X.

doi:10.1109/tim.2004.834070

Cited by 516 publications

(257 citation statements)

References 17 publications

Supporting

Mentioning

248

Contrasting

Unclassified

Order By: Relevance

“…Principal Components Analysis (PCA): The PCA approach was developed to reduce the dimensionality of the input features for both supervised and unsupervised classification purposes by Malhi et al [16]. As they pointed out that the PCA technique transforms n vectors ( ) [17], the IG associated with an attribute and a dataset is calculated using a straightforward implementation of Equation (3).…”

Section: Feature Selection Techniquesmentioning

confidence: 99%

A Combined-Learning Based Framework for Improved Software Fault Prediction

Yohannese¹,

Li²

2017

IJCIS

View full text Add to dashboard Cite

Software Fault Prediction (SFP) is found to be vital to predict the fault-proneness of software modules, which allows software engineers to focus development activities on fault-prone modules, thereby prioritize and optimize tests, improve software quality and make better use of resources. In this regard, machine learning has been successfully applied to solve classification problems for SFP. Nevertheless, the presence of different software metrics, the redundant and irrelevant features and the imbalanced nature of software datasets have created more and more challenges for the classification problems. Therefore, the objective of this study is to independently examine software metrics with multiple Feature Selection (FS) combined with Data Balancing (DB) using Synthetic Minority Oversampling Techniques for improving classification performance. Accordingly, a new framework that efficiently handles those challenges in a combined form on both Object Oriented Metrics (OOM) and Static Code Metrics (SCM) datasets is proposed. The experimental results confirm that the prediction performance could be compromised without suitable Feature Selection Techniques (FST). To mitigate that, data must be balanced. Thus our combined technique assures the robust performance. Furthermore, a combination of Random Forts (RF) with Information Gain (IG) FS yields the highest Receiver Operating Characteristic (ROC) curve (0.993) value, which is found to be the best combination when SCM are used, whereas the combination of RF with Correlation-based Feature Selection (CFS) guarantees the highest ROC (0.909) value, which is found to be the best choice when OOM are used. Therefore, as shown in this study, software metrics used to predict the fault proneness of the software modules must be carefully examined and suitable FST for software metrics must be cautiously selected. Moreover, DB must be applied in order to obtain robust performance. In addition to that, dealing with the challenges mentioned above, the proposed framework ensures the remarkable classification performance and lays the pathway to quality assurance of software.

show abstract

Section: Feature Selection Techniquesmentioning

confidence: 99%

A Combined-Learning Based Framework for Improved Software Fault Prediction

Yohannese¹,

Li²

2017

IJCIS

View full text Add to dashboard Cite

show abstract

“…Eqs. (25)(26)(27) are used as follows. First, when two features have a high correlation value, ρ k ( f p , f q ) > 0.9 for all the classes, the one with the larger inter-class variability is discarded.…”

Section: Features Evaluationmentioning

confidence: 99%

“…Moreover, some of these techniques can learn incrementally and adapt to the current defect population. These techniques can be subdivided into three broad categories: soft-computing, [24][25][26][27], support vector machines [28,29] and boosting [30][31][32].…”

Section: Introductionmentioning

confidence: 99%

Automatic Defect Classification on a Production Line

Borghese

Fomasi

2015

Intell Ind Syst

View full text Add to dashboard Cite

We describe here a novel defect classification system that works in real-time on the images of material running on the production line, provided by a video-inspection module. The classifier is constituted of a two-levels hierarchical architecture: a set of adequate features are extracted from the image first; the defect type is then identified from them. Statistical analysis allows greatly reducing the number of features, leaving only the most significant ones. A particular version of multi-class boosting has been developed for labeling: differently from its original version, it accepts only one label for each image, the true one, and does not require the rank for the other classes. Nevertheless, the classifier is able to produce a valid rank of the defect with respect to all the classes, information that can be used to achieve an identification rate of the dangerous defects very close to 100 % on a real data set.

show abstract

“…A common PCA derivation in terms of a standardized linear projection maximizing variance is the projected space. PCA is useful for data compression, reducing dimensions number without information loss.PCA is used to reduce the dimension of the feature vector extracted [17].…”

Section: Principal Component Analysis(pca)mentioning

confidence: 99%