Benchmarking Attribute Selection Techniques for Discrete Class Data Mining

By Mark A. Hall, G. Holmes. IEEE Transactions on Knowledge and Data Engineering, 2003

1,313 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/92f160c2ccf58c8098a062abae5f3c99c6951aa0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Flood Detection and Susceptibility Mapping Using Sentinel-1 Remote Sensing Data and a Machine Learning Approach: Hybrid Intelligence of Bagging Ensemble Based on K-Nearest Neighbor Classifier (2020, methodology): “RAE is a distance-based attribute/factor ranking approach proposed by Kira and Rendell [124], and later improved by Kononenko [125] and Hall and Holmes [126].”
StressSense: detecting stress in unconstrained acoustic environments using smartphones (2012, methodology): “In the ﬁrst experiment, we measure the relevance of the different features using information gain based feature ranking [35].”
Comparing and experimenting machine learning techniques for code smell detection (2015, methodology): “The last two best algorithms are B-Naïve Bayes and B-Random Forest with 19 victories.”
A survey of techniques for internet traffic classification using machine learning (2008, background): “(Additional details can be found in [29] [40] [41] [42] [43].”
Data Mining Static Code Attributes to Learn Defect Predictors (2007, background): “The baseline definition draws from standard practices in the data mining community [23], [24].”
A General Software Defect-Proneness Prediction Framework (2011, methodology): “For example, Hall and Holmes [32] concluded that the forward selection (FS) search was well suited to Naive Bayes but the backward elimination (BE) search is more suitable for C4.5.”
Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease: a 5-year multicentre prospective registry analysis (2016, methodology): “Machine learning involved automated feature selection by information gain ranking, model building with a boosted ensemble algorithm, and 10-fold stratified cross-validation for the entire process (Figure 2).(11) – 14 Machine learning techniques were implemented in the open-source Waikato Environment”
Evaluating defect prediction approaches: a benchmark and an extensive comparison (2011, background): “However, according to a study of Hall and Holmes (2003) , who performed an extensive comparison of attribute selection techniques, PCA is one of the worst performers.”
Incremental wrapper-based gene selection from microarray data for cancer classification (2006, neutral): “In [18], a rank search method is proposed to compare feature selection algorithms.”
Easy over hard: a case study on deep learning (2017, neutral): “More complex methods: • New feature selection[1].”
Benchmarking Classification Models for Software Defect Prediction: A Proposed Framework and Novel Findings (2008, methodology): “Especially, data preprocessing or engineering activities such as the removal of noninformative features or the discretization of continuous attributes may improve the performance of some classifiers (see, e.g., [13], [25]).”
Financial Statement Fraud Detection: An Analysis of Statistical and Machine Learning Algorithms (2011, neutral): “To evaluate the utility of the attributes, a wrapper attribute selection technique was used that has been shown in prior research to be effective (Hall and Holmes 2003).”
Stability of feature selection algorithms: a study on high-dimensional spaces (2005, neutral): “Uncertainty (SYM), [ 9 ], ReliefF (RELIEF), [15], and SVMRFE [8].”
A feature selection model based on genetic rank aggregation for text sentiment classification (2017, methodology): “For each attribute A i , the information gain between the attribute and the class is determined as given by equation (3) [25]:”