Stability Of Feature Selection Algorithms Research Articles

A new hybrid stability measure for feature selection

Feature Selection (FS) algorithms are applied in bioinformatics applications to identify the disease causing genes. Performance of such algorithms is measured in terms of accuracy of the model and stability of FS algorithms. Stability evaluates the identical replication of feature sets obtained after every execution. Recently research has shown that a stability measure must satisfy set of properties like, fully defined, monotonicity, boundedness, deterministic maximum stability, and correction for chance. Among the existing stability measures, only Nogueira’s frequency based stability measure satisfies all the required properties. However, frequency based stability measures fail to discriminate among the cases when overall frequency of features are same. In order to address this issue, the paper proposes a hybrid similarity based stability measure which satisfies all the desirable properties, as mentioned earlier. The proposed stability measure is unique as it is the first similarity based stability measure that satisfies all the required properties. Also, all these essential properties are mathematically established. Further, the paper also proposes a combination of frequency based and similarity based measure which preserves all the aspects of both the approaches. The work presented also analyzes the stability performance of LASSO and Elastic Net, using synthetic and microarray gene expression datasets. Elastic Net depicts higher stability and selection of relevant features.

Stability of Indexed Microarray and Text Data

The common challenge for machine learning and data mining tasks is the curse of High Dimensionality. Feature selection reduces the dimensionality by selecting the relevant and optimal features from the huge dataset. In this research work, a clustering and genetic algorithm based feature selection (CLUST-GA-FS) is proposed that has three stages namely irrelevant feature removal, redundant feature removal, and optimal feature generation. The performance of the feature selection algorithms are analyzed using the parameters like classification accuracy, precision, recall and error rate. Recently, an increasing attention is given to the stability of feature selection algorithms which is an indicator that requires that similar subsets of features are selected every time the algorithm is executed on the same dataset. This work analyzes the stability of the algorithm on four publicly available dataset using stability measurements Average normal hamming distance(ANHD), Dices coefficient, Tanimoto distance, Jaccards index and Kuncheva index.

On the stability of feature selection algorithms

Feature Selection is central to modern data science, from exploratory data analysis to predictive model-building. The stability of a feature selection algorithm refers to the robustness of its fe...

A variance reduction framework for stable feature selection

AbstractStability of feature selection is an important but under‐addressed issue in knowledge discovery from high‐dimensional data. In this study, we present a theoretical framework about the relationship between the stability and the accuracy of feature selection based on a formal bias‐variance decomposition of feature selection error. The framework also reveals the connection between stability and sample size and suggests a variance reduction approach for improving the stability of feature selection algorithms under small sample size. Following the theoretical framework, we propose an empirical variance reduction framework, margin‐based instance weighting, which weights training instances according to their importance to feature evaluation. Our extensive experimental study first verifies the theoretical and empirical frameworks based on synthetic data sets and a popular feature selection algorithm SVM‐RFE. Experiments based on real‐world microarray data sets further verify that the empirical framework is effective at reducing the variance and improving the subset stability of two representative feature selection algorithms, SVM‐RFE and ReliefF, while maintaining comparable predictive accuracy based on the selected features. The proposed instance weighting framework is also shown to be more effective and efficient than the ensemble framework at improving the subset stability of the feature selection algorithms under study. © 2012 Wiley Periodicals, Inc. Statistical Analysis and Data Mining, 2012

Stability of feature selection algorithms: a study on high-dimensional spaces

With the proliferation of extremely high-dimensional data, feature selection algorithms have become indispensable components of the learning process. Strangely, despite extensive work on the stability of learning algorithms, the stability of feature selection algorithms has been relatively neglected. This study is an attempt to fill that gap by quantifying the sensitivity of feature selection algorithms to variations in the training set. We assess the stability of feature selection algorithms based on the stability of the feature preferences that they express in the form of weights-scores, ranks, or a selected feature subset. We examine a number of measures to quantify the stability of feature preferences and propose an empirical way to estimate them. We perform a series of experiments with several feature selection algorithms on a set of proteomics datasets. The experiments allow us to explore the merits of each stability measure and create stability profiles of the feature selection algorithms. Finally, we show how stability profiles can support the choice of a feature selection algorithm.