Dimensionality reduction using genetic algorithms

By Michael L. Raymer, William F. Punch III, Erik D. Goodman, Leslie A. Kuhn, Anil K. Jain. IEEE Transactions on Evolutionary Computation, 2000

900 citations recorded by the source corpus.

Pattern recognition generally requires that objects be described in terms of a set of measurable features. The selection and quality of the features representing each pattern affect the success of subsequent classification. Feature extraction is the process of deriving new features from original features to reduce the cost of feature measurement, increase classifier efficiency, and allow higher accuracy. Many feature extraction techniques involve linear transformations of the original pattern vectors to new vectors of lower dimensionality. While this is useful for data visualization and classification efficiency, it does not necessarily reduce the number of features to be measured since each new feature may be a linear combination of all of the features in the original pattern vector. Here, we present a new approach to feature extraction in which feature selection and extraction and classifier training are performed simultaneously using a genetic algorithm. The genetic algorithm optimizes a feature weight vector used to scale the individual features in the original pattern vectors. A masking vector is also employed for simultaneous selection of a feature subset. We employ this technique in combination with the k nearest neighbor classification rule, and compare the results with classical feature selection and extraction techniques, including sequential floating forward feature selection, and linear discriminant analysis. We also present results for the identification of favorable water-binding sites on protein surfaces.

Subfields: Artificial Intelligence, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/a81cf97778664a960e4f2770bc8c703f8a1afa1b

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 Molecular Biology

Citations in the corpus

Feature selection for support vector machines by means of genetic algorithm (2003, methodology): “RicLan96, RayPunGooKuhJai00 ]. As in this thesis SVMs are used for classi-”
Normalized Mutual Information Feature Selection (2009, neutral): “Genetic algorithms (GAs) have been successfully applied to feature selection [22], [23].”
Hybrid Genetic Algorithms for Feature Selection (2004, background): “Subsequently, many literatures were published that have shown advantages of the GAs for feature selection [1], [30], [13], [27].”
Improved salp swarm algorithm based on particle swarm optimization for feature selection (2018, methodology): “A variety of search techniques have been applied to feature selection, such as complete search, greedy search, heuristic search (Zhong 2001), and random search (Lai et al. 2006; Raymer et al. 2000).”
Feature selection methods for big data bioinformatics: A survey from the search perspective. (2016, methodology): “They applied LLDA-RFE to several public data sets of cancer microarrays and compared its performance with those of Laplacian score and SVD-entropy, two state-of-theart unsupervised methods, and with that of Fisher score, a supervised filter method.”
Wrapper–Filter Feature Selection Algorithm Using a Memetic Framework (2007, methodology): “In this paper, we propose a novel wrapper-filter feature selection algorithm (WFFSA) using a memetic framework [11-16], i.e., a combination of genetic algorithm (GA) [ 17-20 ] and local search (LS).”
A rough set approach to feature selection based on ant colony optimization (2010, methodology): “, 2004), heuristic (Zhong and Dong, 2001), random (Raymer et al., 2000; Lai et al., 2006) strategies were proposed.”
A Mixture Model for Clustering Ensembles (2004, methodology): “either conserved or non-conserved type of molecules in the bound structure of proteins [1, 28].”
A GA-based feature selection and parameters optimizationfor support vector machines (2006, positive): “Some other GA-based feature selection methods were proposed ( Raymer, Punch, Goodman, Kuhn, & Jain, 2000; Yang and Honavar, 1998; Salcedo-Sanz, Prado-Cumplido, Pe´rez-Cruz, & Bouson˜o-Calzo´n, 2002).”
Clustering ensembles: models of consensus and weak partitions (2005, methodology): “Moreover, the clustering error obtained by EM and MCLA algorithms with k 1⁄4 4 for “Biochemistry” data [1] was the same as found by supervised classifiers applied to this data set [45].”
Markov blanket-embedded genetic algorithm for gene selection (2007, background): “It has been naturally used for feature selection (or in the case of microarray data, gene selection) and has shown promising performance [6,7,11,17–20,28,29] (for a survey of GA based methods on microarray analysis, the reader is referred to [15]).”
A hybrid genetic algorithm for feature selection wrapper based on mutual information (2007, background): “…(Bishop, 1995), hill-climbing (Caruana and Freitag, 1994), branch and bound algorithms (Somol et al., 2004), and the stochastic algorithms like simulated annealing (Doak, 1992) and genetic algorithms (GAs) (Bhanu and Lin, 2003; Zhu and Guan, 2004; Raymer et al., 2000; Yang and Honavar, 1998).”
A discrete particle swarm optimization method for feature selection in binary classification problems (2010, background): “Raymer et al. (2000) propose a GA which simultaneously performs feature selection and extraction.”
A Survey of Dimension Reduction Techniques (2002, methodology): “For example, [45] use GAs in combination with a k-nearest neighbor (knn) classifier to reduce the dimension of a feature set: starting with a population of random transformation matrices {Wk×p} , they use GAs to find the transformation Wk×p such that the knn classifier using the new features Sk×n = ”
Dimensionality Reduction: A Comparative Review (2008, methodology): “backpropagation. As an alternative approach, genetic algorithms [ 57 ] can be used to train an”
ActiveTrust: Secure and Trustable Routing in Wireless Sensor Networks (2016, background): “com 132 following sections will discuss the architecture of WSN and related issues and challenges in it [5].”