SEARCH · PubMed Health

Results for “feature selection”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources

PubMed2005

Classification of gasoline data obtained by gas chromatography using a piecewise alignment algorithm combined with feature selection and principal component analysis.

A fast and objective chemometric classification method is developed and applied to the analysis of gas chromatography (GC) data from five commercial gasoline samples. The gasoline samples serve as model mixtures, whereas the focus is on the development and demonstration of the classification method. The method is based on objective retention time alignment (referred to as piecewise alignment) coupled with analysis of variance (ANOVA) feature selection prior to classification by principal component analysis (PCA) using optimal parameters. The degree-of-class-separation is used as a metric to objectively optimize the alignment and feature selection parameters using a suitable training set thereby reducing user subjectivity, as well as to indicate the success of the PCA clustering and classification. The degree-of-class-separation is calculated using Euclidean distances between the PCA scores of a subset of the replicate runs from two of the five fuel types, i.e., the training set. The unaligned training set that was directly submitted to PCA had a low degree-of-class-separation (0.4), and the PCA scores plot for the raw training set combined with the raw test set failed to correctly cluster the five sample types. After submitting the training set to piecewise alignment, the degree-of-class-separation increased (1.2), but when the same alignment parameters were applied to the training set combined with the test set, the scores plot clustering still did not yield five distinct groups. Applying feature selection to the unaligned training set increased the degree-of-class-separation (4.8), but chemical variations were still obscured by retention time variation and when the same feature selection conditions were used for the training set combined with the test set, only one of the five fuels was clustered correctly. However, piecewise alignment coupled with feature selection yielded a reasonably optimal degree-of-class-separation for the training set (9.2), and when the same alignment and ANOVA parameters were applied to the training set combined with the test set, the PCA scores plot correctly classified the gasoline fingerprints into five distinct clusters.

Algorithms↗

PubMed2000

Feature selection and classifier performance in computer-aided diagnosis: the effect of finite sample size.

In computer-aided diagnosis (CAD), a frequently used approach for distinguishing normal and abnormal cases is first to extract potentially useful features for the classification task. Effective features are then selected from this entire pool of available features. Finally, a classifier is designed using the selected features. In this study, we investigated the effect of finite sample size on classification accuracy when classifier design involves stepwise feature selection in linear discriminant analysis, which is the most commonly used feature selection algorithm for linear classifiers. The feature selection and the classifier coefficient estimation steps were considered to be cascading stages in the classifier design process. We compared the performance of the classifier when feature selection was performed on the design samples alone and on the entire set of available samples, which consisted of design and test samples. The area Az under the receiver operating characteristic curve was used as our performance measure. After linear classifier coefficient estimation using the design samples, we studied the hold-out and resubstitution performance estimates. The two classes were assumed to have multidimensional Gaussian distributions, with a large number of features available for feature selection. We investigated the dependence of feature selection performance on the covariance matrices and means for the two classes, and examined the effects of sample size, number of available features, and parameters of stepwise feature selection on classifier bias. Our results indicated that the resubstitution estimate was always optimistically biased, except in cases where the parameters of stepwise feature selection were chosen such that too few features were selected by the stepwise procedure. When feature selection was performed using only the design samples, the hold-out estimate was always pessimistically biased. When feature selection was performed using the entire finite sample space, the hold-out estimates could be pessimistically or optimistically biased, depending on the number of features available for selection, the number of available samples, and their statistical distribution. For our simulation conditions, these estimates were always pessimistically (conservatively) biased if the ratio of the total number of available samples per class to the number of available features was greater than five.

Algorithms↗

PubMedDate not supplied

Implementing the Fisher's discriminant ratio in a k-means clustering algorithm for feature selection and data set trimming.

The Fisher's discriminant ratio has been used as a class separability criterion and implemented in a k-means clustering algorithm for performing simultaneous feature selection and data set trimming on a set of 221 HIV-1 protease inhibitors. The total number of molecular descriptors computed for each inhibitor is 43, and they are scaled to lie between 1 and 0 before being subjected to the feature selection process. Since the purpose is to select some of the most class sensitive descriptors, several feature evaluation indices such as the Shannon entropy, the linear regression of selected descriptors on the pKi of selected inhibitors, and a stepwise variable selection program are used to filter them. While the Shannon entropy provides the information content for each descriptor computed, more class sensitive descriptors are searched by both the linear regression and stepwise variable selection procedures. The inhibitors are divided into several different numbers of classes. They are subsequently divided into five classes due to the fact that the best feature selection result is obtained by the division. Most of the good features selected are the topological descriptors, and they are correlated well with the pKi values. The outliers or the inhibitors with less class-sensitive descriptor values computed for each selected descriptor are identified and gathered by the k-means clustering algorithm. These are the trimmed inhibitors, while the remaining ones are retained or selected. We find that 44% or 98 inhibitors can be retained when the number of good descriptors selected for clustering is three. The descriptor values of these selected inhibitors are far more class sensitive than the original ones as evidenced by substantial increasing in statistical significance when they are subjected to both the SYBYL CoMFA PLS and Cerius2 PLS regression analyses.

Journal Article↗

PubMed1998

Feature selection for genetic sequence classification.

MOTIVATION: Most of the existing methods for genetic sequence classification are based on a computer search for homologies in nucleotide or amino acid sequences. The standard sequence alignment programs scale very poorly as the number of sequences increases or the degree of sequence identity is <30%. Some new computationally inexpensive methods based on nucleotide or amino acid compositional analysis have been proposed, but prediction results are still unsatisfactory and depend on the features chosen to represent the sequences. RESULTS: In this paper, a feature selection method based on the Gamma (or near-neighbour) test is proposed. If there is a continuous or smooth map from feature space to the classification target values, the Gamma test gives an estimate for the mean-squared error of the classification, despite the fact that one has no a priori knowledge of the smooth mapping. We can search a large space of possible feature combinations for a combination which gives a smallest estimated mean-squared error using a genetic algorithm. The method was used for feature selection and classification of the large subunits of rRNA according to RDP (Ribosomal Database Project) phylogenetic classes. The sequences were represented by dinucleotide frequency distribution. The nearest-neighbour criterion has been used to estimate the predictive accuracy of the classification based on the selected features. For examples discussed, we found that the classification according to the first nearest neighbour is correct for 80% of the test samples. If we consider the set of the 10 nearest neighbours, then 94% of the test samples are classified correctly. AVAILABILITY: The principal novel component of this method is the Gamma test and this can be downloaded compiled for Unix Sun 4, Windows 95 and MS-DOS from http://www.cs.cf.ac.uk/ec/ CONTACT: s.margetts@cs.cf.ac.uk

Algorithms↗

PubMed2006

Differential prioritization between relevance and redundancy in correlation-based feature selection techniques for multiclass gene expression data.

BACKGROUND: Due to the large number of genes in a typical microarray dataset, feature selection looks set to play an important role in reducing noise and computational cost in gene expression-based tissue classification while improving accuracy at the same time. Surprisingly, this does not appear to be the case for all multiclass microarray datasets. The reason is that many feature selection techniques applied on microarray datasets are either rank-based and hence do not take into account correlations between genes, or are wrapper-based, which require high computational cost, and often yield difficult-to-reproduce results. In studies where correlations between genes are considered, attempts to establish the merit of the proposed techniques are hampered by evaluation procedures which are less than meticulous, resulting in overly optimistic estimates of accuracy. RESULTS: We present two realistically evaluated correlation-based feature selection techniques which incorporate, in addition to the two existing criteria involved in forming a predictor set (relevance and redundancy), a third criterion called the degree of differential prioritization (DDP). DDP functions as a parameter to strike the balance between relevance and redundancy, providing our techniques with the novel ability to differentially prioritize the optimization of relevance against redundancy (and vice versa). This ability proves useful in producing optimal classification accuracy while using reasonably small predictor set sizes for nine well-known multiclass microarray datasets. CONCLUSION: For multiclass microarray datasets, especially the GCM and NCI60 datasets, DDP enables our filter-based techniques to produce accuracies better than those reported in previous studies which employed similarly realistic evaluation procedures.

Animals↗

PubMed1999

Feature selection for optimized skin tumor recognition using genetic algorithms.

In this paper, a new approach to computer supported diagnosis of skin tumors in dermatology is presented. High resolution skin surface profiles are analyzed to recognize malignant melanomas and nevocytic nevi (moles), automatically. In the first step, several types of features are extracted by 2D image analysis methods characterizing the structure of skin surface profiles: texture features based on cooccurrence matrices, Fourier features and fractal features. Then, feature selection algorithms are applied to determine suitable feature subsets for the recognition process. Feature selection is described as an optimization problem and several approaches including heuristic strategies, greedy and genetic algorithms are compared. As quality measure for feature subsets, the classification rate of the nearest neighbor classifier computed with the leaving-one-out method is used. Genetic algorithms show the best results. Finally, neural networks with error back-propagation as learning paradigm are trained using the selected feature sets. Different network topologies, learning parameters and pruning algorithms are investigated to optimize the classification performance of the neural classifiers. With the optimized recognition system a classification performance of 97.7% is achieved.

Algorithms↗

PubMed2004

A comparative study of feature selection and multiclass classification methods for tissue classification based on gene expression.

This paper studies the problem of building multiclass classifiers for tissue classification based on gene expression. The recent development of microarray technologies has enabled biologists to quantify gene expression of tens of thousands of genes in a single experiment. Biologists have begun collecting gene expression for a large number of samples. One of the urgent issues in the use of microarray data is to develop methods for characterizing samples based on their gene expression. The most basic step in the research direction is binary sample classification, which has been studied extensively over the past few years. This paper investigates the next step-multiclass classification of samples based on gene expression. The characteristics of expression data (e.g. large number of genes with small sample size) makes the classification problem more challenging. The process of building multiclass classifiers is divided into two components: (i) selection of the features (i.e. genes) to be used for training and testing and (ii) selection of the classification method. This paper compares various feature selection methods as well as various state-of-the-art classification methods on various multiclass gene expression datasets. Our study indicates that multiclass classification problem is much more difficult than the binary one for the gene expression datasets. The difficulty lies in the fact that the data are of high dimensionality and that the sample size is small. The classification accuracy appears to degrade very rapidly as the number of classes increases. In particular, the accuracy was very low regardless of the choices of the methods for large-class datasets (e.g. NCI60 and GCM). While increasing the number of samples is a plausible solution to the problem of accuracy degradation, it is important to develop algorithms that are able to analyze effectively multiple-class expression data for these special datasets.

Algorithms↗

PubMed2003

Tumor classification and marker gene prediction by feature selection and fuzzy c-means clustering using microarray data.

BACKGROUND: Using DNA microarrays, we have developed two novel models for tumor classification and target gene prediction. First, gene expression profiles are summarized by optimally selected Self-Organizing Maps (SOMs), followed by tumor sample classification by Fuzzy C-means clustering. Then, the prediction of marker genes is accomplished by either manual feature selection (visualizing the weighted/mean SOM component plane) or automatic feature selection (by pair-wise Fisher's linear discriminant). RESULTS: The proposed models were tested on four published datasets: (1) Leukemia (2) Colon cancer (3) Brain tumors and (4) NCI cancer cell lines. The models gave class prediction with markedly reduced error rates compared to other class prediction approaches, and the importance of feature selection on microarray data analysis was also emphasized. CONCLUSIONS: Our models identify marker genes with predictive potential, often better than other available methods in the literature. The models are potentially useful for medical diagnostics and may reveal some insights into cancer classification. Additionally, we illustrated two limitations in tumor classification from microarray data related to the biology underlying the data, in terms of (1) the class size of data, and (2) the internal structure of classes. These limitations are not specific for the classification models used.

Biomarkers, Tumor↗

PubMed2005

Stimulus-dependent changes in spike threshold enhance feature selectivity in rat barrel cortex neurons.

Feature selectivity is a fundamental property of sensory cortex neurons, yet the mechanisms underlying its genesis are not fully understood. Using intracellular recordings in vivo from layers 2-6 of rat barrel cortex, we studied the selectivity of neurons to the angular direction of whisker deflection. The spike output and the underlying synaptic response decreased exponentially in magnitude as the direction of deflection diverged from the preferred. However, the spike output was more sharply tuned for direction than the underlying synaptic response amplitude. This difference in selectivity was attributable to the rectification imposed by the spike threshold on the input-output function of cells. As in the visual system, spike threshold was not constant and showed trial-to-trial variability. However, here we show that the mean spike threshold was direction dependent and increased as the direction diverged from the preferred. Spike threshold was also related to the rate of rise of the synaptic response, which was direction dependent and steepest for the preferred direction. To assess the impact of the direction-dependent changes in spike threshold on direction selectivity, we applied a fixed threshold to the synaptic responses and calculated a predicted spike output. The predicted output was more broadly tuned than the obtained spike response, demonstrating for the first time that the regulation of the spike threshold by the properties of the synaptic response effectively enhances the selectivity of the spike output.

Action Potentials↗

PubMed2005

Texture analysis for tissue discrimination on T1-weighted MR images of the knee joint in a multicenter study: Transferability of texture features and comparison of feature selection methods and classifiers.

PURPOSE: To investigate the reproducibility and transferability of texture features between MR centers, and to compare two feature selection methods and two classifiers. MATERIALS AND METHODS: Coronal T1-weighted MR images of the knees of 63 patients, divided into three groups, were included in the study. MR images were obtained at three different MR centers. Regions of interest (ROIs) were drawn in the bone marrow and fat tissue. Then texture analysis (TA) of the ROIs was performed, and the most discriminant features were identified using Fisher coefficients and POE+ACC (probability of classification error and average correlation coefficients). Based on these features, artificial neural network (ANN) and k-nearest-neighbor (k-NN) classifiers were used for tissue discrimination. RESULTS: Although the texture features differed among the MR centers, features from one center could be successfully used for tissue discrimination in texture data on MR images from other centers. The best results were achieved using the ANN classifier in combination with features selected by POE+ACC. CONCLUSION: The differences in texture features extracted from MR images from different centers seem to have only a small impact on the results of tissue discrimination.

Adipose Tissue↗

PubMed2005

Feature selection and classification for microarray data analysis: evolutionary methods for identifying predictive genes.

BACKGROUND: In the clinical context, samples assayed by microarray are often classified by cell line or tumour type and it is of interest to discover a set of genes that can be used as class predictors. The leukemia dataset of Golub et al. 1 and the NCI60 dataset of Ross et al. 2 present multiclass classification problems where three tumour types and nine cell lines respectively must be identified. We apply an evolutionary algorithm to identify the near-optimal set of predictive genes that classify the data. We also examine the initial gene selection step whereby the most informative genes are selected from the genes assayed. RESULTS: In the absence of feature selection, classification accuracy on the training data is typically good, but not replicated on the testing data. Gene selection using the RankGene software 3 is shown to significantly improve performance on the testing data. Further, we show that the choice of feature selection criteria can have a significant effect on accuracy. The evolutionary algorithm is shown to perform stably across the space of possible parameter settings - indicating the robustness of the approach. We assess performance using a low variance estimation technique, and present an analysis of the genes most often selected as predictors. CONCLUSION: The computational methods we have developed perform robustly and accurately, and yield results in accord with clinical knowledge: A Z-score analysis of the genes most frequently selected identifies genes known to discriminate AML and Pre-T ALL leukemia. This study also confirms that significantly different sets of genes are found to be most discriminatory as the sample classes are refined.

Algorithms↗

PubMed2005

Automated neonatal seizure detection: a multistage classification system through feature selection based on relevance and redundancy analysis.

OBJECTIVE: Automatic seizure detection obtains valuable information concerning duration and timing of seizures. Commonly used methods for EEG seizure detection in adults are inadequate for the same task in neonates because they lack the specific age-dependant characteristics of normal and pathological EEG. This paper presents an automatic seizure detection system for newborn with focus on feature selection via relevance and redundancy analysis. METHODS: Two linear correlation-based feature selection methods and the ReliefF method were applied to parameterized EEG data acquired from six neonates aged between 39 and 42 weeks. To evaluate the effectiveness of these methods, features extracted from seizure and non-seizure segments were ranked by these methods. The optimized ranked feature subsets were fed into a backpropagation neural network for classifying. Its performance was used as indicator for the feature selection effectiveness. RESULTS: Results showed an average seizure detection rate of 91%, an average non-seizure detection rate of 95%, an average false rejection rate of 95% and an overall average detection rate of 93% with a false seizure detection rate of 1.17/h. CONCLUSIONS: This good performance in detecting newborn ictal activities has been achieved based on an optimized subset of 30 features determined by the ReliefF-based detector, which corresponds to a reduction of the number of features of up to 75%. SIGNIFICANCE: The presented approach takes into account specific characteristics of normal and pathological EEG. Thus, it can improve the accuracy of conventional seizure detection systems in newborn.

Algorithms↗

PubMed2002

Feature selection for structure-activity correlation using binary particle swarms.

We present a new feature selection algorithm for structure-activity and structure-property correlation based on particle swarms. Particle swarms explore the search space through a population of individuals that adapt by returning stochastically toward previously successful regions, influenced by the success of their neighbors. This method, which was originally intended for searching multidimensional continuous spaces, is adapted to the problem of feature selection by viewing the location vectors of the particles as probabilities and employing roulette wheel selection to construct candidate subsets. The algorithm is applied in the construction of parsimonious quantitative structure-activity relationship (QSAR) models based on feed-forward neural networks and is tested on three classical data sets from the QSAR literature. It is shown that the method compares favorably with simulated annealing and is able to identify a better and more diverse set of solutions given the same amount of simulation time.

Algorithms↗

PubMed2002

Using feature generation and feature selection for accurate prediction of translation initiation sites.

Correct prediction of the translation initiation site (TIS) is an important issue in genomic research. We show that feature generation together with correlation based feature selection can be used with a variety of machine learning algorithms to give highly accurate translation initiation site prediction. Only very few features are needed and the results achieve comparable accuracy to the best existing approaches. Our approach has the advantage that it does not require one to devise a special prediction method; rather standard machine learning classifiers are shown to give very good performance on the selected features. The raw and generated features which we have found to be important are the following: positions -3 and -1 in the sequence; upstream k-grams for k=3, 4, and 5; stop-codon frequency; downstream in-frame 3-gram; and the distance of ATG to the beginning of the sequence. The best result, with an overall accuracy of 90%, is obtained by selecting only seven features from this set. The same features retrained with the use of a scanning model achieves an overall accuracy of 94% on this dataset.

Codon, Initiator↗

PubMed2005

Dejittered spike-conditioned stimulus waveforms yield improved estimates of neuronal feature selectivity and spike-timing precision of sensory interneurons.

What is the meaning associated with a single action potential in a neural spike train? The answer depends on the way the question is formulated. One general approach toward formulating this question involves estimating the average stimulus waveform preceding spikes in a spike train. Many different algorithms have been used to obtain such estimates, ranging from spike-triggered averaging of stimuli to correlation-based extraction of "stimulus-reconstruction" kernels or spatiotemporal receptive fields. We demonstrate that all of these approaches miscalculate the stimulus feature selectivity of a neuron. Their errors arise from the manner in which the stimulus waveforms are aligned to one another during the calculations. Specifically, the waveform segments are locked to the precise time of spike occurrence, ignoring the intrinsic "jitter" in the stimulus-to-spike latency. We present an algorithm that takes this jitter into account. "Dejittered" estimates of the feature selectivity of a neuron are more accurate (i.e., provide a better estimate of the mean waveform eliciting a spike) and more precise (i.e., have smaller variance around that waveform) than estimates obtained using standard techniques. Moreover, this approach yields an explicit measure of spike-timing precision. We applied this technique to study feature selectivity and spike-timing precision in two types of sensory interneurons in the cricket cercal system. The dejittered estimates of the mean stimulus waveforms preceding spikes were up to three times larger than estimates based on the standard techniques used in previous studies and had power that extended into higher-frequency ranges. Spike timing precision was approximately 5 ms.

Action Potentials↗

PubMedDate not supplied

Effect of molecular descriptor feature selection in support vector machine classification of pharmacokinetic and toxicological properties of chemical agents.

Statistical-learning methods have been developed for facilitating the prediction of pharmacokinetic and toxicological properties of chemical agents. These methods employ a variety of molecular descriptors to characterize structural and physicochemical properties of molecules. Some of these descriptors are specifically designed for the study of a particular type of properties or agents, and their use for other properties or agents might generate noise and affect the prediction accuracy of a statistical learning system. This work examines to what extent the reduction of this noise can improve the prediction accuracy of a statistical learning system. A feature selection method, recursive feature elimination (RFE), is used to automatically select molecular descriptors for support vector machines (SVM) prediction of P-glycoprotein substrates (P-gp), human intestinal absorption of molecules (HIA), and agents that cause torsades de pointes (TdP), a rare but serious side effect. RFE significantly reduces the number of descriptors for each of these properties thereby increasing the computational speed for their classification. The SVM prediction accuracies of P-gp and HIA are substantially increased and that of TdP remains unchanged by RFE. These prediction accuracies are comparable to those of earlier studies derived from a selective set of descriptors. Our study suggests that molecular feature selection is useful for improving the speed and, in some cases, the accuracy of statistical learning methods for the prediction of pharmacokinetic and toxicological properties of chemical agents.

Algorithms↗

PubMed2005

FrankSum: new feature selection method for protein function prediction.

In the study of in silico functional genomics, improving the performance of protein function prediction is the ultimate goal for identifying proteins associated with defined cellular functions. The classical prediction approach is to employ pairwise sequence alignments. However this method often faces difficulties when no statistically significant homologous sequences are identified. An alternative way is to predict protein function from sequence-derived features using machine learning. In this case the choice of possible features which can be derived from the sequence is of vital importance to ensure adequate discrimination to predict function. In this paper we have successfully selected biologically significant features for protein function prediction. This was performed using a new feature selection method (FrankSum) that avoids data distribution assumptions, uses a data independent measurement (p-value) within the feature, identifies redundancy between features and uses an appropriate ranking criterion for feature selection. We have shown that classifiers generated from features selected by FrankSum outperforms classifiers generated from full feature sets, randomly selected features and features selected from the Wrapper method. We have also shown the features are concordant across all species and top ranking features are biologically informative. We conclude that feature selection is vital for successful protein function prediction and FrankSum is one of the feature selection methods that can be applied successfully to such a domain.

Amino Acid Sequence↗

PubMed2003

On combining multiple microarray studies for improved functional classification by whole-dataset feature selection.

As microarray technologies become routinely applied in genome laboratories for studying gene expression, it is not uncommon that experiments on identical or similar sets of genes are conducted by multiple laboratories for various functional studies of these genes. Much of such data are often available to researchers for their data analysis, either through collaborators or from online gene expression databases. It will be useful to combine data from different microarray studies to improve the microarray data mining results. We show that the functional classification of genes from microarray data can be improved further by combining gene expression data from multiple microarray studies, even if the experimental focus or conditions for each experimental study may differ. However, blindly combining all available datasets may not always improve the analysis results---it is important to be selective of the datasets for inclusion. In our approach, we consider each dataset to be one feature, and then apply feature selection strategies to select appropriate datasets for training. With a simple hill-climbing method, we show that gene classification performances can be improved by whole-dataset feature selection.

Algorithms↗