Term Selection Methods Research Articles

An improved binary particle swarm optimization algorithm for clinical cancer biomarker identification in microarray data

Background and ObjectiveThe limited number of samples and high-dimensional features in microarray data make selecting a small number of features for disease diagnosis a challenging problem. Traditional feature selection methods based on evolutionary algorithms are difficult to search for the optimal set of features in a limited time when dealing with the high-dimensional feature selection problem. New solutions are proposed to solve the above problems. MethodsIn this paper, we propose a hybrid feature selection method (C-IFBPFE) for biomarker identification in microarray data, which combines clustering and improved binary particle swarm optimization while incorporating an embedded feature elimination strategy. Firstly, an adaptive redundant feature judgment method based on correlation clustering is proposed for feature screening to reduce the search space in the subsequent stage. Secondly, we propose an improved flipping probability-based binary particle swarm optimization (IFBPSO), better applicable to the binary particle swarm optimization problem. Finally, we also design a new feature elimination (FE) strategy embedded in the binary particle swarm optimization algorithm. This strategy gradually removes poorer features during iterations to reduce the number of features and improve accuracy. ResultsWe compared C-IFBPFE with other published hybrid feature selection methods on eight public datasets and analyzed the impact of each improvement. The proposed method outperforms other current state-of-the-art feature selection methods in terms of accuracy, number of features, sensitivity, and specificity. The ablation study of this method validates the efficacy of each component, especially the proposed feature elimination strategy significantly improves the performance of the algorithm. ConclusionsThe hybrid feature selection method proposed in this paper helps address the issue of high-dimensional microarray data with few samples. It can select a small subset of features and achieve high classification accuracy on microarray datasets. Additionally, independent validation of the selected features shows that those chosen by C-IFBPFE have strong correlations with disease phenotypes and can identify important biomarkers from data related to biomedical problems.

MCDM-EFS: A novel ensemble feature selection method for software defect prediction using multi-criteria decision making

Software defect prediction models are used for predicting high risk software components. Feature selection has significant impact on the prediction performance of the software defect prediction models since redundant and unimportant features make the prediction model more difficult to learn. Ensemble feature selection has recently emerged as a new methodology for enhancing feature selection performance. This paper proposes a new multi-criteria-decision-making (MCDM) based ensemble feature selection (EFS) method. This new method is termed as MCDM-EFS. The proposed method, MCDM-EFS, first generates the decision matrix signifying the feature’s importance score with respect to various existing feature selection methods. Next, the decision matrix is used as the input to well-known MCDM method TOPSIS for assigning a final rank to each feature. The proposed approach is validated by an experimental study for predicting software defects using two classifiers K-nearest neighbor (KNN) and naïve bayes (NB) over five open-source datasets. The predictive performance of the proposed approach is compared with existing feature selection algorithms. Two evaluation metrics – nMCC and G-measure are used to compare predictive performance. The experimental results show that the MCDM-EFS significantly improves the predictive performance of software defect prediction models against other feature selection methods in terms of nMCC as well as G-measure.

Intelligent Selection Method for Gravity-suitable Area Based on Improved Support Vector Machine by the Genetic Algorithm

The gravity-aided inertial navigation system (GAINS) can achieve precise positioning of underwater vehicles in a gravity-suitable area. However, there are generally shortcomings in the existing intelligent suitable area selection methods in terms of selecting gravity feature parameters and learning parameters. In this paper, an intelligent suitable area selection method is proposed based on an improved support vector machine by the genetic algorithm (GA-SVM) to address the aforementioned problems. Firstly, the genetic algorithm (GA) is utilized to independently pick out the optimal feature subset of 15 existing gravity feature parameters and obtain the optimal support vector machine (SVM) learning parameters while eliminating irrelevant redundant features, thus improving the classifier’s performance and generalization ability. Then, the SVM classifier is trained according to the optimal information output by GA, and the accuracy of the test set is 95.5%. Finally, the classifier is utilized to distinguish suitable and unsuitable areas in the application region to evaluate the proposed method’s performance. The TERCOM experiment in the suitable area resulted in an average positioning error of less than 170 m.

CellBRF: a feature selection method for single-cell clustering using cell balance and random forest.

Single-cell RNA sequencing (scRNA-seq) offers a powerful tool to dissect the complexity of biological tissues through cell sub-population identification in combination with clustering approaches. Feature selection is a critical step for improving the accuracy and interpretability of single-cell clustering. Existing feature selection methods underutilize the discriminatory potential of genes across distinct cell types. We hypothesize that incorporating such information could further boost the performance of single cell clustering. We develop CellBRF, a feature selection method that considers genes' relevance to cell types for single-cell clustering. The key idea is to identify genes that are most important for discriminating cell types through random forests guided by predicted cell labels. Moreover, it proposes a class balancing strategy to mitigate the impact of unbalanced cell type distributions on feature importance evaluation. We benchmark CellBRF on 33 scRNA-seq datasets representing diverse biological scenarios and demonstrate that it substantially outperforms state-of-the-art feature selection methods in terms of clustering accuracy and cell neighborhood consistency. Furthermore, we demonstrate the outstanding performance of our selected features through three case studies on cell differentiation stage identification, non-malignant cell subtype identification, and rare cell identification. CellBRF provides a new and effective tool to boost single-cell clustering accuracy. All source codes of CellBRF are freely available at https://github.com/xuyp-csu/CellBRF.

P-178 The interaction between spermatozoa and cumulus cells: a more physiological approach to the selection of good quality spermatozoa for assisted reproduction

Abstract Study question Does selection by cumulus cells (CCs) improve sperm quality in terms of acrosome integrity and mitochondrial functionality compared to standard techniques? Summary answer The sperm-CCs interaction method appears to be effective in selecting sperm with an intact acrosome and functioning mitochondria What is known already In nature, the selection of the best quality sperm is very demanding and involves high morphological qualities and dynamic features. Semen preparation in ART has long been carried out using either the swim-up or density gradient method; although both methods provide motile sperm, they cannot really replicate the physiological selection processes. As a result, several methods have been proposed to mimic the natural selection that occurs in the female reproductive tract, allowing the most competent sperm to provide the paternal contribution to the zygote. However, the literature on these techniques is limited and does not provide consistent conclusions Study design, size, duration Proof-of-principle study conducted at the New Fertility Group (NFG) in Rome and the Department of Molecular and Developmental Medicine at the University of Siena (Italy) from July 2022 to December 2022, including 9 semen samples. Comparison of the CCs selection model with the standard swim-up by assessing mitochondrial and acrosome integrity. In addition, mitochondrial activity was assessed by oxygraphic analysis in spermatozoa exposed to immortalised human granulosa cells (hGL5) compared to standard swim-up sorted spermatozoa Participants/materials, setting, methods The CCs-model was prepared by running two channels of medium, one with CCs (study group) and one with medium alone (control). Spermatozoa were added to each channel and incubated at 37 °C for 1 hour. Moreover, spermatozoa were also added on hGL5 and incubated for 1.5h. The collected spermatozoa were compared with controls for acrosomal integrity, assessed by FITC-labelled PSA, and mitochondrial membrane potential, assessed by Mitotracker. Finally, sperm oxygen uptake was measured by oxygraphic analysis Main results and the role of chance Assessment of sperm acrosome integrity showed a higher percentage of responding acrosomes in the CCs model compared to swim-up and controls (61% ± 0.3; 39% ± 0.4; 19% ± 0.4 p < 0.05). The CCs model also showed 83% ± 0.3 of spermatozoa with active mitochondria compared to 65% ± 0.3 in the control group (p < 0.005). A similar pattern was observed in the swim-up group. Quantitative analysis of mitochondrial membrane potential (MMP) confirmed that CCs and swim-up groups had higher MMP compared to the control group (p < 0.005). Evaluation of mitochondria in the hGL5 treated group showed active mitochondria in 78% ± 0.49 of the sperm compared to 59% ± 0.65 in the control group (p = 0.1). The same profile was detected by fluorescence quantification of MMP in the hGL5 group compared to the control group (p < 0.01). Oxygraphic analysis, as an index of oxidative phosphorylation and consequently of ATP production, was performed on hGL5 exposed spermatozoa, swim-up selected spermatozoa and controls. hGL5 exposed and swim-up selected spermatozoa showed a similar O2 consumption, about 4 times higher than unselected spermatozoa (p < 0.05) Limitations, reasons for caution The limited sample size and the heterogeneity of the samples were the main limitations of this study. Although CC and hGL5 sperm selection models appear to be more physiological, further studies on other parameters such as sperm DNA integrity are needed. Wider implications of the findings Both the CCs model and the hGL5 model appear to be effective in selecting good quality sperm. As these models avoid sperm centrifugation, further research is needed to understand their efficacy and reliability and whether they are a valid alternative to conventional sperm selection methods in terms of IVF outcomes. Trial registration number none

New criteria for wrapper feature selection to enhance bearing fault classification

Classification is a critical task in many fields, including signal processing and data analysis. The accuracy and stability of classification results can be improved by selecting the most relevant features from the data. In this paper, a new criterion for feature selection using wrapper method is proposed, which is based on the evaluation of the classification results according to the accuracy and stability (standard deviation) of each class and the number of selected features. The proposed method is evaluated using Random Forest (RF) and Ant Colony Optimization (ACO) algorithms on a benchmark dataset. Results show that the proposed method outperforms classical feature selection methods in terms of accuracy and stability of classification results, especially for the difficult-to-classify combined damage class. This study demonstrates the effectiveness of the proposed new wrapper feature selection criterion to improve the performance of classification algorithms with higher stability (STD: C1 = 0.5, C2 = 0.8, C3 = 0.6, C4 = 1.8) and better accuracy (average C1 = 98.5%, C2 = 96.6%, C3 = 9.5%, C4 = 93) for the both; the statoric current and the vibration signal compared to other techniques. Machine learning methods had proven their efficiency in time-varying machines fault diagnosis when taking vibration signals and statoric currents extracted features as inputs. However, the use of the both demonstrated a higher robustness and a remarkable superiority.

Efficiency of index-based selection for potential yield in durum wheat [<em>Triticum turgidum</em> (L.) ssp. <em>turgidum convar. durum</em> (Desf.) Mackey] lines

Wheat is a socioeconomically important crop in Algeria. Improving genetic gain of quantitative traits through selection is at the core of every successful breeding program. Selection is usually performed on grain yield, but other agronomically related characteristics can also help increase genetic gain through indirect or multi-trait selection. The objective of this work was to quantify genetic parameters and compare the efficiency of direct, indirect and simultaneous selection methods in terms of predicted genetic values of wheat progenies. For this purpose, 418 F4-derived lines were evaluated for six agronomic traits including heading date, flag leaf area, plant height, number of spikes, thousand kernel weight and grain yield in an augmented block design with three check varieties. Wide genetic variation with moderately high broad-sense heritability were observed for the recorded traits, except for heading date. The results from genetic gain revealed variation in gains for assessed traits and breeding methods employed. The classic index of Smith and Hazel (SHI) demonstrated a similar genetic gain in grain yield compared to gain from direct selection. Generally, the selection-based index showed the highest responses considering all traits simultaneously with a slight inferiority of the SHI index. The coincidence rates among the evaluated indices were higher than those obtained between the measured traits. Based on the comparisons between the selected lines, the SHI index and the selection base index of Williams were similar to grain yield and can reach up to 79.51% coincidence of breeding lines identified by these selection criteria. Breeding lines L252, L34, L24, L130 and L413 were the most common individuals identified according to number of coincidences from the different selection methods used. Of these, L34 and L24, and to a lesser extent L15 can be considered promising wheat lines for improving grain yield.

Feature clustering-Assisted feature selection with differential evolution

Modern data collection technologies may produce thousands of or even more features in a single dataset. The high dimensionality of data poses a barrier to determining discriminating features due to the curse of dimensionality. Thanks to the global search ability, many population-based feature selection approaches have been proposed. However, very few studies pay attention on that a feature selection task has multiple optimal feature subsets. To search for multiple optimal feature subsets, we propose a feature clustering-assisted feature selection method. The proposed method employs the knowledge of correlation measures to group features. And, this correlation knowledge is embedded into the encoding method and the search process. A niching-based mutation operator is also used to explore the vicinity of a target individual. The aim is to find different feature subsets with very similar or the same classification performance. In addition, a modification operator is proposed aiming to increase the population diversity to improve the feature selection performance. The experiments on 16 datasets show that the proposed algorithm outperforms other popular feature selection methods in terms of classification accuracy and feature subset size.

Robust Unsupervised Feature Selection via Multi-Group Adaptive Graph Representation

Unsupervised feature selection can play an important role in addressing the issue of processing massive unlabelled high-dimensional data in the domain of machine learning and data mining. This paper presents a novel unsupervised feature selection method, referred to as Multi-Group Adaptive Graph Representation (MGAGR). Different from existing methods, the relationship between features is explored via the global similarity matrix, which is reconstructed by local similarities of multiple groups. Specifically, the similarity of a feature compared to other features can be represented by the linear combination of all the local similarities. The local similarity of a representative group is given a large weight to reconstruct the global similarity. Besides, an iterative algorithm is given to solve the optimization problem, in which the global similarity matrix, its corresponding reconstruction weights and the self-representation matrix are iteratively improved. Experimental results on 8 benchmark datasets demonstrates that the proposed method outperforms the state-of-the-art unsupervised feature selection methods in terms of clustering performance. The source code is available at: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/misteru/MGAGR</uri> .

Unsupervised Feature Selection via Orthogonal Basis Clustering and Local Structure Preserving.

Due to the "curse of dimensionality" issue, how to discard redundant features and select informative features in high-dimensional data has become a critical problem, hence there are many research studies dedicated to solving this problem. Unsupervised feature selection technique, which does not require any prior category information to conduct with, has gained a prominent place in preprocessing high-dimensional data among all feature selection techniques, and it has been applied to many neural networks and learning systems related applications, e.g., pattern classification. In this article, we propose an efficient method for unsupervised feature selection via orthogonal basis clustering and reliable local structure preserving, which is referred to as OCLSP briefly. Our OCLSP method consists of an orthogonal basis clustering together with an adaptive graph regularization, which realizes the functionality of simultaneously achieving excellent cluster separation and preserving the local information of data. Besides, we exploit an efficient alternative optimization algorithm to solve the challenging optimization problem of our proposed OCLSP method, and we perform a theoretical analysis of its computational complexity and convergence. Eventually, we conduct comprehensive experiments on nine real-world datasets to test the validity of our proposed OCLSP method, and the experimental results demonstrate that our proposed OCLSP method outperforms many state-of-the-art unsupervised feature selection methods in terms of clustering accuracy and normalized mutual information, which indicates that our proposed OCLSP method has a strong ability in identifying more important features.

Sequential Backward Feature Selection Method Based on Stacking Framework

Aiming at the problems of noise and irrelevant feature filtering in data processing, a feature selection method based on the stacking framework is proposed. Use the K-Fold cross-validation method to train and save DNN and SVM-based learners. The prediction results of the base learners are used as the input of the meta-learner, and the logistic regression learning model is trained and saved; comprehensively analyze the correlation coefficients of the fully connected neural network weight matrix and support vector machine, According to the learning results of the meta-learner model, assign different weights to each base learner, calculate the influence factors of each feature, and call the sequence backward search algorithm (SBS) to generate the optimal feature subset. In the experimental stage, a disease diagnosis model was constructed based on the open data set of heart disease research on the Kaggle website, and Stacking-SBS was called to generate the optimal feature subset in the feature space, and the performance comparison experiment of the diagnostic model before and after feature selection was performed, and the method was improved with information. (IG), Chi-square test (Chi) and correlation-based feature selection method (CFS) are compared. The results show that the application of this method can not only reduce model training time, but also significantly improve the model's recall rate and F1 value. In addition, this method is significantly better than the other three feature selection methods in terms of performance improvement. Finally, the open data set of cardiovascular research on the Kaggle website is used to verify the generalization ability of Stacking-SBS. The experimental results show that this method can also significantly improve the performance of the disease diagnosis model.

Online and offline streaming feature selection methods with bat algorithm for redundancy analysis

Streaming feature selection (SFS), is the task of selecting the most informative features in dealing with high-dimensional or incrementally growing problems. Several SFS algorithms have been proposed in the literature. However, they do not consider all feature subsets at the redundancy analysis step due to computational concerns. Moreover, they do not reconsider previously removed features which leads to losing most of the useful information. In this paper, the redundancy analysis step is defined as a binary optimization problem. Then, a binary bat algorithm (BBA) is adopted to find the minimal informative subsets. In this way, a large number of feature subsets can be considered effectively at the redundancy analysis step. In addition, an effective priority list is used to maintain previously removed redundant features. Such a list allows the re-examination of informative features. As a result, it is possible to consider the mutual information between features that are not streamed in an small time interval. Experimental studies on fifteen different types of datasets show that our approach is superior to state-of-the-art online and offline streaming feature selection methods in terms of classification accuracy.

Ranking based variable selection for censored data using AFT models

Numerous variable selection techniques have been developed for complete high-dimensional data but very few of them for censored data. The techniques for complete data must be modified if censoring is present. In this paper, we consider the variable selection technique for accelerated failure time (AFT) models by extending the ranking-based variable selection (RBVS) algorithm and its iterative procedure as proposed in the work of Baranowski et al. through the Stute’s weighted least square technique. Simulation studies are conducted to demonstrate the performance of the proposed methods. We further illustrate the performance of this method with a mantle cell lymphoma microarray example. When there is no correlation among the covariates, the proposed method outperforms the iterative sure independence screening and stability selection methods in terms of overall performance for high-dimensional data. Real data analysis also suggests that the proposed method can be chosen for high-dimensional censored data analysis in parallel to other methods in the literature.

Orthogonally constrained matrix factorization for robust unsupervised feature selection with local preserving

Feature selection is a significant preprocessing technique that involves discarding redundant and irrelevant features, so as to reduce the data dimensionality and build a better understanding of data. Pruning superfluous features tend to build better generalization models while improving the computation efficiency extremely. In practices, obtaining the labels of data is time-consuming and labor-intensive, which brings great challenges for feature selection. In this paper, we present a novel robust unsupervised feature selection method for unlabeled dimensionality reduction, which obtains feature importance information by predicting cluster labels of data with matrix factorization. The orthogonal constraints on two decomposing matrices facilitate achieving more accurate class labels of clusters so as to select features with highly discrimination power. Then, the local preserving term is integrated into the projected data for selecting features with local retention ability. Independent of that, an alternative iterative algorithm is incorporated into the optimization of ℓ2,1-norm objective function for efficient and robust feature selection. Extensive comparative experiments are carried out on six benchmark datasets to evaluate the performance of the proposed method. The results show that our method outperforms several well-known unsupervised feature selection methods in terms of both clustering accuracy and normalized mutual information.

An Empirical Study of Training Data Selection Methods for Ranking-Oriented Cross-Project Defect Prediction.

Ranking-oriented cross-project defect prediction (ROCPDP), which ranks software modules of a new target industrial project based on the predicted defect number or density, has been suggested in the literature. A major concern of ROCPDP is the distribution difference between the source project (aka. within-project) data and target project (aka. cross-project) data, which evidently degrades prediction performance. To investigate the impacts of training data selection methods on the performances of ROCPDP models, we examined the practical effects of nine training data selection methods, including a global filter, which does not filter out any cross-project data. Additionally, the prediction performances of ROCPDP models trained on the filtered cross-project data using the training data selection methods were compared with those of ranking-oriented within-project defect prediction (ROWPDP) models trained on sufficient and limited within-project data. Eleven available defect datasets from the industrial projects were considered and evaluated using two ranking performance measures, i.e., FPA and Norm(Popt). The results showed no statistically significant differences among these nine training data selection methods in terms of FPA and Norm(Popt). The performances of ROCPDP models trained on filtered cross-project data were not comparable with those of ROWPDP models trained on sufficient historical within-project data. However, ROCPDP models trained on filtered cross-project data achieved better performance values than ROWPDP models trained on limited historical within-project data. Therefore, we recommended that software quality teams exploit other project datasets to perform ROCPDP when there is no or limited within-project data.

Assessing the performance of a novel method for genomic selection: rrBLUP-method6

The aim of this study was to compare the predictive performance of ridge regression best linear unbiased prediction-method 6 (rrBLUPm6) with well-known genomic selection methods (rrBLUP, GBLUP and BayesA) in terms of accuracy of prediction, computing time and memory requirement. The impact of the genetic architecture and heritability on the accuracy of genomic evaluation was also studied. To this end, a genome was simulated which consisted of five chromosomes, one Morgan each, on which 5000 biallelic single-nucleotide polymorphisms (SNP) were distributed. Prediction of genomic breeding values was done in different scenarios of number of QTL (50 and 500 QTL), distribution of QTL effects (uniform, normal and gamma) and different heritability levels (0.1, 0.3 and 0.5). Pearson’s correlation between true and predicted genomic breeding values (rp,t) was used as the measure of prediction accuracy. Computing time and memory requirement were also measured for studied methods. The accuracy of rrBLUPm6 was higher than GBLUP and rrBLUP, and was comparable with BayesA. In addition, regarding computing time and memory requirement, rrBLUPm6 outperformed other methods and ranked first. A significant increase in accuracy of prediction was observed following increase in heritability. However, the number and distribution of QTL effects did not affect the accuracy of prediction significantly. As rrBLUPm6 showed a great performance regarding accuracy of prediction, computing time and memory requirement, we recommend it for genomic selection.

Dimensionality reduction using singular vectors

A common problem in machine learning and pattern recognition is the process of identifying the most relevant features, specifically in dealing with high-dimensional datasets in bioinformatics. In this paper, we propose a new feature selection method, called Singular-Vectors Feature Selection (SVFS). Let D= [A mid mathbf {b}] be a labeled dataset, where mathbf {b} is the class label and features (attributes) are columns of matrix A. We show that the signature matrix S_A=I-A^{dagger }A can be used to partition the columns of A into clusters so that columns in a cluster correlate only with the columns in the same cluster. In the first step, SVFS uses the signature matrix S_D of D to find the cluster that contains mathbf {b}. We reduce the size of A by discarding features in the other clusters as irrelevant features. In the next step, SVFS uses the signature matrix S_A of reduced A to partition the remaining features into clusters and choose the most important features from each cluster. Even though SVFS works perfectly on synthetic datasets, comprehensive experiments on real world benchmark and genomic datasets shows that SVFS exhibits overall superior performance compared to the state-of-the-art feature selection methods in terms of accuracy, running time, and memory usage. A Python implementation of SVFS along with the datasets used in this paper are available at https://github.com/Majid1292/SVFS.

A neurodynamic optimization approach to supervised feature selection via fractional programming

Feature selection is an important issue in machine learning and data mining. Most existing feature selection methods are greedy in nature thus are prone to sub-optimality. Though some global feature selection methods based on unsupervised redundancy minimization can potentiate clustering performance improvements, their efficacy for classification may be limited. In this paper, a neurodynamics-based holistic feature selection approach is proposed via feature redundancy minimization and relevance maximization. An information-theoretic similarity coefficient matrix is defined based on multi-information and entropy to measure feature redundancy with respect to class labels. Supervised feature selection is formulated as a fractional programming problem based on the similarity coefficients. A neurodynamic approach based on two one-layer recurrent neural networks is developed for solving the formulated feature selection problem. Experimental results with eight benchmark datasets are discussed to demonstrate the global convergence of the neural networks and superiority of the proposed neurodynamic approach to several existing feature selection methods in terms of classification accuracy, precision, recall, and F-measure.

Index Term Selection Heuristics for Arabic Text Retrieval

The Arabic index term selection is a challenging process due to the complex morphological nature of the Arabic language. Index term selection is a significant factor that affects the efficiency of any information retrieval system. Many methods of index term selection were proposed in the literature. The majority of them were based on root extraction and stemming. Other proposals apply complex linguistic rules and machine learning tools. This paper proposes a simple index term selection method using some heuristics such that a representative subset of terms is selected to form the index. The proposed heuristics essentially select index terms from Arabic words having the prefix ‘AL’ (definite words) as a basis. Besides, the proposed method selects new words according to any of the following heuristics: the words preceding or words succeeding definite terms, choosing words that follow some linking words and words following propositions in semi-sentences, and selecting words that represent named entities. The proposed heuristics were tested using the TREC-2001/2002 Arabic test collection. The results show the effectiveness of the proposed method since it outperforms selecting all terms stemmed by two well-known stemmers. For example, choosing definite words and words that represent named entities outperforms selecting all terms stemmed by the LIGHT10 stemmer according to the mean average precision by 8.4% and at the same time decreases the index size by 27.8%.

SuRF: A new method for sparse variable selection, with application in microbiome data analysis.

In this article, we present a new variable selection method for regression and classification purposes, particularly for microbiome analysis. Our method, called subsampling ranking forward selection (SuRF), is based on LASSO penalized regression, subsampling and forward-selection methods. SuRF offers major advantages over existing variable selection methods in terms of both sparsity of selected models and model inference. We provide an R package that can implement our method for generalized linear models. We apply our method to classification problems from microbiome data, using a novel agglomeration approach to deal with the special tree-like correlation structure of the variables. Existing methods arbitrarily choose a taxonomic level a priori before performing the analysis, whereas by combining SuRF with these aggregated variables, we are able to identify the key biomarkers at the appropriate taxonomic level, as suggested by the data. We present simulations in multiple sparse settings to demonstrate that our approach performs better than several other popularly used existing approaches in recovering the true variables. We apply SuRF to two microbiome datasets: one about prediction of pouchitis and another for identifying samples from two healthy individuals. We find that SuRF can provide a better or comparable prediction with other methods while controlling the false positive rate of variable selection.