Incremental granularity feature selection based on representative samples

In this study, the authors propose a new feature selection scheme, the incremental forward feature selection, which is inspired by incremental reduced support vector machines. In their method, a new feature is added into the current selected feature subset if it will bring in the most extra information. This information is measured by using the distance between the new feature vector and the column space spanned by current feature subset. The incremental forward feature selection scheme can exclude highly linear correlated features that provide redundant information and might degrade the efficiency of learning algorithms. The method is compared with the weight score approach and the 1-norm support vector machine on two well-known microarray gene expression data sets, the acute leukemia and colon cancer data sets. These two data sets have a very few observations but huge number of genes. The linear smooth support vector machine was applied to the feature subsets selected by these three schemes respectively and obtained a slightly better classification results in the 1-norm support vector machine and incremental forward feature selection. Finally, the authors claim that the rest of genes still contain some useful information. The previous selected features are iteratively removed from the data sets and the feature selection and classification steps are repeated for four rounds. The results show that there are many distinct feature subsets that can provide enough information for classification tasks in these two microarray gene expression data sets.

Incremental feature selection refreshes a subset of information-rich features from added-in samples without forgetting the previously learned knowledge. However, most existing algorithms for incremental feature selection have no explicit mechanisms to handle heterogeneous data with symbolic and real-valued features. Therefore, this paper presents an incremental feature selection method for heterogeneous data with the sequential arrival of samples in group. Discernible neighborhood counting that measures different types of features, is first introduced to establish a framework for feature selection from heterogeneous data. With the arrival of new samples, the discernible neighborhood counting of a feature subset is then updated to reveal the incremental feature selection scheme. This scheme determines the criterion for efficiently adding informative features and deleting redundant features. Based on the incremental scheme, our incremental feature selection algorithm is further formulated to select valuable features from heterogeneous data. Extensive experiments are finally conducted to demonstrate the effectiveness and the efficiency of the proposed incremental feature selection algorithm.

Feature selection is a popular technique of preprocessing data. In order to deal with dynamic or large data, incremental feature selection has been developed, in which the features selected from existing data are integrated with those mined from both existing and dynamic data in the manner of incremental computation. Fuzzy rough set theory is powerful in handling uncertainty in real-valued data or even mixed data, and one of its most important applications is feature selection. Nevertheless, not much work has been found on fuzzy-rough-set-based incremental feature selection. Therefore, in this article, we investigate the incremental feature selection using a fuzzy-rough-set-based information entropy with incoming instances. Specifically, the representative instances from the incoming ones are first selected according to the information coverage of fuzzy granules generated by fuzzy rough sets. Then, the incremental mechanism of the fuzzy-rough-set-based information entropy is formulated by adding newcome instances. Finally, an incremental feature selection procedure, which we call active incremental feature selection, is proposed. Furthermore, some numerical experiments are conducted to assess the performance of the proposed feature selection algorithm, and the results show that our algorithm is of a prominent advantage in terms of computational time, especially for a dataset with large number of instances.

Gene expression data is illustration levels of genes that DNA encode into the protein such as muscle or brain cells. However, some abnormal cells may evolve from unnatural expression levels. So, finding a subset of informative gene would be beneficial to biologists because it can help to identify discriminate genes. Unfortunately, genes grow rapidly up into the tens of thousands gene which make it difficult for classifying processes such as curse of dimensionality and misclassification problems. This paper proposes classification models based-on incremental learning algorithm and feature selection on gene expression data. Three feature selection methods: Correlation based Feature Selection (Cfs), Gain Ratio (GR), and Information Gain (Info) combined with Incremental Learning Algorithm based-on Mahalanobis Distance (ILM). Result of the experiment represented proposes models CfsILM, GRILM and InfoILM not only reduce many dimensions, save time-resource but also improve accuracy rate. Particularly, CfsILM was outstanding than other models on three public gene expression datasets.

This Gene expression data illustrates levels of genes that DNA encodes into the protein such as muscle or brain cells. However, some abnormal cells may evolve from unnatural expression levels. Therefore, finding a subset of informative gene would be beneficial to biologists because it can identify discriminative genes. Unfortunately, genes grow up rapidlyinto the tens of thousands gene which make it difficult for classifying processes such as curse of dimensionality and misclassification problems. This paper proposed classification model based-on incremental learning algorithm and feature selection on gene expression data. Three feature selection methods: Correlation based Feature Selection (Cfs), Gain Ratio (GR), and Information Gain (Info) combined with Incremental Learning Algorithm based-on Mahalanobis Distance (ILM). Result of the experiment represented proposed models CfsILM, GRILM and InfoILM not only to reduce many dimensions from 2001, 7130 and 4026 into 26, 135, and 135 that save time-resource but also to improve accuracy rate 64.52%, 34.29%, and 8.33% into 90%, 97.14%, and 83.33% respectively. Particularly, CfsILM is more outstanding than other models on three public gene expression datasets.

The classification of humanoid locomotion is a troublesome exercise because of non-linearity associate with gait signals. The classification using the different machine learning technique leads for over fitting and under fitting. To select the optimized feature is a difficult task. The high dimension feature vector requires a high computational cost. The hand craft feature selection machine learning techniques performed poor. We have used the incremental feature selection strategy for feature selection. In this paper we first selected the feature and identify the principle feature then we classify gait data using different machine learning technique (KNN, ANN, SVM, DNN and classifier fusion) and shown the performance comparison. Experimental result on real time datasets propose method is better than previous method as far as humanoid locomotion classification is concerned and the generalization accuracy provided by new feature selection method i.e. incremental feature selection (IFS) with analysis of variance (ANOVA) (Zhang et al., 20). During the feature extraction, 17 features were selected from the existing literatures (Wang et al. in IEEE Trans Circ Syst Video Technol 14(2):149–158, 15). Using all the features could lead to over fitting, information redundancy and dimension disaster. Thus, a system with optimal features was selected using ANOVA combined with IFA. These selected features were then fed as an input to the ANN, SVM, KNN and DNN model. These individual classifiers were combined to produce classifier fusion model. The 5-fold cross-validation was used to evaluate the performance of the proposed model. Based on the empirical results it may be concluded that classifier fusion provides satisfactory results (92.23 %) compared to other individual classifiers. One-way analysis of variance test, Friedman’s test and Kruskal-Wallis test has also been conducted to validate the statistical significance of the results. The proposed system can be used as recommender system based on behavioral gait pattern about the performance of player of Indian cricket team, Biometric and help to diagnosis Parkinson disease.

Incremental feature selection by sample selection and feature-based accelerator

Incremental approaches for heterogeneous feature selection in dynamic ordered data

GalNAc-transferase specificity prediction based on feature selection method

Graph classification is an important tool for analyzing data with structure dependency, where subgraphs are often used as features for learning. In reality, the dimension of the subgraphs crucially depends on the threshold setting of the frequency support parameter, and the number may become extremely large. As a result, subgraphs may be incrementally discovered to form a feature stream and require the underlying graph classifier to effectively discover representative subgraph features from the subgraph feature stream. In this paper, we propose a primal-dual incremental subgraph feature selection algorithm (ISF) based on a max-margin graph classifier. The ISF algorithm constructs a sequence of solutions that are both primal and dual feasible. Each primal-dual pair shrinks the dual gap and renders a better solution for the optimal subgraph feature set. To avoid bias of ISF algorithm on short-pattern subgraph features, we present a new incremental subgraph join feature selection algorithm (ISJF) by forcing graph classifiers to join short-pattern subgraphs and generate long-pattern subgraph features. We evaluate the performance of the proposed models on both synthetic networks and real-world social network data sets. Experimental results demonstrate the effectiveness of the proposed methods.

Incremental unsupervised feature selection for dynamic incomplete multi-view data

Feature selection method with rough sets based on incremental learning has the major advantage of the higher efficiency in a dynamic information system, which has attracted extensive research. However, the incremental approximation feature selection with an accelerator (IAFSA) remains ambiguous for a dynamic information system with fuzzy decisions (ISFD). Driven by this concern, the non-incremental approximation feature selection is first presented by fuzzy knowledge distance (FKD). Second, the incremental theory of FKD is constructed with a batch of objects appended to or removed from the dynamic ISFD. Subsequently, an acceleration mechanism to eliminate redundant information granules is developed to reduce the sample space. Eventually, two categories of IAFSA based on FKD are presented. The experiments reflect the efficiency and effectiveness of the developed IAFSA algorithms.

Predicting N-terminal acetylation based on feature selection method

It is important to identify which proteins can interact with nucleic acids for the purpose of protein annotation, since interactions between nucleic acids and proteins involve in numerous cellular processes such as replication, transcription, splicing, and DNA repair. This research tries to identify proteins that can interact with DNA, RNA, and rRNA, respectively. mRMR (Minimum redundancy and maximum relevance), with its elegant mathematical formulation, has been applied widely in processing biological data and feature analysis since its introduction in 2005. mRMR plus incremental feature selection (IFS) is known to be very efficient in feature selection and analysis, and able to improve both effectiveness and efficiency of a prediction model. IFS is applied to decide how many features should be selected from feature list provided by mRMR. In the end, the selected features of mRMR and IFS are further refined by a conventional feature selection method--forward feature wrapper (FFW), by reordering the features. Each protein is coded by 132 features including amino acid compositions and physicochemical properties. After the feature selection, k-Nearest Neighbor algorithm, the adopted prediction model, is trained and tested. As a result, the optimized prediction accuracies for the DNA, RNA, and rRNA are 82.0, 83.4, and 92.3%, respectively. Furthermore, the most important features that contribute to the prediction are identified and analyzed biologically. The predictor, developed for this research, is available for public access at http://chemdata.shu.edu.cn/protein_na_mrmr/.

Incremental capacity feature selection for lithiumion battery state of health estimation considering estimation capability and efficiency