GLEm-Net: Unified framework for data reduction with categorical and numerical features

Mixed feature selection in incomplete decision table

We focus on modelling categorical features and improving predictive power of neural networks with mixed categorical and numerical features in supervised learning tasks. The goal of this paper is to challenge the current dominant approach in actuarial data science with a new architecture of a neural network and a new training algorithm. The key proposal is to use a joint embedding for all categorical features, instead of separate entity embeddings, to determine the numerical representation of the categorical features which is fed, together with all other numerical features, into hidden layers of a neural network with a target response. In addition, we postulate that we should initialize the numerical representation of the categorical features and other parameters of the hidden layers of the neural network with parameters trained with (denoising) autoencoders in unsupervised learning tasks, instead of using random initialization of parameters. Since autoencoders for categorical data play an important role in this research, they are investigated in more depth in the paper. We illustrate our ideas with experiments on a real data set with claim numbers, and we demonstrate that we can achieve a higher predictive power of the network.

In the era of digital information, online platforms play a crucial role in shaping public opinion. However, the extensive spread of misinformation and fake news poses a significant challenge, largely fueled by non-credible users. Detecting user credibility is vital for ensuring the reliability of information on these platforms. This study employs supervised machine learning algorithms, leveraging key user features to enhance credibility detection. Feature selection methods, specifically SelectKBest and correlation-based algorithms, are explored for their impact on X-Platform user credibility detection. Utilizing various classifiers, including support vector machine, logistic regression, and XGBoost, experiments are conducted on the ArPFN dataset, which is a labeled, balanced, publicly available dataset. The evaluation includes measures like accuracy, precision, recall, and F1-score to assess efficiency. This research considers feature categories and selection methods with SML to detect their impact on the accuracy of X-Platform user credibility detection, making this research a reference for researchers and practitioners working in the field of SML, feature engineering, and social media analysis. We aim to advance the field’s understanding of effective strategies for mitigating the spread of fake news. The novelty of this study lies in the comprehensive exploration of feature selection methods and their influence on credibility detection, contributing valuable insights for future research in this domain.

Computer-aided diagnosis has the potential of increasing diagnostic accuracy by providing a second reading to radiologists. In many computerized schemes, numerous features can be extracted to describe suspect image regions. A subset of these features is then employed in a data classifier to determine whether the suspect region is abnormal or normal. Different subsets of features will, in general, result in different classification performances. A feature selection method is often used to determine an "optimal" subset of features to use with a particular classifier. A classifier performance measure (such as the area under the receiver operating characteristic curve) must be incorporated into this feature selection process. With limited datasets, however, there is a distribution in the classifier performance measure for a given classifier and subset of features. In this paper, we investigate the variation in the selected subset of "optimal" features as compared with the true optimal subset of features caused by this distribution of classifier performance. We consider examples in which the probability that the optimal subset of features is selected can be analytically computed. We show the dependence of this probability on the dataset sample size, the total number of features from which to select, the number of features selected, and the performance of the true optimal subset. Once a subset of features has been selected, the parameters of the data classifier must be determined. We show that, with limited datasets and/or a large number of features from which to choose, bias is introduced if the classifier parameters are determined using the same data that were employed to select the "optimal" subset of features.

In machine learning, feature selection is a commonly used technique for improving the predictive performance and interpretability of a trained model. Feature selection techniques are classified into three approaches: the filter, wrapper, and embedded approaches. The embedded approach performs the feature selection process during the model training and achieves a good balance between performance and computational cost in general. In the paper, we propose a novel embedded feature selection method using probabilistic model-based evolutionary optimization. We introduce the multivariate Bernoulli distribution, which determines the selection of features, and we optimize its parameters during the training. The distribution parameter update rule is the same as that of the population-based incremental learning (PBIL), but we simultaneously update the parameters of the machine learning model using an ordinary gradient descent method. This method can be easily implemented into non-linear models, such as neural networks. Moreover, we incorporate the penalty term into the objective function to control the number of selected feature. We apply the proposed method with the neural network model to the feature selection of three classification problems. The proposed method achieves competitive performance and reasonable computational cost compared with conventional feature selection methods.

Neighborhood effective information ratio for hybrid feature subset evaluation and selection

Neural networks have proven to be a good alternative in application fields such as healthcare, time-series forecasting and artificial vision, among others, for tasks like regression or classification. Their potential has been particularly remarkable in unstructured data, but recently developed architectures or their ensemble with other classical methods have produced competitive results in structured data. Feature selection has several beneficial properties: improve efficacy, performance, problem understanding and data recollection time. However, as new data sources become available and new features are generated using feature engineering techniques, more computational resources are required for feature selection methods. Feature selection takes an exorbitant amount of time in datasets with numerous features, making it impossible to use or achieving suboptimal selections that do not reflect the underlying behavior of the problem. We propose a nonparametric neural network layer which provides all the benefits of feature selection while requiring few changes to the architecture. Our method adds a novel layer at the beginning of the neural network, which removes the influence of features during training, adding inherent interpretability to the model without extra parameterization. In contrast to other feature selection methods, we propose an efficient and model-aware method to select the features with no need to train the model several times. We compared our method with a variety of popular feature selection strategies and datasets, showing remarkable results

Neighborhood rough set based heterogeneous feature subset selection

With the rapid development of artificial intelligence in recent years, the research on image processing, text mining, and genome informatics has gradually deepened, and the mining of large-scale databases has begun to receive more and more attention. The objects of data mining have also become more complex, and the data dimensions of mining objects have become higher and higher. Compared with the ultra-high data dimensions, the number of samples available for analysis is too small, resulting in the production of high-dimensional small sample data. High-dimensional small sample data will bring serious dimensional disasters to the mining process. Through feature selection, redundancy and noise features in high-dimensional small sample data can be effectively eliminated, avoiding dimensional disasters and improving the actual efficiency of mining algorithms. However, the existing feature selection methods emphasize the classification or clustering performance of the feature selection results and ignore the stability of the feature selection results, which will lead to unstable feature selection results, and it is difficult to obtain real and understandable features. Based on the traditional feature selection method, this paper proposes an ensemble feature selection method, Random Bits Forest Recursive Clustering Eliminate (RBF-RCE) feature selection method, combined with multiple sets of basic classifiers to carry out parallel learning and screen out the best feature classification results, optimizes the classification performance of traditional feature selection methods, and can also improve the stability of feature selection. Then, this paper analyzes the reasons for the instability of feature selection and introduces a feature selection stability measurement method, the Intersection Measurement (IM), to evaluate whether the feature selection process is stable. The effectiveness of the proposed method is verified by experiments on several groups of high-dimensional small sample data sets.

Fault diagnosis methods based on machine learning have made great progress for rotating machinery. The main steps of the machine learning process involve feature extraction, selection, and classification. Feature selection improves classification accuracy and reduces diagnosis time by selecting the better features. Due to the difficulty of traditional feature selection methods to rank the feature importance of each class, the best subset of features could hardly be obtained. Therefore, this article proposes a new feature selection method to address the shortcomings of the above traditional methods, called Feature Ranking based on Optimal Class Distance Ratio (FROCDR), which can choose the optimal features between every two classes of samples to obtain feature ranking that is conducive to classification. In order to comprehensively extract the fault information in the signal, the multiscale analysis and the variational mode decomposition (VMD) method are applied to process the vibration signals under different scales and frequency bands, and the processed signals are visualized by symmetrized dot pattern (SDP). In addition, features are extracted from the obtained SDP images, and the proposed FROCDR method is used to select the best subset of features. The final diagnosis task is accomplished by a random forest (RF) classifier. Experimental cases of bearing and gear data show that the proposed method has higher diagnostic accuracy and stability.

Prediction of the stage of cancer plays an important role in planning the course of treatment and has been largely reliant on imaging tools which do not capture molecular events that cause cancer progression. Gene-expression data-based analyses are able to identify these events, allowing RNA-sequence and microarray cancer data to be used for cancer analyses. Breast cancer is the most common cancer worldwide, and is classified into four stages - stages 1, 2, 3, and 4 [2]. While machine learning models have previously been explored to perform stage classification with limited success, multi-class stage classification has not had significant progress. There is a need for improved multi-class classification models, such as by investigating deep learning models. Gene-expression-based cancer data is characterised by the small size of available datasets, class imbalance, and high dimensionality. Class balancing methods must be applied to the dataset. Since all the genes are not necessary for stage prediction, retaining only the necessary genes can improve classification accuracy. The breast cancer samples are to be classified into 4 classes of stages 1 to 4. Invasive ductal carcinoma breast cancer samples are obtained from The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) datasets and combined. Two class balancing techniques are explored, synthetic minority oversampling technique (SMOTE) and SMOTE followed by random undersampling. A hybrid feature selection pipeline is proposed, with three pipelines explored involving combinations of filter and embedded feature selection methods: Pipeline 1 - minimum-redundancy maximum-relevancy (mRMR) and correlation feature selection (CFS), Pipeline 2 - mRMR, mutual information (MI) and CFS, and Pipeline 3 - mRMR and support vector machine-recursive feature elimination (SVM-RFE). The classification is done using deep learning models, namely deep neural network, convolutional neural network, recurrent neural network, a modified deep neural network, and an AutoKeras generated model. Classification performance post class-balancing and various feature selection techniques show marked improvement over classification prior to feature selection. The best multiclass classification was found to be by a deep neural network post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with a Cohen-Kappa score of 0.303 and a classification accuracy of 53.1%. For binary classification into early and late-stage cancer, the best performance is obtained by a modified deep neural network (DNN) post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with an accuracy of 81.0% and a Cohen-Kappa score (CKS) of 0.280. This pipeline also showed improved multiclass classification performance on neuroblastoma cancer data, with a best area under the receiver operating characteristic (auROC) curve score of 0.872, as compared to 0.71 obtained in previous work, an improvement of 22.81%. The results and analysis reveal that feature selection techniques play a vital role in gene-expression data-based classification, and the proposed hybrid feature selection pipeline improves classification performance. Multi-class classification is possible using deep learning models, though further improvement particularly in late-stage classification is necessary and should be explored further.

In many pattern recognition applications feature selection and instance selection can be used as two data preprocessing methods that aim at reducing the computational cost of the learning process. Moreover, in some cases, feature subset selection can improve the classification performance. Feature selection and instance selection can be interesting since the choice of features and instances greatly influence the performance of the learnt models as well as their training costs. In the past, unifying both problems was carried out by solving a global optimization problem using meta-heuristics. This paradigm not only does not exploit the manifold structure of data but can be computationally expensive. To the best of our knowledge, the joint use of sparse modeling representative and feature subset relevance have not been exploited by the joint feature and selection methods. In this paper, we target the joint feature and instance selection by adopting feature subset relevance and sparse modeling representative selection. More precisely, we propose three schemes for the joint feature and instance selection. The first is a wrapper technique while the two remaining ones are filter approaches. In the filter approaches, the search process adopts a genetic algorithm in which the evaluation is mainly given by a score that quantify the goodness of the features and instances. An efficient instance selection technique is used and integrated in the search process in order to adapt the instances to the candidate feature subset. We evaluate the performance of the proposed schemes using image classification where classifiers are the nearest neighbor classifier and support vector machine classifier. The study is conducted on five public image datasets. These experiments show the superiority of the proposed schemes over various baselines. The results confirm that the filter approaches leads to promising improvement on classification accuracy when both feature selection and instance selection are adopted.

High-dimensional datasets can lead to overfitting and computationally expensive model building on machine learning. This study uses a dimensionality reduction technique, namely feature selection techniques, to overcome these problems. Five feature selection methods were used, i.e., Chi-Square (CS), Information Gain (IG), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Least Absolute Shrinkage and Selection Operator (LASSO), and three classifier methods viz. Naïve Bayes, Extreme Gradient Boosting (XGB), and RF Classifier. The dataset used is the Heart Attack Analysis & Prediction Dataset. In this study, three scenarios of the best feature selection were carried out, namely: 1. selection of the best feature using a specific feature selection, 2. the intersection of selection of the best feature from the same category, 3. the intersection of selection of the best feature from the five proposed feature selection methods. The performance model is measured using accuracy, precision, recall, f1-score, AUC, and training time. This study reveals that feature selection is very effective in improving the performance of prediction models. Based on the experiment results, the best feature selection is CS and IG in the Filter Category with the XGB model. The best feature selected improved the performance of accuracy, precision, recall, f1-score, and AUC, i.e., 1.7%, 1%, 2.3%, 1.6%, and 0.2%, respectively. Meanwhile, training time requirements decreased by 23.5%. Feature selection with specific techniques performs better than feature selection by selecting the best features from the same category feature selection technique or various other feature selection methods.

This study is aimed at assessing the usefulness of different feature selection and classification methodologies in the context of sleep apnea hypopnea syndrome (SAHS) detection. Feature extraction, selection and classification stages were applied to analyze blood oxygen saturation (SaO2) recordings in order to simplify polysomnography (PSG), the gold standard diagnostic methodology for SAHS. Statistical, spectral and nonlinear measures were computed to compose the initial feature set. Principal component analysis (PCA), forward stepwise feature selection (FSFS) and genetic algorithms (GAs) were applied to select feature subsets. Fisher's linear discriminant (FLD), logistic regression (LR) and support vector machines (SVMs) were applied in the classification stage. Optimum classification algorithms from each combination of these feature selection and classification approaches were prospectively validated on datasets from two independent sleep units. FSFS + LR achieved the highest diagnostic performance using a small feature subset (4 features), reaching 83.2% accuracy in the validation set and 88.7% accuracy in the test set. Similarly, GAs + SVM also achieved high generalization capability using a small number of input features (7 features), with 84.2% accuracy on the validation set and 84.5% accuracy in the test set. Our results suggest that reduced subsets of complementary features (25% to 50% of total features) and classifiers with high generalization ability could provide high-performance screening tools in the context of SAHS.

Sentiment Analysis (SA) research has increased tremendously in recent times. Sentiment analysis deals with the methods that automatically process the text contents and extract the opinion of the users. In this paper, unigram and bi-grams are extracted from the text, and composite features are created using them. Part of Speech (POS) based features adjectives and adverbs are also extracted. Information Gain (IG) and Minimum Redundancy Maximum Relevancy (mRMR) feature selection methods are used to extract prominent features. Further, effect of various feature sets for sentiment classification is investigated using machine learning methods. Effects of different categories of features are investigated on four standard datasets i.e. Movie review, product (book, DVD and electronics) review dataset. Experimental results show that composite features created from prominent features of unigram and bi-gram perform better than other features for sentiment classification. mRMR is better feature selection method as compared to IG for sentiment classification. Boolean Multinomial Naïve Bayes (BMNB) algorithm performs better than Support Vector Machine (SVM) classifier for sentiment analysis in terms of accuracy and execution time.KeywordsSentiment Analysisfeature selection methodsmachine learningInformation GainMinimum Redundancy Maximum Relevancy (mRMR)composite features