Subset Of Informative Features Research Articles

A Review On Early Detection Of Chronic Kidney Disease

Early detection of Chronic Kidney Disease (CKD) is critical for timely intervention and effective treatment. Deep learning algorithms have demonstrated promise in medical applications, including disease detection. In this study, we propose a deep learning-based system for early CKD detection using the Chronic Kidney Disease dataset from Kaggle. Additionally, we incorporate the Grasshopper Optimization Algorithm (GOA) for feature selection to enhance the system's performance and interpretability. Our system employs a convolutional neural network (CNN) architecture to analyze clinical and laboratory attributes from the CKD dataset, obtained from Kaggle, consisting of 4,000 instances with 25 attributes. These attributes encompass patient demographics, blood tests, and medical history, providing a comprehensive representation of CKD-related factors. To improve the system's performance, we integrate the GOA for feature selection. The GOA is a nature-inspired metaheuristic optimization algorithm that mimics the foraging behavior of grasshoppers. It aims to identify the most relevant attributes associated with CKD from the dataset. By selecting a subset of informative features, we enhance the model's predictive accuracy and reduce overfitting. During the training phase, the CNN learns to automatically extract relevant features and patterns associated with CKD from the selected attributes. Additionally, data preprocessing techniques such as normalization and feature scaling are applied to further improve the model's performance and generalizability. To evaluate the system's performance, we conduct experiments using a separate test dataset comprising 1,000 instances from the CKD dataset. The incorporation of the GOA for feature selection in our deep learning system not only improves its performance but also enhances interpretability. By identifying the most relevant attributes associated with CKD, we focus on key biomarkers and risk factors, enhancing the system's accuracy and providing valuable insights into the disease. Our research showcases the potential of deep learning algorithms, coupled with GOA-based feature selection, for early CKD detection. By leveraging the Kaggle CKD dataset and incorporating the GOA, we contribute to improving the accuracy and applicability of the system in real-world clinical settings. To handle Big data we are proposing to implement this problem on Pyspark one of the Big data computational environments for effective learning. In this platform, we can dynamically scale the infrastructure as per the demand of the data. Ultimately, our work aims to advance the early detection and management of CKD, leading to improved patient outcomes and more effective healthcare interventions.

An overlapping conscious relief-based feature subset selection method

Feature selection is considered as a fundamental prepossessing step in various data mining and machine learning based works. The quality of features is essential to achieve good classification performance and to have better data analysis experience. Among several feature selection methods, distance-based methods are gaining popularity because of their eligibility in capturing feature interdependency and relevancy with the endpoints. However, most of the distance-based methods only rank the features and ignore the class overlapping issues. Features with class overlapping data work as an obstacle during classification. Therefore, the objective of this research work is to propose a method named overlapping conscious MultiSURF (OMsurf) to handle data overlapping and select a subset of informative features discarding the noisy ones. Experimental results over 20 benchmark dataset demonstrates the superiority of OMsurf over six existing state-of-the-art methods

Evolving pathway activation from cancer gene expression data using nature-inspired ensemble optimization

Class-imbalanced biological datasets pose significant challenges in machine learning and data analysis tasks. Prior methods to handle imbalance rely on data oversampling, which increases computational costs and overfitting. While feature selection and ensemble learning are promising techniques, current applications in imbalanced contexts are limited. To address these challenges, we present a novel framework called Hybrid Sampling Nature-Inspired Optimization Ensemble (HSNOE) to enhance the identification of hidden responders in imbalanced biological datasets. Our contributions are three-fold: 1) A hybrid undersampling and oversampling technique to mitigate class-imbalance; 2) Integrate an ant colony optimization-based feature selection that identifies informative feature subsets; 3) An ensemble classifier integrating diverse models trained on optimized features to improve performance. The experiments conducted on the five biological datasets demonstrate that HSNOE exhibits more stable comprehensive performance across six evaluation metrics compared to ten benchmark methods. We also conducted a biological analysis specifically on the Pan-cancer dataset. Moreover, the HSNOE method has been made publicly available on GitHub.11https://github.com/wangxb96/HSNOE.

Shared Manifold Regularized Joint Feature Selection for Joint Classification and Regression in Alzheimer's Disease Diagnosis.

In Alzheimer's disease (AD) diagnosis, joint feature selection for predicting disease labels (classification) and estimating cognitive scores (regression) with neuroimaging data has received increasing attention. In this paper, we propose a model named Shared Manifold regularized Joint Feature Selection (SMJFS) that performs classification and regression in a unified framework for AD diagnosis. For classification, unlike the existing works that build least squares regression models which are insufficient in the ability of extracting discriminative information for classification, we design an objective function that integrates linear discriminant analysis and subspace sparsity regularization for acquiring an informative feature subset. Furthermore, the local data relationships are learned according to the samples' transformed distances to exploit the local data structure adaptively. For regression, in contrast to previous works that overlook the correlations among cognitive scores, we learn a latent score space to capture the correlations and employ the latent space to design a regression model with l2,1 -norm regularization, facilitating the feature selection in regression task. Moreover, the missing cognitive scores can be recovered in the latent space for increasing the number of available training samples. Meanwhile, to capture the correlations between the two tasks and describe the local relationships between samples, we construct an adaptive shared graph to guide the subspace learning in classification and the latent cognitive score learning in regression simultaneously. An efficient iterative optimization algorithm is proposed to solve the optimization problem. Extensive experiments on three datasets validate the discriminability of the features selected by SMJFS.

Improved binary differential evolution with dimensionality reduction mechanism and binary stochastic search for feature selection

Computer systems store massive amounts of data with numerous features, leading to the need to extract the most important features for better classification in a wide variety of applications. Poor performance of various machine learning algorithms may be caused by unimportant features that increase the time and memory required to build a classifier. Feature selection (FS) is one of the efficient approaches to reducing the unimportant features. This paper, therefore, presents a new FS, named BDE-BSS-DR, that utilizes Binary Differential Evolution (BDE), Binary Stochastic Search (BSS) algorithm, and Dimensionality Reduction (DR) mechanism. The BSS algorithm increases the search capability of the BDE by escaping from local optimal points and exploring the search space. The DR mechanism then reduces the dimensions of the search space gradually. As a result of using DR, the local optima of the search space and the problem of wrong removal of important features before starting the search process are reduced. The algorithm's efficiency is evaluated on 20 different medical datasets. The obtained outcomes indicate that the BDE-BSS-DR outperforms the BDE and BDE-BSS algorithms significantly. Furthermore, the effectiveness of the proposed algorithms in selecting the most important features of the heart disease data, several cancer diseases, and COVID-19 are also compared with several other state-of-the-art methods. Our results show that the BDE-BSS-DR with SVM classifier has a significant advantage over other methods with an average classification accuracy of 95.05% in heart disease and 99.40% in COVID-19 disease. In addition, the comparisons made with KNN and SVM classification prove the efficiency of the DR and BSS in generating a subset of optimal and informative features.

Breast Cancer Detection and Classification Using Hybrid Feature Selection and DenseXtNet Approach

Breast Cancer (BC) detection and classification are critical tasks in medical diagnostics. The lives of patients can be greatly enhanced by the precise and early detection of BC. This study suggests a novel approach for detecting BC that combines deep learning models and sophisticated image processing techniques to address those shortcomings. The BC dataset was pre-processed using histogram equalization and adaptive filtering. Data augmentation was performed using cycle-consistent GANs (CycleGANs). Handcrafted features like Haralick features, Gabor filters, contour-based features, and morphological features were extracted, along with features from deep learning architecture VGG16. Then, we employed a hybrid optimization model, combining the Sparrow Search Algorithm (SSA) and Red Deer Algorithm (RDA), called Hybrid Red Deer with Sparrow optimization (HRDSO), to select the most informative subset of features. For detecting BC, we proposed a new DenseXtNet architecture by combining DenseNet and optimized ResNeXt, which is optimized using the hybrid optimization model HRDSO. The proposed model was evaluated using various performance metrics and compared with existing methods, demonstrating that its accuracy is 97.58% in BC detection. MATLAB was utilized for implementation and evaluation purposes.

Hybrid distributed feature selection using particle swarm optimization-mutual information

Feature selection (FS) is a data preprocessing step in machine learning (ML) that selects a subset of relevant and informative features from a large feature pool. FS helps ML models improve their predictive accuracy at lower computational costs. Moreover, FS can handle the model overfitting problem on a high-dimensional dataset. A major problem with the filter and wrapper FS methods is that they consume a significant amount of time during FS on high-dimensional datasets. The proposed “HDFS(PSO-MI): hybrid distribute feature selection using particle swarm optimization-mutual information (PSO-MI)”, is a PSO-based hybrid method that can overcome the problem mentioned above. This method hybridizes the filter and wrapper techniques in a distributed manner. A new combiner is also introduced to merge the effective features selected from multiple data distributions. The effectiveness of the proposed HDFS(PSO-MI) method is evaluated using five ML classifiers, i.e., logistic regression (LR), k-NN, support vector machine (SVM), decision tree (DT), and random forest (RF), on various datasets in terms of accuracy and Matthew’s correlation coefficient (MCC). From the experimental analysis, we observed that HDFS(PSO-MI) method yielded more than 98%, 95%, 92%, 90%, and 85% accuracy for the unbalanced, kidney disease, emotions, wafer manufacturing, and breast cancer datasets, respectively. Our method shows promising results comapred to other methods, such as mutual information, gain ratio, Spearman correlation, analysis of variance (ANOVA), Pearson correlation, and an ensemble feature selection with ranking method (EFSRank).

A novel feature ranking algorithm for text classification: Brilliant probabilistic feature selector (BPFS)

AbstractText classification (TC) is a very crucial task in this century of high‐volume text datasets. Feature selection (FS) is one of the most important stages in TC studies. In the literature, numerous feature selection methods are recommended for TC. In the TC domain, filter‐based FS methods are commonly utilized to select a more informative feature subsets. Each method uses a scoring system that is based on its algorithm to order the features. The classification process is then carried out by choosing the top‐N features. However, each method's feature order is distinct from the others. Each method selects by giving the qualities that are critical to its algorithm a high score, but it does not select by giving the features that are unimportant a low value. In this paper, we proposed a novel filter‐based FS method namely, brilliant probabilistic feature selector (BPFS), to assign a fair score and select informative features. While the BPFS method selects unique features, it also aims to select sparse features by assigning higher scores than common features. Extensive experimental studies using three effective classifiers decision tree (DT), support vector machines (SVM), and multinomial naive bayes (MNB) on four widely used datasets named Reuters‐21,578, 20Newsgroup, Enron1, and Polarity with different characteristics demonstrate the success of the BPFS method. For feature dimensions, 20, 50, 100, 200, 500, and 1000 dimensions were used. The experimental results on different benchmark datasets show that the BPFS method is more successful than the well‐known and recent FS methods according to Micro‐F1 and Macro‐F1 scores.

Snake-Efficient Feature Selection-Based Framework for Precise Early Detection of Chronic Kidney Disease.

Chronic kidney disease (CKD) refers to impairment of the kidneys that may worsen over time. Early detection of CKD is crucial for saving millions of lives. As a result, several studies are currently focused on developing computer-aided systems to detect CKD in its early stages. Manual screening is time-consuming and subject to personal judgment. Therefore, methods based on machine learning (ML) and automatic feature selection are used to support graders. The goal of feature selection is to identify the most relevant and informative subset of features in a given dataset. This approach helps mitigate the curse of dimensionality, reduce dimensionality, and enhance model performance. The use of natural-inspired optimization algorithms has been widely adopted to develop appropriate representations of complex problems by conducting a blackbox optimization process without explicitly formulating mathematical formulations. Recently, snake optimization algorithms have been developed to identify optimal or near-optimal solutions to difficult problems by mimicking the behavior of snakes during hunting. The objective of this paper is to develop a novel snake-optimized framework named CKD-SO for CKD data analysis. To select and classify the most suitable medical data, five machine learning algorithms are deployed, along with the snake optimization (SO) algorithm, to create an extremely accurate prediction of kidney and liver disease. The end result is a model that can detect CKD with 99.7% accuracy. These results contribute to our understanding of the medical data preparation pipeline. Furthermore, implementing this method will enable health systems to achieve effective CKD prevention by providing early interventions that reduce the high burden of CKD-related diseases and mortality.

Exploring the synergistic effects of hybrid grid partitioning and rough set method for fuzzy rule generation in dataset classification

This research explores the synergistic effects of hybrid grid partitioning and the rough set method for fuzzy rule generation in dataset classification. The aim is to improve the accuracy and interpretability of the classification process. The rough set-based feature selection technique is employed to identify the most relevant features for classification, leading to a focused and informative feature subset. The hybrid grid partitioning approach combines clustering algorithms and grid-based methods to create an efficient grid structure, capturing the intrinsic data distribution. This enhances the representation and separation of data regions, improving classification accuracy. The generated fuzzy rule base provides interpretable decision rules, enabling domain experts to gain insights into the classification process. The proposed approach strikes a balance between accuracy and interpretability, making it valuable for various domains. However, limitations such as generalizability and scalability should be considered. Comparative analysis with existing methods and real-world case studies would further validate the effectiveness of the approach. Overall, this research contributes to the advancement of dataset classification and provides a novel integrated approach for accurate and interpretable classification.

Differential Evolution-Based Feature Selection: A Niching-Based Multiobjective Approach

Feature selection is to reduce both the dimensionality of data and the classification error rate (i.e., increase the classification accuracy) of a learning algorithm. The two objectives are often conflicting, thus a multiobjective feature selection method can obtain a set of nondominated feature subsets. Each solution in the set has a different size and a corresponding classification error rate. However, most existing feature selection algorithms have ignored that, for a given size, there can be different feature subsets with very similar or the same accuracy. This article introduces a niching-based multiobjective feature selection method that simultaneously minimizes the number of selected features and the classification error rate. The proposed method conceives to identify: 1) a set of feature subsets with good convergence and distribution and 2) multiple feature subsets choosing the same number of features with almost the same lowest classification error rate. The contributions of this article are threefold. First, a niching and global interaction mutation operator is proposed that can produce promising feature subsets. Second, a newly developed environmental selection mechanism allows equal informative feature subsets to be stored by relaxing the Pareto-dominance relationship. Finally, the proposed subset repairing mechanism can generate better feature subsets and further remove the redundant features. The proposed method is compared against seven multiobjective feature selection algorithms on 19 datasets, including both binary and multiclass classification tasks. The results show that the proposed method can evolve a rich and diverse set of nondominated solutions for different feature selection tasks, and their availability helps in understanding the relationships between features.

A comparative analysis of meta-heuristic optimization algorithms for feature selection on ML-based classification of heart-related diseases.

This study aims to use a machine learning (ML)-based enhanced diagnosis and survival model to predict heart disease and survival in heart failure by combining the cuckoo search (CS), flower pollination algorithm (FPA), whale optimization algorithm (WOA), and Harris hawks optimization (HHO) algorithms, which are meta-heuristic feature selection algorithms. To achieve this, experiments are conducted on the Cleveland heart disease dataset and the heart failure dataset collected from the Faisalabad Institute of Cardiology published at UCI. CS, FPA, WOA, and HHO algorithms for feature selection are applied for different population sizes and are realized based on the best fitness values. For the original dataset of heart disease, the maximum prediction F-score of 88% is obtained using K-nearest neighbour (KNN) when compared to logistic regression (LR), support vector machine (SVM), Gaussian Naive Bayes (GNB), and random forest (RF). With the proposed approach, the heart disease prediction F-score of 99.72% is obtained using KNN for population sizes 60 with FPA by selecting eight features. For the original dataset of heart failure, the maximum prediction F-score of 70% is obtained using LR and RF compared to SVM, GNB, and KNN. With the proposed approach, the heart failure prediction F-score of 97.45% is obtained using KNN for population sizes 10 with HHO by selecting five features. Experimental findings show that the applied meta-heuristic algorithms with ML algorithms significantly improve prediction performances compared to performances obtained from the original datasets. The motivation of this paper is to select the most critical and informative feature subset through meta-heuristic algorithms to improve classification accuracy.

MFSJMI: Multi-label feature selection considering join mutual information and interaction weight

Multi-label feature selection captures a reliable and informative feature subset from high-dimensional multi-label data, which plays an important role in pattern recognition. In conventional information-theoretical based multi-label feature selection methods, the high-order feature relevance between feature and label set is evaluated using low-order mutual information. However, existing methods do not establish the theoretical basis for the low-order approximation. To fill this gap, we first identify two underlying assumptions based on high-order label distribution: Label Independence Assumption (LIA) and Paired-label Independence Assumption (PIA). Second, we systematically analyze the strengths and weaknesses of two assumptions and introduce joint mutual information to satisfy more realistic label distribution. Furthermore, by decomposing joint mutual information, an interaction weight is proposed to consider multiple label correlations. Finally, a new method considering join mutual information and interaction weight is proposed. Comprehensive experiments demonstrate the effectiveness of the proposed method on various evaluation metrics.

A modified binary version of aphid–ant mutualism for feature selection: a COVID-19 case study

AbstractThe speedy development of intelligent technologies and gadgets has led to a drastic increment of dimensions within the datasets in recent years. Dimension reduction algorithms, such as feature selection methods, are crucial to resolving this obstacle. Currently, metaheuristic algorithms have been extensively used in feature selection tasks due to their acceptable computational cost and performance. In this article, a binary-modified version of aphid–ant mutualism (AAM) called binary aphid–ant mutualism (BAAM) is introduced to solve the feature selection problems. Like AAM, in BAAM, the intensification and diversification mechanisms are modeled via the intercommunication of aphids with other colonies’ members, including aphids and ants. However, unlike AAM, the number of colonies’ members can change in each iteration based on the attraction power of their leaders. Moreover, the second- and third-best individuals can take the place of the ringleader and lead the pioneer colony. Also, to maintain the population diversity, prevent premature convergence, and facilitate information sharing between individuals of colonies including aphids and ants, a random cross-over operator is utilized in BAAM. The proposed BAAM is compared with five other feature selection algorithms using several evaluation metrics. Twelve medical and nine non-medical benchmark datasets with different numbers of features, instances, and classes from the University of California, Irvine and Arizona State University repositories are considered for all the experiments. Moreover, a coronavirus disease (COVID-19) dataset is used to validate the effectiveness of the BAAM in real-world applications. Based on the acquired outcomes, the proposed BAAM outperformed other comparative methods in terms of classification accuracy using various classifiers, including K nearest neighbor, kernel-based extreme learning machine, and multi-class support vector machine, choosing the most informative features, the best and mean fitness values and convergence speed in most cases. As an instance, in the COVID-19 dataset, BAAM achieved 96.53% average accuracy and selected the most informative feature subset.

JoMIC: A joint MI-based filter feature selection method

Feature selection (FS) is a common preprocessing step of machine learning that selects informative subset of features which fuels a model to perform better during prediction or classification. It helps in the design of an intelligent and expert system used in computer vision, image processing, gene expression data analysis, intrusion detection and natural language processing. In this paper, we introduce an effective filter method called Joint Mutual Information with Class relevance (JoMIC) using multivariate Joint Mutual Information (JMI) and Mutual Information (MI). Our method considers both JMI and MI of a non selected feature with selected ones w.r.t a given class to select a feature that is highly relevant to the class but non redundant to other selected features. We compare our method with seven other filter-based methods using the machine learning classifiers viz., Logistic Regression, Support Vector Machine, K-nearest Neighbor (KNN), Decision Tree, Random Forest, Naïve Bayes, and Stochastic Gradient Descent on various datasets. Experimental results reveal that our method yields better performance in terms of accuracy, Matthew’s Correlation Coefficient (MCC) and F1-score over 16 benchmark datasets, as compared to other competent methods. The superiority of our proposed method is that it uses an effective objective function that combines both JMI and MI to choose the relevant and non redundant features.

Efficient cross-validation traversals in feature subset selection

Sparse and robust classification models have the potential for revealing common predictive patterns that not only allow for categorizing objects into classes but also for generating mechanistic hypotheses. Identifying a small and informative subset of features is their main ingredient. However, the exponential search space of feature subsets and the heuristic nature of selection algorithms limit the coverage of these analyses, even for low-dimensional datasets. We present methods for reducing the computational complexity of feature selection criteria allowing for higher efficiency and coverage of screenings. We achieve this by reducing the preparation costs of high-dimensional subsets {mathscr {O}}({n}m^2) to those of one-dimensional ones {mathscr {O}}(m^2). Our methods are based on a tight interaction between a parallelizable cross-validation traversal strategy and distance-based classification algorithms and can be used with any product distance or kernel. We evaluate the traversal strategy exemplarily in exhaustive feature subset selection experiments (perfect coverage). Its runtime, fitness landscape, and predictive performance are analyzed on publicly available datasets. Even in low-dimensional settings, we achieve approximately a 15-fold increase in exhaustively generating distance matrices for feature combinations bringing a new level of evaluations into reach.

Joint Random Forest and Particle Swarm Optimization for Predictive Pathloss Modeling of Wireless Signals from Cellular Networks

The accurate and reliable predictive estimation of signal attenuation loss is of prime importance in radio resource management. During wireless network design and planning, a reliable path loss model is required for optimal predictive estimation of the received signal strength, coverage, quality, and signal interference-to-noise ratio. A set of trees (100) on the target measured data was employed to determine the most informative and important subset of features, which were in turn employed as input data to the Particle Swarm (PS) model for predictive path loss analysis. The proposed Random Forest (RF-PS) based model exhibited optimal precision performance in the real-time prognostic analysis of measured path loss over operational 4G LTE networks in Nigeria. The relative performance of the proposed RF-PS model was compared to the standard PS and hybrid radial basis function-particle swarm optimization (RBF-PS) algorithm for benchmarking. Generally, results indicate that the proposed RF-PS model gave better prediction accuracy than the standard PS and RBF-PS models across the investigated environments. The projected hybrid model would find useful applications in path loss modeling in related wireless propagation environments.

Semi-supervised feature selection for partially labeled mixed-type data based on multi-criteria measure approach

In many real applications, the data are always collected from different types and they are subjected to obtain partial labeling information of objects. Such data are referred to as partially labeled mixed-type data. There is currently few work on feature selection approaches for these data. Motivated by this issue, this paper aims at selecting the informative feature subset from partially labeled mixed-type data. At first, to improve the classification performance, an improved label propagation algorithm based on K-nearest neighbor is proposed, which marks the decision labels of unlabeled objects by making use of the information between unlabeled objects and labeled objects. On this basis, a feature multi-criteria measure based on the dependency, information entropy and information granulation is proposed for selecting candidate features. Finally, the corresponding semi-supervised feature selection algorithm is developed to select the feature subset for the partially labeled mixed-type data. Experimental results on UCI data sets demonstrate the effectiveness of the proposed feature selection algorithm and the superiority in terms of the classification accuracy compared with other algorithms.

Assessing the Added Value of Sentinel-1 PolSAR Data for Crop Classification

Crop classification is an important remote sensing task with many applications, e.g., food security monitoring, ecosystem service mapping, climate change impact assessment, etc. This work focuses on mapping 10 crop types at the field level in an agricultural region located in the Spanish province of Navarre. For this, multi-temporal Synthetic Aperture Radar Polarimetric (PolSAR) Sentinel-1 imagery and multi-spectral Sentinel-2 data were jointly used. We applied the Cloude–Pottier polarimetric decomposition on PolSAR data to compute 23 polarimetric indicators and extracted vegetation indices from Sentinel-2 time-series to generate a big feature space of 818 features. In order to assess the relevance of the different features for the crop mapping task, we run a number of scenarios using a Support Vector Machines (SVM) classifier. The model that was trained using only the polarimetric data demonstrates a very promising performance, achieving an overall accuracy over 82%. A genetic algorithm was also implemented as a feature selection method for deriving an optimal feature subset. To showcase the positive effect of using polarimetric data over areas suffering from cloud coverage, we contaminated the original Sentinel-2 time-series with simulated cloud masks. By incorporating the genetic algorithm, we derived a high informative feature subset of 120 optical and polarimetric features, as the corresponding classification model increased the overall accuracy by 5% compared to the model trained only with Sentinel-2 features. The feature importance analysis indicated that apart from the Sentinel-2 spectral bands and vegetation indices, several polarimetric parameters, such as Shannon entropy, second eigenvalue and normalised Shannon entropy are of high value in identifying crops. In summary, the findings of our study highlight the significant contribution of Sentinel-1 PolSAR data in crop classification in areas with frequent cloud coverage and the effectiveness of the genetic algorithm in discovering the most informative features.

Neurodynamics-driven holistic approaches to semi-supervised feature selection

Feature selection is a crucial part of machine learning and pattern recognition, which aims at selecting a subset of informative features from the original dataset. Because of label information, supervised feature selection performs better than unsupervised feature selection without label information. However, in the presence of a small number of labeled data and a large number of unlabeled data, it is challenging for supervised feature selection methods to select relevant features. In this paper, we propose three neurodynamics-driven holistic approaches to semi-supervised feature selection via semi-supervised feature redundancy minimization and semi-supervised feature relevancy maximization. We first define information-theoretic semi-supervised similarity coefficient matrix and semi-supervised feature relevancy vector based on multi-information, unsupervised symmetric uncertainty, and entropy to measure feature redundancy and relevancy. We then formulate a fractional programming problem and an iteratively weighted quadratic programming problem based on the semi-supervised similarity coefficient matrix and semi-supervised feature relevancy vector for semi-supervised feature selection. To solve the formulated problems, we delineate three neurodynamic optimization approaches based on two projection neural networks. We elaborate on the experimental results on six benchmark datasets to demonstrate the superior classification performance of the proposed neurodynamic approaches against six existing supervised and semi-supervised feature selection methods.