Feature Subset Research Articles

A New Performance Metric to Evaluate Filter Feature Selection Methods in Text Classification

High dimensionality and sparsity are the primary issues in text classification. Using feature selection approaches, the most effective way to solve the problem is to select a subset of features. The most common and effective methods used for this process are filter techniques. Various performance metrics such as Micro-F1, Macro-F1, and Accuracy are used to evaluate the performance of filter methods used for feature selection on datasets  Such methods work depending on a classification algorithm. However, when selecting features in filter techniques, the information on the individual features is evaluated without considering the relationship between the features. In such an approach, the actual performance of the filter technique used in feature selection may not be determined. In such a case, it causes the existing methods to be insufficient in testing the validity of the proposed method. For this purpose, this study suggests a novel performance metric called Selection Error (SE) to determine the actual performance evaluation of filter techniques. The Selection Error metric allows us to analyze the information value of the selected features more accurately than existing methods without relying on a classifier. The feature selection performance of the filtering approaches was performed on six different datasets with both The Selection Error and traditional performance metrics. When the results are examined, it is seen that there is a strong relationship between the proposed performance metric and the classification performance metric results. The Selection Error aims to significantly contribute to the literature by demonstrating the success of filtering feature selection methods, regardless of classifier performance. 

Gaze-Based Detection of Thoughts across Naturalistic Tasks Using a PSO-Optimized Random Forest Algorithm.

One key aspect of the human experience is our ongoing stream of thoughts. These thoughts can be broadly categorized into various dimensions, which are associated with different impacts on mood, well-being, and productivity. While the past literature has often identified eye movements associated with a specific thought dimension (task-relatedness) during experimental tasks, few studies have determined if these various thought dimensions can be classified by oculomotor activity during naturalistic tasks. Employing thought sampling, eye tracking, and machine learning, we assessed the classification of nine thought dimensions (task-relatedness, freely moving, stickiness, goal-directedness, internal-external orientation, self-orientation, others orientation, visual modality, and auditory modality) across seven multi-day recordings of seven participants during self-selected computer tasks. Our analyses were based on a total of 1715 thought probes across 63 h of recordings. Automated binary-class classification of the thought dimensions was based on statistical features extracted from eye movement measures, including fixation and saccades. These features all served as input into a random forest (RF) classifier, which was then improved with particle swarm optimization (PSO)-based selection of the best subset of features for classifier performance. The mean Matthews correlation coefficient (MCC) values from the PSO-based RF classifier across the thought dimensions ranged from 0.25 to 0.54, indicating above-chance level performance in all nine thought dimensions across participants and improved performance compared to the RF classifier without feature selection. Our findings highlight the potential of machine learning approaches combined with eye movement measures for the real-time prediction of naturalistic ongoing thoughts, particularly in ecologically valid contexts.

Hybrid framework for membrane protein type prediction based on the PSSM

Membrane proteins are considered the major source of drug targets and are indispensable for drug design and disease prevention. However, traditional biomechanical experiments are costly and time-consuming; thus, many computational methods for predicting membrane protein types are gaining popularity. The position-specific scoring matrix (PSSM) method is an excellent method for describing the evolutionary information of protein sequences. In this study, we propose an improved capsule neural network (ICNN) model based on a capsule neural network to acquire sufficient relevant information from the PSSM. Furthermore, accounting for the complementarity between traditional machine learning and deep learning, we propose a hybrid framework that combines both approaches to predict protein types. This framework trains 41 baseline models based on the PSSM. The optimal subset features, selected after traversal, are fused using a two-level decision-level feature fusion approach. Subsequently, comparisons are made using three combined strategies within an ensemble learning framework. The experimental results demonstrate that solely relying on PSSM input, the proposed method not only surpasses the optimal methods by 1.52%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, 2.26%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} and 2.67%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} on Dataset1, Dataset2, and Datasets3, respectively, but also exhibits superior generalizability. Furthermore, the code and dataset can be free download at https://github.com/ruanxiaoli/membrane-protein-types.

Test-retest repeatability of myocardial radiomic features from quantitative cardiac magnetic resonance T1 and T2 mapping

Radiomics of cardiac magnetic resonance (MR) imaging has proved to be potentially useful in the study of various myocardial diseases. Therefore, assessing the repeatability degree in radiomic features measurement is of fundamental importance.The aim of this study was to assess test-retest repeatability of myocardial radiomic features extracted from quantitative T1 and T2 maps.A representative group of 24 subjects (mean age 54 ± 18 years) referred for clinical cardiac MR imaging were enrolled in the study. For each subject, T1 and T2 mapping through MOLLI and T2-prepared TrueFISP acquisition sequences, respectively, were performed at 1.5 T. Then, 98 radiomic features of different classes (shape, first-order, second-order) were extracted from a region of interest encompassing the whole left ventricle myocardium in a short axis slice. The repeatability was assessed performing different and complementary analyses: intraclass correlation coefficient (ICC) and limits of agreement (LOA) (i.e., the interval within which 95% of the percentage differences between two repeated measures are expected to lie).Radiomic features were characterized by a relatively wide range of repeatability degree in terms of both ICC and LOA. Overall, 44.9% and 38.8% of radiomic features showed ICC values > 0.75 for T1 and T2 maps, respectively, while 25.5% and 23.4% of radiomic features showed LOA between ±10%. A subset of radiomic features for T1 (Mean, Median, 10Percentile, 90Percentile, RootMeanSquared, Imc2, RunLengthNonUniformityNormalized, RunPercentage and ShortRunEmphasis) and T2 (MaximumDiameter, RunLengthNonUniformityNormalized, RunPercentage, ShortRunEmphasis) maps presented both ICC > 0.75 and LOA between ±5%.Overall, radiomic features extracted from T1 maps showed better repeatability performance than those extracted from T2 maps, with shape features characterized by better repeatability than first-order and textural features. Moreover, only a limited subset of 9 and 4 radiomic features for T1 and T2 maps, respectively, showed high repeatability degree in terms of both ICC and LOA. These results confirm the importance of assessing test-retest repeatability degree in radiomic feature estimation and might be useful for a more effective/reliable use of myocardial T1 and T2 mapping radiomics in clinical or research studies.

A novel ensemble causal feature selection approach with mutual information and group fusion strategy for multi-label data

PurposeAs intelligent technology advances, practical applications often involve data with multiple labels. Therefore, multi-label feature selection methods have attracted much attention to extract valuable information. However, current methods tend to lack interpretability when evaluating the relationship between different types of variables without considering the potential causal relationship.Design/methodology/approachTo address the above problems, we propose an ensemble causal feature selection method based on mutual information and group fusion strategy (CMIFS) for multi-label data. First, the causal relationship between labels and features is analyzed by local causal structure learning, respectively, to obtain a causal feature set. Second, we eliminate false positive features from the obtained feature set using mutual information to improve the feature subset reliability. Eventually, we employ a group fusion strategy to fuse the obtained feature subsets from multiple data sub-space to enhance the stability of the results.FindingsExperimental comparisons are performed on six datasets to validate that our proposal can enhance the interpretation and robustness of the model compared with other methods in different metrics. Furthermore, the statistical analyses further validate the effectiveness of our approach.Originality/valueThe present study makes a noteworthy contribution to proposing a causal feature selection approach based on mutual information to obtain an approximate optimal feature subset for multi-label data. Additionally, our proposal adopts the group fusion strategy to guarantee the robustness of the obtained feature subset.

Inversion Method for Transformer Winding Hot Spot Temperature Based on Gated Recurrent Unit and Self-Attention and Temperature Lag.

The hot spot temperature of transformer windings is an important indicator for measuring insulation performance, and its accurate inversion is crucial to ensure the timely and accurate fault prediction of transformers. However, existing studies mostly directly input obtained experimental or operational data into networks to construct data-driven models, without considering the lag between temperatures, which may lead to the insufficient accuracy of the inversion model. In this paper, a method for inverting the hot spot temperature of transformer windings based on the SA-GRU model is proposed. Firstly, temperature rise experiments are designed to collect the temperatures of the entire side and top of the transformer tank, top oil temperature, ambient temperature, the cooling inlet and outlet temperatures, and winding hot spot temperature. Secondly, experimental data are integrated, considering the lag of the data, to obtain candidate input feature parameters. Then, a feature selection algorithm based on mutual information (MI) is used to analyze the correlation of the data and construct the optimal feature subset to ensure the maximum information gain. Finally, Self-Attention (SA) is applied to optimize the Gate Recurrent Unit (GRU) network, establishing the GRU-SA model to perceive the potential patterns between output feature parameters and input feature parameters, achieving the precise inversion of the hot spot temperature of the transformer windings. The experimental results show that considering the lag of the data can more accurately invert the hot spot temperature of the windings. The inversion method proposed in this paper can reduce redundant input features, lower the complexity of the model, accurately invert the changing trend of the hot spot temperature, and achieve higher inversion accuracy than other classical models, thereby obtaining better inversion results.

An improved ensemble learning model-based strategy for fault diagnosis of lithium battery double roller press equipment

Abstract The production process of lithium batteries is intricate, involving the coordination of various types of equipment.The sta-bility and precision of double roller press equipment directly affect product performance. With the increasing global de-mand for green energy, the application of lithium batteries in electric vehicles and energy storage systems is expanding, which imposes higher requirements on the stability and quality of lithium battery production. It is an important topic to address the challenges brought about by the gradual intelligentization of double roller presses, such as the complexifica-tion of control systems and the diversification of fault reasons. This paper proposes an enhanced ensemble learning model-based fault diagnosis strategy for lithium battery double roller press equipment. Firstly, the K-nearest neighbors (KNN) algorithm is employed to handle missing data, combined with normalization and standardization methods to improve fea-ture processing, thereby enhancing data quality. Secondly, the Maximum Information Coefficient (MIC) algorithm is utilized to select features highly correlated with fault labels, combined with the Recursive Feature Elimination with Cross-Validation (RFECV) to further optimize feature selection, creating an optimal feature subset. Finally, a RXS-XGBoost model is constructed through the Stacking ensemble learning method, selecting Random Forest (RF), XGBoost, and Sup-port Vector Machines (SVM) as base learners, with XGBoost as the meta-learner. This ensemble approach aims to lever-age the complementary advantages of different algorithms, enhancing the accuracy and robustness of fault diagnosis. The experimental results demonstrate that this improved ensemble learning diagnostic strategy achieves an accuracy rate of up to 99.05%, which is significantly better than other fault diagnosis strategies. It not only effectively reduces the model's training complexity and the risk of overfitting but also significantly enhances the efficiency and precision of fault diagno-sis for lithium battery double roller press equipment.

Robust weighted fuzzy margin-based feature selection with three-way decision

Feature selection has shown noticeable benefits to the tasks of machine learning and data mining, and an extensive variety of feature selection methods has been proposed to remove redundant and irrelevant features. However, most of the existing methods aim to find a feature subset to perfectly fit data with the minimum empirical risk, thus causing the problems of overfitting and noise sensitivity. In this study, a robust weighted fuzzy margin-based feature selection is proposed for uncertain data with noise. Concretely, a robust weighted fuzzy margin based on fuzzy rough sets is first introduced to evaluate the significance of different features. Then, a gradient ascent algorithm based on the noise filtering strategy and three-way decision is developed to optimize the sample and feature weights to further enlarge the fuzzy margin. Finally, an adaptive feature selection algorithm based on the robust weighted fuzzy margin is presented to generate an optimal feature subset with a large margin. Extensive experiments on the UCI benchmark datasets show that the proposed method could obtain high-quality feature subsets and outperform other representative methods under different noise rates.

A fast dual-module hybrid high-dimensional feature selection algorithm

When dealing with large-scale datasets, high-dimensional feature selection plays a crucial role in improving the performance and interpretability of machine learning models. However, it still faces the problems of the “dimensionality curse” and high computational cost in dealing with high-dimensional datasets. In this paper, we develop a dual-module hybrid feature selection algorithm based on correlation filtering and a multi-objective evolutionary algorithm based on dynamic variation distance (HMOFS-CFDVD), which aims to reduce the computational cost of high-dimensional features and search space. Firstly, a coarse-grained feature filtering method based on correlation is proposed, enabling the algorithm to quickly identify potentially better feature subsets in the later stages. After that, a fine-grained feature selection is further implemented based on the multi-objective evolutionary algorithm with sample variation distance to obtain a high-quality feature subset. This stage incorporates a novel population initialization method with self-regulation, an adaptive evolution strategy, an optimal sample selection mechanism based on dynamic sample distance, and an improved diversity mechanism to enhance the evolutionary performance. Moreover, the feature relevance metric is introduced as a third objective to improve the overall algorithm's performance. Experimental results on 12 high-dimensional feature datasets demonstrate that the proposed HMOFS-CFDVD achieves high accuracy and produces a smaller subset of features.

FSBOA: feature selection using bat optimization algorithm for software fault detection

Feature selection (FS) plays a crucial role in software fault prediction (SFP), aiming to identify a subset of relevant and discriminative features from a large pool of software metrics. It serves as a critical preprocessing step in building accurate fault prediction models, enabling the identification of potential software faults early in the development life cycle and facilitating effective resource allocation for testing and maintenance activities. The study's objective is to determine how well the bat optimization algorithm (BOA) can extract the features that are most important for correctly predicting software flaws, improve the accuracy of fault prediction, reduce the dimensionality of the feature space, and mitigate the risk of overfitting, thereby enabling more effective resource utilization and better allocation of testing efforts. The forecasting models underwent testing and training utilizing a collection of software metrics, with the datasets undergoing evaluation using several different FS algorithms. An assessment was conducted by contrasting the effectiveness of multiple optimization algorithms, including evolutionary methods such as FS employing genetic algorithm (FSGA), FS employing differential evolution (FSDE), and swarm-based techniques such as FS employing ant colony optimization (FSACO), FS employing particle swarm optimization (FSPSO), FS employing firefly algorithm (FSFA), and FS employing binary grey wolf optimization algorithm (FSBGWO) in relation to FS employing bat optimization algorithm (FSBAO). The results obtained from FSBAO approach demonstrate the effectiveness in solving FS optimization problems with at most accuracy of 98.92%. Furthermore, the experimental results have been statistically validated for the greater efficiency of the proposed FSBAO algorithm. This study's findings have crucial implications for developing a software failure prediction models that is more accurate and efficient.

A data-driven approach to agricultural machinery working states analysis during ploughing operations

In the field of precision agriculture, there is a significant shift towards data-driven methodologies that considerably enhance the efficiency and sustainability of agricultural operations. This research investigates the application of CAN-Bus and GNSS data to develop a comprehensive analysis of agricultural machinery's operational states during ploughing operations through advanced data analytics techniques, including machine learning. The primary tool utilized in this study is the Random Forest classifier, a robust algorithm well-suited for handling the complexity and volume of data typical in modern agricultural settings. The study evaluates Random Forest models trained on various feature subsets to accurately identify different operational states of agricultural machinery, including idle, moving, turning, and working states. By merging CAN-Bus data, which capture real-time operational parameters, with GNSS data, providing spatial and temporal context, it is possible to achieve a comprehensive understanding of machinery behaviour and its interaction with field conditions. This integration significantly enhances decision-making capabilities in farm management, leading to more effective and efficient operations.Furthermore, the findings from this study contribute to the broader agricultural community by illustrating how data-driven approaches can harness the vast amounts of data generated by modern agricultural machinery. This research underscores the potential of machine learning modelsnot only to interpret complex data sets but also to transform these insights into actionable knowledge, which can lead to more precise and sustainable agricultural practices. Overall, this study offers a systematic approach for analysing agricultural data and lays the groundwork for future advancements in incorporating machine learning and IoT technologies into the agricultural sector. This aims to enhance productivity and sustainability in farming practices.

Monitoring soil nutrients using machine learning based on UAV hyperspectral remote sensing

ABSTRACT Unmanned aerial vehicles (UAV) are rapidly evolving experimental platforms that play an important role in remote sensing. In this study, we investigated a machine learning method for monitoring soil nutrient content using UAV hyperspectral remote sensing. We employed machine learning techniques for feature extraction and soil hyperspectral information modelling. In contrast to traditional mathematical transformation methods, we adopted a combination of random forest and differential evolution algorithms to rank the weights of individual hyperspectral data, thereby obtaining a series of spectral feature subsets for soil organic matter, total nitrogen and available phosphorus and potassium. Furthermore, the analytic hierarchy process was used for weight analysis, and the characteristic bands of the four soil nutrients were successfully extracted. Next, a quantitative inversion model based on a back-propagation (BP) neural network was established to estimate soil nutrient content, with determination coefficients higher than 0.7 and 0.6 for the modelling and verification sets, respectively. The relative percent difference values were greater than 2, among which the highest was for available potassium, with determination coefficients of 0.95 and 0.84 for the modelling and verification sets, respectively. In addition, visualization distribution maps of soil nutrients were obtained by combining the BP model and original reflectance hyperspectral images, and the comparisons of content histograms showed a relatively consistent distribution between the sampling and inversion points. The results verified the effectiveness of the combined machine learning method for large-scale and high-precision monitoring and visualization of soil nutrient contents.

Integration of simulated annealing into pigeon inspired optimizer algorithm for feature selection in network intrusion detection systems.

In the context of the 5G network, the proliferation of access devices results in heightened network traffic and shifts in traffic patterns, and network intrusion detection faces greater challenges. A feature selection algorithm is proposed for network intrusion detection systems that uses an improved binary pigeon-inspired optimizer (SABPIO) algorithm to tackle the challenges posed by the high dimensionality and complexity of network traffic, resulting in complex models, reduced accuracy, and longer detection times. First, the raw dataset is pre-processed by uniquely one-hot encoded and standardized. Next, feature selection is performed using SABPIO, which employs simulated annealing and the population decay factor to identify the most relevant subset of features for subsequent review and evaluation. Finally, the selected subset of features is fed into decision trees and random forest classifiers to evaluate the effectiveness of SABPIO. The proposed algorithm has been validated through experimentation on three publicly available datasets: UNSW-NB15, NLS-KDD, and CIC-IDS-2017. The experimental findings demonstrate that SABPIO identifies the most indicative subset of features through rational computation. This method significantly abbreviates the system's training duration, enhances detection rates, and compared to the use of all features, minimally reduces the training and testing times by factors of 3.2 and 0.3, respectively. Furthermore, it enhances the F1-score of the feature subset selected by CPIO and Boost algorithms when compared to CPIO and XGBoost, resulting in improvements ranging from 1.21% to 2.19%, and 1.79% to 4.52%.

Biclustering Multivariate Longitudinal Data with Application to Recovery Trajectories of White Matter After Sport-Related Concussion.

Biclustering is the task of simultaneously clustering the samples and features of a data set. In doing so, subsets of samples that exhibit similar behaviors across subsets of features can be identified. Motivated by a longitudinal diffusion tensor imaging study of sport-related concussion (SRC), we present the problem of biclustering multivariate longitudinal data in which subjects and features are grouped simultaneously based on longitudinal patterns rather than magnitude. We propose a penalized regression based method for solving this problem by exploiting the heterogeneity in the longitudinal patterns within subjects and features. We evaluate the performance of the proposed methods via a simulation study and apply them to the motivating dataset, revealing distinctive patterns of white-matter abnormalities within subgroups of SRC cases.

Computational Cost Reduction in Multi-Objective Feature Selection Using Permutational-Based Differential Evolution

Feature selection is a preprocessing step in machine learning that aims to reduce dimensionality and improve performance. The approaches for feature selection are often classified according to the evaluation of a subset of features as filter, wrapper, and embedded approaches. The high performance of wrapper approaches for feature selection is associated at the same time with the disadvantage of high computational cost. Cost-reduction mechanisms for feature selection have been proposed in the literature, where competitive performance is achieved more efficiently. This work applies the simple and effective resource-saving mechanisms of the fixed and incremental sampling fraction strategies with memory to avoid repeated evaluations in multi-objective permutational-based differential evolution for feature selection. The selected multi-objective approach is an extension of the DE-FSPM algorithm with the selection mechanism of the GDE3 algorithm. The results showed high resource savings, especially in computational time and the number of evaluations required for the search process. Nonetheless, it was also detected that the algorithm’s performance was diminished. Therefore, the results reported in the literature on the effectiveness of the strategies for cost reduction in single-objective feature selection were only partially sustained in multi-objective feature selection.

Integrating metaheuristics and artificial intelligence for healthcare: basics, challenging and future directions

Accurate and rapid disease detection is necessary to manage health problems early. Rapid increases in data amount and dimensionality caused challenges in many disciplines, with the primary issues being high computing costs, memory costs, and low accuracy performance. These issues will arise since Machine Learning (ML) classifiers are mostly used in these fields. However, noisy and irrelevant features have an impact on ML accuracy. Therefore, to choose the best subset of features and decrease the dimensionality of the data, Metaheuristics (MHs) optimization algorithms are applied to Feature Selection (FS) using various modalities of medical imaging or disease datasets with different dimensions. The review starts by giving a general overview of the many approaches to AI algorithms, followed by a general overview of the various MH algorithms for healthcare applications, an analysis of MHs boosted AI for healthcare applications, and using a wide range of research databases as a data source for access to numerous field publications. The final section of this review discusses the problems and challenges facing healthcare application development.

Myelodysplastic syndrome risk assessment using priority linked correlated feature set using ResNet50

Myelodysplastic syndrome (MDS) is more common as people get older, and the best way to treat these conditions is to catch them early. Acute myeloid leukemia (AML) is a more aggressive form of MDS; about 30 % of cases advance to this disease. The key to intervening before MDS develops into AML is early identification. If doctors can diagnose MDS early, they can start treatment sooner, which could slow the disease’s progression and enhance patients’ quality of life. But because it is so similar to other conditions, MDS is frequently misdiagnosed. Early detection of MDS can let medical experts to provide better treatment, which may slow the disease’s progression and boost survival rates. Assessment of Bone Marrow (BM) histopathology is used to diagnose MDS and myeloproliferative neoplasm (MPN) by identifying dysplastic cellular shape, cellularity, and blast excess. Deep Learning (DL) algorithms can account for the genome’s numerous complicated interconnections, allowing for the creation of more accurate prognostication models. The ResNet50 network consists of 50 layers: 48 convolutional layers, 1 MaxPool layer, and 1 average pool layer. This research presents a Priority Linked Correlated Feature Set using ResNet50 (PLCFS-ResNet50) for risk assessment of MDS. The proposed model uses the feature subset for training the Resnet50 model and for risk assessment so that proper diagnosis can be done in advance. The proposed model achieves 98.6 % accuracy in feature set generation and 98.7 % accuracy in risk assessment of MDS. The proposed model when contrasted with the traditional model exhibits better prediction levels.

Maximum pitting corrosion depth prediction of buried pipeline based on theory-guided machine learning

Buried pipelines are crucial for the transportation of oil and natural gas resources. However, pipeline failure accidents have frequently occurred due to corrosion. Therefore, an accurate corrosion depth prediction model is necessary for the reliable supply of energy. In this paper, a theory-guided machine learning (ML) model is developed for maximum pitting corrosion depth prediction, the engineering theory and domain knowledge are integrated into feature space to improve the model interpretability. Firstly, several new feature variables are constructed based on the interactions between independent variables. Then, feature importance of all feature variables is obtained using random forest (RF). A hybrid multi-objective grey wolf optimization (HMOGWO) is proposed to optimize the hyper-parameters of RF model, considering feature number, prediction accuracy, and stability simultaneously. Finally, a comprehensive pitting corrosion dataset is utilized for performance evaluation. The results indicate that the proposed theory-guided model can achieve high prediction accuracy and stability, the optimal feature subset can be determined using multi-objective optimization method simultaneously, which solves the problems of model interpretability and feature selection for traditional ML models with the single-objective optimizer. This study is of great significance to the transportation safety of buried pipelines.

A Four-Grade Astrocytoma Classification System with Optimal Feature Selection Using Ant Colony Optimization Algorithm

An efficient Four-Grade Astrocytoma brain tumor classification is proposed in this research work. The work highlights the optimal feature selection using a hybrid combination of Ant Colony Optimization (ACO) algorithm and machine learning techniques, leading to better classification accuracy. A diverse set of features including spatial and frequency domain features will provide better discrimination of patterns. ACO provides a robust and efficient mechanism for exploring the vast space of possible feature subsets. A total of 135 features are extracted including Tamura features, Gabor Wavelet Features, Coiflet Wavelet Coefficients, Speeded Up Robust Features, and GLCM features. Ant Colony Optimization is used to find out the optimum set of features. The selected feature set is used for training a Decision Tree classifier, where the Astrocytoma MRI images are categorized into four grades of astrocyoma namely, pilocytic astrocytoma, diffuse astrocytoma, anaplastic astrocytoma and glioblastoma multiforne. Classification is also performed using KNN classifier, SVM classifier and Ensemble Bagged Tree Classifier. Performance metrics like accuracy, sensitivity, specificity, precision and FScore are utilized to evaluate the proposed system’s performance. It is found out that the decision tree classifier outperforms the other classifiers. Thus this integrated approach of ACO with machine learning classifiers results in improved Astrocytoma classification performance aiding clinicians in making accurate treatment decisions.

A Classification Method for Incomplete Mixed Data Using Imputation and Feature Selection

Data missing is a ubiquitous problem in real-world systems that adversely affects the performance of machine learning algorithms. Although many useful imputation methods are available to address this issue, they often fail to consider the information provided by both features and labels. As a result, the performance of these methods might be constrained. Furthermore, feature selection as a data quality improvement technique has been widely used and has demonstrated its efficiency. To overcome the limitation of imputation methods, we propose a novel algorithm that combines data imputation and feature selection to tackle classification problems for mixed data. Based on the mean and standard deviation of quantitative features and the selecting probabilities of unique values of categorical features, our algorithm constructs different imputation models for quantitative and categorical features. Particle swarm optimization is used to optimize the parameters of the imputation models and select feature subsets simultaneously. Additionally, we introduce a legacy learning mechanism to enhance the optimization capability of our method. To evaluate the performance of the proposed method, seven algorithms and twelve datasets are used for comparison. The results show that our algorithm outperforms other algorithms in terms of accuracy and F1 score and has reasonable time overhead.