Search

A critical aspect of software engineering is Software fault prediction which aims to identify and prevent errors in software systems before their release which can cause failures or issues for its users. Various techniques and tools have been developed to detect software faults, including static code analysis, dynamic testing, and machine learning-based approaches. In past few years, the world has seen a growing interest in the use of ML models for predicting software faults, as they can effectively analyse high dimensional datasets and detect complex patterns which are difficult for human experts to detect. However, developing accurate and reliable software fault detection models requires careful selection of data, feature engineering, and model evaluation. This purpose of this paper is to present a comprehensive analysis of potential applications and future research directions in the field of software fault detection. The study emphasizes the importance of identifying and addressing software faults to ensure the reliability and efficiency of software systems. Additionally, the paper outlines various approaches and techniques that can be employed for effective software fault detection. [ABSTRACT FROM AUTHOR]

Feature Selection (FS) is critically important for optimising the performance of Intrusion Detection Systems (IDSs) used in Wireless Sensor Networks (WSNs). However, selecting an optimal number of relevant features from massive IDS data sets has been an ongoing FS research problem. Many approaches use Machine Learning (ML), such as nature-inspired population-based and Differential Evolution (DE)-based algorithms for FS. However, the main drawback of DE is that it has a premature convergence issue, which leads to false convergence. In this paper, a novel mutation strategy is proposed to alleviate premature convergence and to develop a modified DE-based FS algorithm called DE with maturity extension (DE-ME). The novel mutation is achieved by changing the mutation factor in the DE\rnd\1 based on K-Nearest Neighbour (KNN) as the most efficient implementation without adding further complexity to the design process. The proposed DE-ME algorithm improves the FS operation with significant performance improvements in terms of overcoming premature convergence and increasing accuracy for selection. The DE-ME algorithm was used to select the most effective features from 41 features in the Network Security Laboratory Knowledge Discovery in Databases (NSL-KDD) dataset. The accuracy of DE-ME is high at 99.66 %, its False Positive Rate (FPR) is low at 0.464 %, and the true positive rate (TPR) achieved is 99.80 %. In addition, the number of features selected by DE-ME is 6 out of 41 features in the dataset. [ABSTRACT FROM AUTHOR]

• A feature selection method based on binary dandelion algorithm is proposed. • The algorithm applies a sine-cosine operator and a restart strategy. • Mutual information and quick bit mutation improve the performance of the algorithm. • The algorithm performs well on datasets with larger dimensions. Feature selection (FS) is an important data preprocessing technology for machine learning and data mining. Metaheuristic algorithm (MH) has been widely used in feature selection because of its powerful search function. This paper presents an improved Binary Dandelion Algorithm using Sine Cosine operator and Restart strategy (SCRBDA) for feature selection. First, the sine cosine operator is used in the radius formula of the core dandelions (CD), which significantly enhances the ability of algorithm development and exploration. Secondly, the algorithm uses a restart strategy to increase its ability to get rid of local optimum. Thirdly, mutual information is used to guide the generation of some dandelions, which pays more attention to the correlation between the selected features and categories. Finally, quick bit mutation is used as the mutation strategy to improve the diversity of the population. The SCRBDA proposed in this paper was tested on 18 datasets of different sizes from UCI machine learning database. The SCRBDA was compared with 8 other classical feature selection algorithms, and the performance of the proposed algorithm was evaluated through feature subset size, classification accuracy, fitness value, and F1-score. The experimental results show that SCRBDA achieves the best performance, which has stronger feature reduction ability and achieves better overall performance on most datasets. Especially on large-scale datasets, SCRBDA can obtain extremely smaller feature subsets while maintaining much higher classification accuracy, and satisfactory F1-score. [ABSTRACT FROM AUTHOR]

This paper proposes a new, novel ranking method called Inverse-Reinforcement Learning Ranking. The main goal is to find a reward function representing the user's perceived utility after clicking on each result. It is necessary to utilize log information of all users' queries in the search engine dataset to reach this goal while assuming that the decisions (clicks) of the users are the best (optimal policy). The respective reward function is constructed using features extracted through a feature selection, and their corresponding weights are obtained by two optimization models, which are applicable for ranking results represented to the new users. In addition, new performance criteria were developed to illustrate the performance of the presented ranking method. To evaluate and test the proposed ranking algorithm, a real medium-sized dataset from a search engine is preprocessed and used in this research. Findings show promising results and decisive superiority over the default ranking method. It is illustrated that clicks on the top five results, top ten, and even the first results are remarkably improved by about 13–19% in all experiments, and the perplexity remarkably decreases by almost 23% after applying the ranking method. [ABSTRACT FROM AUTHOR]

Highlights • Fault prediction improves the effectiveness of software quality assurance activities. • This paper focuses on building an effective fault prediction classifier. • Fault prediction model using Iterated feature selection algorithms with L-RNN. • We perform experiments on 19 open source projects. • Fault prediction model is best suitable for projects with faulty classes less than the threshold value. Abstract Software fault prediction (SFP) is typically used to predict faults in software components. Machine learning techniques (e.g., classification) are widely used to tackle this problem. With the availability of the huge amount of data that can be obtained from mining software historical repositories, it becomes possible to have some features (metrics) that are not correlated with the faults, which consequently mislead the learning algorithm and thus decrease its performance. One possible solution to eliminate those metrics is Feature Selection (FS). In this paper, a novel FS approach is proposed to enhance the performance of a layered recurrent neural network (L-RNN), which is used as a classification technique for the SFP problem. Three different wrapper FS algorithms (i.e, Binary Genetic Algorithm (BGA), Binary Particle Swarm Optimization (BPSO), and Binary Ant Colony Optimization (BACO)) were employed iteratively. To assess the performance of the proposed approach, 19 real-world software projects from PROMISE repository are investigated and the experimental results are discussed. Receiver operating characteristic - area under the curve (ROC-AUC) is used as a performance measure. The results are compared with other state-of-art approaches including Naïve Bayes (NB), Artificial Neural Network (ANN), logistic regression (LR), the k-nearest neighbors (k-NN) and C4.5 decision trees, in terms of area under the curve (AUC). Our results have demonstrated that the proposed approach can outperform other existing methods. [ABSTRACT FROM AUTHOR]

Abstract Dealing with high-dimensional dataset has always been an important problem and feature selection is one of useful tools. In this paper, we develop a new filter based supervised feature selection method by combining Trace Ratio Criterion of Linear Discriminant Analysis (TRC-LDA) and group sparsity regularization. The filter based supervised feature selection method is a classifier-independent method while the TRC-LDA criterion is a recently developed criterion for dimensionality reduction that can well preserve discriminative information of dataset. However, there are seldom methods by utilizing TRC-LDA criterion for feature selection. On the other hand, imposing the l2,0-norm to the projection matrix of TRC-LDA will force some rows in it to be zero while keep other rows nonzero making the index of nonzero rows to be the selected features, however, l 2,0 -nom minimizing problem is NP-hard and intractable. To solve the above problem, in this paper, we develop a new method, namely, Trace Ratio Criterion Discriminative Feature Selection (TRC-DFS), for feature selection. The proposed TRC-DFS has imposed l 2,1 -norm, i.e. an approximation of l 2,0 -norm, to the projection matrix W of TRC-LDA to achieve feature selection. As a result, the proposed TRC-DFS can both achieve feature selection as well as capture the discriminative structure of data. We also extend the proposed method with l 2, p -norm (0 < p < 1) regularization to grasp more sparse property and develop an iterative approach to calculate the optimal solution, which is rigorously proved to be converged. Extensive simulations based on several real-world datasets verify the effectiveness of the proposed methods. [ABSTRACT FROM AUTHOR]

This paper presents optimization techniques for automatic personality recognition (APR) based on Twitter in Bahasa Indonesia, the mother tongue of Indonesians. Foremost, we discuss Twitter and its utilization as a resource for many types of research. Several previous studies have been attempted to predict users’ personality automatically. However, only a few of them have done their research for Bahasa Indonesia data. Therefore, this paper discusses the optimization of APR in Bahasa Indonesia. We evaluate a series of techniques implementing hyperparameter tuning, feature selection, and sampling to improve the machine learning algorithms used. The personality prediction system is built on machine learning algorithms. There are three machine learning algorithms used in this study, namely Stochastic Gradient Descent (SGD), and two ensemble learning algorithms, Gradient Boosting (XGBoost), and stacking (super learner). By implementing this series of optimization techniques, the current study’s evaluation results show huge improvement by achieving 1.0 ROC AUC score with SGD and Super Learner. [ABSTRACT FROM AUTHOR]

• MMODE is developed as a new hybrid gene selection method for tumor identification. • MMODE combines multi-tasking and multi-objective frameworks. • MMODE uses a new elite strategy and a new guidance strategy. • MMODE selects a few genes and achieves high classification accuracy. A key preprocessing step in tumor recognition based on microarray expression profile data and machine learning is to identify tumor marker genes. Gene selection aims to select the most relevant gene subset from the original ultra-high dimensional microarray expression profile data to improve tumor identification performance. Inspired by evolutionary multitasking (EMT) and multi-objective optimization, this paper puts forward a novel multitasking multi-objective differential evolution gene selection algorithm (MMODE) which uses new elite and guidance strategies to select the best gene subsets. MMODE initializes two different populations according to different filtering criteria to increase the diversity of the search. These two populations guide their respective populations to search in the optimal direction through knowledge transfer in the evolutionary process. In addition, MMODE employs new elite and guidance strategies that enables individuals to narrow the search range and jump out of local optima. The proposed algorithm is validated on 13 publicly available microarray expression datasets in comparison with state-of-the-art gene selection algorithms. The experimental results show that MMODE can find smaller gene subsets and achieve higher classification accuracy compared with other algorithms. [ABSTRACT FROM AUTHOR]

Fuzzy rough set is a theoretical framework of fuzzy uncertainty management, and discernibility matrix offers a mathematical foundation for algorithm construction of feature learning. The approaches of fuzzy rough set and discernibility matrix have been successfully applied in single-label learning. However, few works have been done on investigating the foundation of fuzzy rough discernibility matrix on multi-label data. There will be two pivotal problems to be addressed when using fuzzy rough discernibility matrix for multi-label data analysis. One is how to extract sample-level and label-level correlations; and the other is how to utilize the discernibility matrix for algorithm construction. For this reason, in this paper the fuzzy rough discrimination matrix is introduced to deal with the problem of multi-label feature selection. First, the significance of labels in the label space is captured based on the label correlation. Labels with different significances contribute to different weights for measuring the similarity between samples. Hence, a sample similarity matrix in the label space is computed based on the label weighting strategy. Then, a framework of a fuzzy decision system is formalized, in which the discernibility matrix of fuzzy rough sets is introduced as a foundation to evaluate the sample discrimination ability of features. Under the discernibility matrix criterion, a multi-label learning algorithm is developed to select discriminative features from multi-label data. A series of experimental analysis verifies the effectiveness and efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

Sparse Bayesian learning (SBL) provides state-of-the-art performance in accuracy, sparsity and probabilistic prediction for classification. In SBL, the regularization priors are automatically determined that avoids an exhaustive hyperparameter selection by cross-validation. However, scalability to large problems is a drawback of SBL because of the inversion of a potentially enormous covariance matrix for updating the regularization priors in every iteration. This paper develops an approximate SBL algorithm called ARP-SBL, where the regularization priors are approximated without inverting the covariance matrix. Therefore, our approach can easily scale up to problems with large data size or feature dimension. It alleviates the long training time and high memory complexity in SBL. Based on ARP-SBL, two scalable nonlinear SBL models: scalable relevance vector machine (ARP-RVM) and scalable sparse Bayesian extreme learning machine (ARP-SBELM) are developed for problems of large data size and large feature size respectively. Experiments on a variety of benchmarks have shown that the proposed models are with competitive accuracy compared to existing methods while i) converging faster; ii) requiring thousands of times less memory; iii) without exhaustive regularized hyperparameter selection; and iv) easily scaling up to large data size and high dimensional features. [ABSTRACT FROM AUTHOR]

This paper investigates a new bacterial colony-based feature selection algorithm to improve the classification accuracy of customers for personalized products recommendation. An attribute learning strategy is developed in this study to update the feature related population. Specifically, the features can be weighted according to their historic contributions to both the individual- and group-based subsets. Additionally, the frequency of appearance is also recorded for the feature candidates to improve the diversity of feature distribution and avoid the over-fitting. Based on the weight-based feature indexes and frequency of appearance records, the performance of feature subsets are enhanced by replacing the features being repeatedly appeared in a same vector. To explore the feasibility of the proposed method for the missing feature problems, the objective of the optimization is to minimize the classification error using the acceptable number of features. K-Nearest Neighbor is employed as the learning technique to cooperate with the proposed feature selection method. The effectiveness of the proposed feature selection method is demonstrated by performing test on the datasets from UCI machine learning repository and real-world data from Amazon customer reviews of products. Compared with other seven feature selections methods, the proposed feature selection algorithm outperforms the other algorithms by achieving higher classification accuracy rate using smaller features. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is considered a challenging machine learning problem that handles the large size of features. The main purpose of FS is to remove irrelevant and redundant variables to improve the performance of the learning algorithms. Consequently, the FS process is considered as an optimization problem where metaheuristics approaches prove efficiency in solving it. In this paper, we propose new binary versions of a new ancient inspired metaheuristic approach called Giza Pyramids Construction (GPC) to select the most relevant subset of features. The proposed binary versions of the algorithm called BGPC-S and BGPC-V are implemented with two transfer functions, with the main objective of maximizing classification accuracy and minimizing the number of selected features. The two versions of BGPC were compared to six well-known binary metaheuristics for feature selection problem, namely Binary Atom Search Optimisation (BASO), Binary Bat Algorithm (BBA), Binary Differential Evolution (BDE), Binary Grey Wolf Optimizer (BGWO), Binary Particle Swarm Optimization (BPSO), and Binary Harris Hawks Optimizer (BHHO), and evaluated over 20 datasets from the UCI repository. Experiments have demonstrated that the proposed approaches outperformed the other algorithms in terms of classification accuracy and the number of selected features. [ABSTRACT FROM AUTHOR]

The sentiment classification method is a research field that is proliferating in Indonesia since it is fast in extracting public opinion and provides essential and valuable information for stakeholders. Of the best-performing sentiment classification approaches, machine learning is one of them that has excellent performance. However, the method has several problems, such as noisy features and high dimensionality of features that significantly affect the sentiment classification performance. Therefore, to overcome the problems, this paper presents a novel feature selection using a combination of Query Expansion Ranking (QER) and Genetic Algorithm-Support Vector Machine (GA-SVM) for improving sentiment classification performance. Based on the experimental results, the proposed method could significantly improve sentiment classification performance, outperform all state-of-the-art algorithms, and decrease computational time. The method achieved the best performance in average precision, recall, and f-measure with the value of 96.78%, 96.76%, and 96.75%, respectively. [ABSTRACT FROM AUTHOR]

Attribute reduction can remove data noise and redundancy, thus reducing computational complexity, which is very important for machine learning. Because the difference between nominal attribute values is difficult to measure, attribute reduction for hybrid data faces challenges. In addition, most of the existing methods are sensitive to noise due to the lack of an anti-noise mechanism. Decision attribute contains the most important information of data. This paper proposes some techniques that consider the above problems from the perspective of fuzzy evidence theory. First of all, a new distance incorporating decision attributes is defined, and then a new fuzzy relation with an anti-noise mechanism is defined. Furthermore, fuzzy belief and fuzzy plausibility are defined based on the defined new distance and new fuzzy relation. In this framework, two anti-noise attribute reduction algorithms for hybrid data are proposed. Experiments on 12 data sets of various types show that compared with the other 8 state-of-the-art algorithms, the proposed algorithms improve the classification accuracy by at least 2% and the anti-noise ability by at least 11%. Therefore, it can be concluded that the proposed algorithms have excellent anti-noise ability while maintaining good feature selection ability. [ABSTRACT FROM AUTHOR]

Hyperparameter optimization is a challenging process that has the potential to improve machine learning algorithms. Since it creates a remarkable computational burden for machine learning tasks, there have been few works coping with tuning strategies of a specific algorithm. In this paper, an improved Stochastic Gradient Descent (SGD) based on Fisher Maximization is developed for tuning hyperparameters of an Echo State Network (ESN) which has a wide range of applications. The results of the method are then compared with those of traditional Gradient Descent and Grid Search. According to the obtained results; 1) The scale of the data sets greatly affects the reliability of hyperparameter optimization results; 2) Feature selection is critical in terms of mean error of training when hyperparameter optimization is applied on some methods such as ESN; 3) SGD falls in a good local minima if Fisher Maximization is performed to find a good starting point. [ABSTRACT FROM AUTHOR]

• A comprehensive review of LL fault diagnosis methods in PV systems. • An ensemble learning model based on probabilistic strategy is developed. • The main features of the fault are extracted from I-V curves using a Simulink based model of PV arrays. • A feature selection algorithm is used for each learning algorithm to reduce the dataset and increase the classification accuracy. • The proposed model is very reliable and able to detect and classify all the LL faults accurately. The fault diagnosis of photovoltaic (PV) arrays aims to increase the reliability and service life of PV systems. Line-Line (LL) faults may remain undetected under low mismatch level and high impedance due to low currents of faults, resulting in power losses and fire potential disaster. This paper proposes a novel and intelligent fault diagnosis method based on an ensemble learning model and Current-Voltage (I-V) characteristics to detect and classify LL faults at the DC side of PV systems. For this purpose, first, the key features are extracted via analyzing I-V characteristics under various LL fault events and normal operation. Second, a feature selection algorithm has been applied to select the best features for each learning algorithm in order to reduce the amount of data required for the learning process. Third, an ensemble learning model is developed that combines several learning algorithms based on the probabilistic strategy to achieve superior diagnostic performance. Here, we find an excellent agreement between simulation and experimental results that the proposed method can obtain higher accuracy in detecting and classifying the LL faults, even under low mismatch levels and high fault impedances. In addition, the comparison results demonstrate that the performance of the proposed method is better than individual machine learning algorithms, so that the proposed method precisely detects and classifies LL faults on PV systems under the different conditions with an average accuracy of 99% and 99.5%, respectively. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is an important pre-processing step in data mining and pattern recognition. It can effectively compress the dimensionality of the feature space to reduce computation time and improve classification performance. The meta-heuristic algorithm-based feature selection method by finding the optimal set of features in the solution space has been widely used. However, this method is prone to trap into local optimality in a sufficiently large solution space. In this paper, we first propose a binary dandelion algorithm (BDA) to improve classification accuracy. In addition, to improve the performance of the algorithm, a binary dandelion algorithm using an improved seeding strategy and chaotic populations (SBDA) is proposed in this paper. Firstly, the strategy of optimizing the seeding radius by using the vibrational function and the historical optimal population increases the complexity of the search process and improves the search performance of the algorithm in the solution space. Secondly, when generating seeds, chaotic populations are generated using chaotic operators, which improves the ability of the algorithm to jump out of the local optimum and improves the stability of the algorithm. In this paper, 15 well-established datasets collected from the UCI machine learning database were adopted to compare four variants of BDA using only chaotic population improvement and in the next experiments, both mechanisms are verified to be effective in improving the performance of the algorithm. In addition, this paper compares the proposed BDA algorithm and SBDA algorithm with eight other classical algorithms. The experimental results show that SBDA can obtain fewer features with higher classification accuracy in most cases. • Proposed a BDA algorithm for solving binary problems. • Proposed a SBDA algorithm by incorporating chaos mapping. • Presented an improved seeding strategy in relation to the historical optimal population. • Studied and applied the feature selection problem with extreme learning machine. [ABSTRACT FROM AUTHOR]

This paper introduces a new hybrid bio-inspired solver which combines elements from the recently proposed Coral Reefs Optimization (CRO) algorithm with operators from the Harmony Search (HS) approach, which gives rise to the coined CRO-HS optimization technique. Specifically, this novel bio-inspired optimizer is utilized in the context of short-term wind speed prediction as a means to obtain the best set of meteorological variables to be input to a neural Extreme Learning Machine (ELM) network. The paper elaborates on the main characteristics of the proposed scheme and discusses its performance when predicting the wind speed based on the measures of two meteorological towers located in USA and Spain. The good results obtained in these experiments when compared to naïve versions of the CRO and HS algorithms are promising and pave the way towards the utilization of the derived hybrid solver in other optimization problems arising from diverse disciplines. [ABSTRACT FROM AUTHOR]

• The integer-valued vector representing a candidate solution encodes all dataset features. • The feature subset size is determined by using an additional element in the vector. • Feature relevance is represented through its relative position in the vector. • A permutation-matrix-based mutation operator can be used to create new solutions. • A simple repair-based recombination operator is used to ensure feasible solution creation. This paper describes a permutational-based Differential Evolution algorithm implemented in a wrapper scheme to find a feature subset to be applied in the construction of a near-optimal classifier. In this approach, the relevance of a feature chosen to build a better classifier is represented through its relative position in an integer-valued vector, and by using a permutational-based mutation operator, it is possible to create new feasible candidate solutions only. Furthermore, to provide a controlled diversity rate in the population, a straightforward repair-based recombination operator is utilized to evolve a population of candidate solutions. Unlike the other approaches in the existing literature using integer-valued vectors and requiring a predefined subset size, in this approach, this size is determined by an additional element included in the encoding scheme, allowing to find an adequate feature subset size to each specific dataset. Experimental results show that this approach is an effective way to create more accurate classifiers as they are compared with those obtained by other similar approaches. [ABSTRACT FROM AUTHOR]

• The first attempt to design a GEP-based codematrix generation algorithm. The individual structure of GEP allows the codematrix to be optimized in a much faster and more effective way. • The feature selection scheme is combined with the design of codematrix, to accomplish the feature selection and codematrix optimization simultaneously. • A greedy strategy-based adaptive optimization method to further improve the individual's quality to deal with hard classes by appending proper columns to the codematrix. The multi-class classification task is one of the most common tasks in machine learning. As a typical solution based on a partitioning strategy, Error-Correcting Output Codes (ECOC) can transform a multi-class classification problem into multiple binary classification problems. The key of ECOC is to construct an effective codematrix to represent a set of class decomposition schemes, which transforms a multiclass problem into a group of binary class problems. Consequently, the design of a fast and effective ECOC codematrix generation method is of great research significance and value for solving multi-class classification problems. In ECOC algorithms, the design of codematrix is treated as a combination problem between different code columns, in which the evolutionary algorithm shows a great advantage. Based on this consideration, the Gene Expression Programming (GEP) is applied to search for the codematrix with high performance because its expressive tree structure makes it well represent codematrcies for subsequent optimization operations. This paper proposes an adaptive ECOC algorithm based on Gene Expression Programming (GEP) and similarity measurement matrix, named GEP-ECOC. In our GEP, each individual represents a set of columns to form a random ECOC codematrix, which is optimized in the evolutionary process. Meanwhile, the crossover and mutation operations are modified to include a legality checking process to ensure that the generated codematrix satisfies the ECOC constraints. The GEP-based ECOC codematrix generation algorithm can quickly produce a codematrix with better performance, which ensures the efficiency of the algorithm to a certain extent. In addition, an adaptive algorithm based on a similarity measurement matrix is proposed to add new columns to the current codematrix, aiming to better handle hard classes. Our algorithm is compared with other algorithms on various data sets, and the experimental results confirm that our GEP-ECOC can balance the efficiency and performance of the algorithm and achieve higher performance. [ABSTRACT FROM AUTHOR]

Age estimation has become inordinately significant for human beings for many reasons, such as detecting legal and criminal responsibility and other social events like a marriage license, birth certificate, etc. This paper aims to decide on the most desirable machine learning algorithm (from conventional machine learning algorithms to deep learning) for dental age estimation based on buccal bone level. The database consisted of 150 CBCT images (73 males and 77 females) from an existing base of the Faculty of Dental Medicine with Clinics, University of Sarajevo, aged 20-69. Results were obtained using the Waikato Environment for Knowledge Analysis (Weka), machine learning software in Java. Left and Right Buccal Alveolar Bone Levels are increasing with age, so they showed to be the most important attributes, especially the latter. Random Forest classifier provided the greatest result with the correlation coefficient of 0.803 and the mean absolute error of 6.022. We have also shown that considering sinus-related features can be a significant addition to the databases. Our paper is probably one of the first studies where regression algorithms based on the Support Vector Machines and Random Forest were utilized. [ABSTRACT FROM AUTHOR]

Highlights • Feature selection over high-dimensional colon cancer Microarray Datasets. • Features selected from both Gene Expression and DNA-Methylation Microarray datasets. • Resultant six biomarker genes for colon cancer validated using Enrichment Analysis. • Biomarker genes validated on independent datasets with 99.9% classification accuracy. Abstract Cancer is a dangerous disease that causes death worldwide. Discovering few genes relevant to one cancer disease can result in effective treatments. The challenge associated with the Microarray datasets is its high dimensionality; the huge number of features compared to the modest number of samples in these datasets. Recent research efforts attempted to reduce this high-dimensionality using different feature selection techniques. This paper presents an ensemble feature selection technique based on t -test and genetic algorithm. After preprocessing the data using t -test, a Nested Genetic Algorithm, namely Nested-GA, is used to get the optimal subset of features by combining data from two different datasets. Nested-GA consists of two Nested Genetic Algorithms (outer and inner) that run on two different kinds of datasets. The Outer Genetic Algorithm (OGA-SVM) works on Microarray gene expression datasets, whereas the Inner Genetic Algorithm (IGA-NNW) runs on DNA Methylation datasets. Nested-GA is performed on a colon cancer dataset with 5-fold cross validation. After applying Nested-GA, the Incremental Feature Selection (IFS) strategy is used to get the smallest optimal genes subset. The genes subset has been validated on an independent dataset resulting in 99.9% classification accuracy. Consequently, the biological significance of the resulting optimal genes is validated using Enrichment Analysis. Moreover, the results of Nested-GA have been compared to the results of other feature selection algorithms that have been run on either Gene Expression or DNA Methylation datasets. From the experimental results, Nested-GA showed the highest classification performance with a small optimal feature subset compared to the other algorithms. Furthermore, by running Nested-GA on lung cancer datasets that contain two different cancer subtypes, it resulted in significantly better classification accuracy (98.4%) compared to the accuracy of a previous research (84.6%) that utilized lung cancer DNA-Methylation data only. [ABSTRACT FROM AUTHOR]

Highlights • Two binary variants of the butterfly optimization algorithm (BOA) are proposed. • Two transfer functions are used to map the continuous search space to discrete one. • Twenty one UCI datasets are utilized in the experiments. • A superior performance of the proposed binary variants is proved in the experiments. Abstract In this paper, binary variants of the Butterfly Optimization Algorithm (BOA) are proposed and used to select the optimal feature subset for classification purposes in a wrapper-mode. BOA is a recently proposed algorithm that has not been systematically applied to feature selection problems yet. BOA can efficiently explore the feature space for optimal or near-optimal feature subset minimizing a given fitness function. The two proposed binary variants of BOA are applied to select the optimal feature combination that maximizes classification accuracy while minimizing the number of selected features. In these variants, the native BOA is utilized while its continuous steps are bounded in a threshold using a suitable threshold function after squashing them. The proposed binary algorithms are compared with five state-of-the-art approaches and four latest high performing optimization algorithms. A number of assessment indicators are utilized to properly assess and compare the performance of these algorithms over 21 datasets from the UCI repository. The experimental results confirm the efficiency of the proposed approaches in improving the classification accuracy compared to other wrapper-based algorithms, which proves the ability of BOA algorithm in searching the feature space and selecting the most informative attributes for classification tasks. [ABSTRACT FROM AUTHOR]

Background: Classification and localization of focal epileptic seizures provide a proper diagnostic procedure for epilepsy patients. Visual identification of seizure activity from long-term electroencephalography (EEG) is tedious, time-consuming and leads to human error. Therefore, there is a need for an automated classification system. Methods: In this paper, we introduce a tool called CADFES: computerized automated detection of focal epileptic seizures. For the study, total 41.66 hours of EEG data from the Bern-Barcelona database was used. Set of 28 features were extracted from time, frequency, and statistical domain and significant features were selected using neighborhood component analysis (NCA). In NCA, optimization of regularization parameter ensured better classification accuracy (less classification loss) with seven features. The performance of the algorithm was assessed using support vector machine (SVM), K-nearest neighbor (K-NN), random forest and adaptive boosting (AdaBoost) classifiers. Results: Experimental results revealed sensitivity, specificity, accuracy, positive predictive rate, negative predictive rate, and area under the curve of 97.6%, 94.4%, 96.1%, 92.9%, 98.8% and 0.96 respectively using the SVM classifier. Finally, MATLAB based software tool referred to as CADFES was introduced for automated classification of focal and non-focal seizures. Comparison results ensure that proposed study is superior to existing methods. Hence, it is expected to perform better at the hospitals for automated classification of focal epileptic seizures in real-time. [ABSTRACT FROM AUTHOR]

Abstract Feature selection is used in many applications in machine learning and bioinformatics. As a popular approach, feature selection can be implemented in the filter-manner based on the sparse solution of the l 1 regularization. Most study of the l 1 regularization concentrates on investigating the iteration solution of the problem or focuses on adapting sparsity to different applications. It is necessary to explore more deeply about how the sparsity learned with the l 1 regularization contributes to feature selection. In this paper, we make an effort to analyze the role of the l 1 regularization in feature selection from the perspective of information theory. We discover that the l 1 regularization contributes to minimizing the redundancy in feature selection. To avoid the complex computation of the l 1 optimization, we propose a novel feature selection algorithm, i.e. the Laplacian regularization based on mutual information (LRMI), which realizes the minimization of the redundancy in a new way, and incorporates the l 2 norm to achieve automatic grouping. Extensive experimental results demonstrate the superiority of LRMI over several traditional l 1 regularization based feature selection algorithms with less time consumption. [ABSTRACT FROM AUTHOR]

In recent years, cyber-attacks against businesses and organizations have increased. Such attacks usually result in significant damage to the organization, such as the loss and/or leakage of sensitive and confidential information. Because email communication is an integral part of daily business operations, attackers frequently leverage email as an attack vector in order to initially penetrate the targeted organization. Email message allows the attacker to deliver dangerous content to the victim, such as malicious attachments or links to malicious websites. Existing email analysis solutions analyze only specific parts of the email using rule-based methods, while other important parts remain unanalyzed. Existing anti-virus engines primarily use signature-based detection methods, and therefore are insufficient for detecting new unknown malicious emails. Machine learning methods have been shown to be effective at detecting maliciousness in various domains and particularly in email. Previous works which used machine learning methods suggested sets of features which offer a limited perspective over the whole email message. In this paper, we propose a novel set of general descriptive features extracted from all email components (header, body, and attachments) for enhanced detection of malicious emails using machine learning methods. The proposed features are extracted just from the email itself; therefore, our features are independent, since the extraction process does not require an Internet connection or the use of external services or other tools, thereby meeting the needs of real-time detection systems. We conducted an extensive evaluation of our new novel features against sets of features suggested by previous academic work using a collection of 33,142 emails which contains 38.73% malicious and 61.27% benign emails. The results show that malicious emails can be detected effectively when using our novel features with machine learning algorithms. Moreover, our novel features enhance the detection of malicious emails when used in conjunction with features suggested by related work. The Random Forest classifier achieved the highest detection rates, with an AUC of 0.929, true positive rate (TPR) of 0.947, and false positive rate (FPR) of 0.03. We also present the IDR (integrated detection rate), a new measure which helps calibrate the threshold of a machine learning classifier in order to achieve the optimal TP and FP rates, which are the most important measures for a real-time and practical cyber-security application. [ABSTRACT FROM AUTHOR]

Abstract Multi-view unsupervised feature selection has been proven to be an effective approach to reduce the dimensionality of multi-view data. One of its key issues is how to exploit the underlying common structures across different views. In this paper, we propose a consensus learning guided multi-view unsupervised feature selection method, which embeds multi-view feature selection into a non-negative matrix factorization based clustering with sparse constrain. The proposed method learns latent feature matrices from all the views, and optimizes a consensus matrix such that the difference between the cluster indicator matrix of each view and the consensus matrix is minimized. The parameters for balancing the weights of different views are automatically adjusted, and a sparse constraint is imposed on the latent feature matrices to perform feature selection. After that, we design an effective iterative algorithm to solve the resultant optimization problem. Extensive experiments have been conducted on six publicly multi-view datasets, and the results demonstrate that the proposed algorithm outperforms several other state-of-the-art single view and multi-view unsupervised feature selection methods in terms of clustering tasks, validating the effectiveness of the proposed multi-view unsupervised feature selection method. The source code of our algorithm will be available on our on-line page: http://tangchang.net/. [ABSTRACT FROM AUTHOR]

Objective: This paper constructs a mortality prediction system based on a real-world dataset. This mortality prediction system aims to predict mortality in heart failure (HF) patients. Effective mortality prediction can improve resources allocation and clinical outcomes, avoiding inappropriate overtreatment of low-mortality patients and discharging of high-mortality patients. This system covers three mortality prediction targets: prediction of in-hospital mortality, prediction of 30-day mortality and prediction of 1-year mortality.Materials and Methods: HF data are collected from the Shanghai Shuguang hospital. 10,203 in-patients records are extracted from encounters occurring between March 2009 and April 2016. The records involve 4682 patients, including 539 death cases. A feature selection method called Orthogonal Relief (OR) algorithm is first used to reduce the dimensionality. Then, a classification algorithm named Dynamic Radius Means (DRM) is proposed to predict the mortality in HF patients.Results and Discussions: The comparative experimental results demonstrate that mortality prediction system achieves high performance in all targets by DRM. It is noteworthy that the performance of in-hospital mortality prediction achieves 87.3% in AUC (35.07% improvement). Moreover, the AUC of 30-day and 1-year mortality prediction reach to 88.45% and 84.84%, respectively. Especially, the system could keep itself effective and not deteriorate when the dimension of samples is sharply reduced.Conclusions: The proposed system with its own method DRM can predict mortality in HF patients and achieve high performance in all three mortality targets. Furthermore, effective feature selection strategy can boost the system. This system shows its importance in real-world applications, assisting clinicians in HF treatment by providing crucial decision information. [ABSTRACT FROM AUTHOR]

Eye fixation models, which try to quantitatively predict human eye attended areas in visual fields, have received increasing interest in recent years. In this paper, a novel framework is proposed for the detection of eye fixations. First, a multi-channel detection module, which extracts information of color contrast, salient object proposals and center bias from input image, is conducted to introduce various useful information into the subsequent fixations detection. In salient object detection channel, we employ the multi-instance learning (MIL) algorithm to determine which object proposal can attract attention, which avoids the fuzzyness of positive sample selection. Second, an adaptive weighted fusion method achieved by deep learning framework is proposed to fuse the multi-level information (i.e., contrast, objective, center bias) together for the detection of fixations, so that the integration of information between each level becomes more scientific. Finally, the detection result is optimized by embedding semantic information. Experimental results show that the algorithm has achieved competitive results in MIT1003, MIT300 and Toronto120 dataset. [ABSTRACT FROM AUTHOR]

Nowadays, there has been recent interest in applying machine learning techniques to the neurodegenerative disorders. Dementia is one such emerging global health issue and its early detection can be very helpful. A comparative analysis of four machine learning algorithms i.e., J48, Naïve Bayes, Random Forest and Multilayer Perceptron is presented in this paper. For the attribute reduction, the CFSSubsetEval is used. It has been shown that J48 is performing best among all the algorithms for the detection of Dementia. [ABSTRACT FROM AUTHOR]

Feature selection has playing an important role in many pattern recognition and machine learning applications, where meaningful features are desired to be extracted from high dimensional raw data and noisy ones are expected to be eliminated. l 2,1 -norm regularization based Robust Feature Selection (RFS) has extracted a lot of attention due to its efficiency and high performance of joint sparsity. In this paper, we propose a more general framework for robust and discriminative multi-class feature selection. Four types of weighting, which are based on correlation information between features and labels, are adopted to strengthen the discriminative performance of l 2,1 -norm joint sparsity. F-norm regularization, which is extended from multi-class Elastic Net, is added to improve the stability of the method. An efficient algorithm and its corresponding convergence proof are provided. Experimental results on several two-class and multi-class datasets are performed to verify the effectiveness of the proposed feature selection method. [ABSTRACT FROM AUTHOR]

Assigning weights in features has been an important topic in some classification learning algorithms. In this paper, we propose a new paradigm of assigning weights in classification learning, called value weighting method. While the current weighting methods assign a weight to each feature, we assign a different weight to the values of each feature. The performance of naive Bayes learning with value weighting method is compared with that of some other traditional methods for a number of datasets. The experimental results show that the value weighting method could improve the performance of naive Bayes significantly. [ABSTRACT FROM AUTHOR]

Naïve Bayes, k-nearest neighbors, Adaboost, support vector machines and neural networks are five among others commonly used text classifiers. Evaluation of these classifiers involves a variety of factors to be considered including benchmark used, feature selections, parameter settings of algorithms, and the measurement criteria employed. Researchers have demonstrated that some algorithms outperform others on some corpus, however, inconsistency of human labeling and high dimensionality of feature spaces are two issues to be addressed in text categorization. This paper focuses on evaluating the five commonly used text classifiers by using an automatically generated text document collection which is labeled by a group of experts to alleviate subjectivity of human category assignments, and at the same time to examine the influence of the number of features on the performance of the algorithms. [ABSTRACT FROM AUTHOR]

In Data Mining, during the preprocessing step, there is a considerable diversity of candidate algorithms to select important features, according to some criteria. This broad availability of algorithms that perform the Feature Selection task gives rise to the difficulty of choosing, a priori , between the algorithms at hand, the most promising one for a particular problem. In this paper, we present the proposal and evaluation of a new architecture for the recommendation of Feature Selection algorithms based on the use of Metalearning. Our framework is very flexible since the user can adapt it to its proper needs. This flexibility is one of the main advantages of our proposal over other approaches in the literature, which involve steps that cannot be adapted to the user’s local requirements. Furthermore, it combines several concepts of intelligent systems, including Machine Learning and Data Mining, with topics derived from expert systems, as user and data-driven knowledge, with meta-knowledge. This set of solutions coupled with leading-edge technologies allows our architecture to be integrated into any information system, which impact on the automation of services and in reducing human effort during the process. Regarding the Metalearning process, our framework considers several types of properties inherent to the data sets, as well as, Feature Selection algorithms based on many information, distance, dependence and consistency measures. The quality of the methods for Feature Selection was estimated according to a multicriteria performance measure, which guided the ranking process of these algorithms for the construction of data metabases. Proposed by the authors of this work, this multicriteria performance measure combines any three measurements on a single one, creating an interesting and powerful tool to evaluate not only FS algorithms but also to assess any context where it is necessary a combination to maximize a measure or minimize it. The recommendation models, represented by decision trees and induced from the training metabases, allowed us to see in what circumstances a Feature Selection algorithm outperforms the other and what aspects of the data present greater influence in determining the performance of these algorithms. Nevertheless, if the user wishes, any other learning algorithm may be used to induce the recommendation model. This versatility is another strong point of this proposal. Results show that with the characterization of data, through statistical, information and complexity measures, it is possible to reach an accuracy higher than 90%. Besides yielding recommendation models that are interpretable and robust to overfitting, the developed architecture is less computationally expensive than approaches recently proposed in the literature. [ABSTRACT FROM AUTHOR]

Dragonfly Algorithm (DA) is a recent swarm-based optimization method that imitates the hunting and migration mechanisms of idealized dragonflies. Recently, a binary DA (BDA) has been proposed. During the algorithm iterative process, the BDA updates its five main coefficients using random values. This updating mechanism can be improved to utilize the survival-of-the-fittest principle by adopting different functions such as linear, quadratic, and sinusoidal. In this paper, a novel BDA is proposed. The algorithm uses different strategies to update the values of its five main coefficients to tackle Feature Selection (FS) problems. Three versions of BDA have been proposed and compared against the original DA. The proposed algorithms are Linear-BDA, Quadratic-BDA, and Sinusoidal-BDA. The algorithms are evaluated using 18 well-known datasets. Thereafter, they are compared in terms of classification accuracy, the number of selected features, and fitness value. The results show that Sinusoidal-BDA outperforms other proposed methods in almost all datasets. Furthermore, Sinusoidal-BDA exceeds three swarm-based methods in all the datasets in terms of classification accuracy and it excels in most datasets when compared in terms of the fitness function value. In a nutshell, the proposed Sinusoidal-BDA outperforms the comparable feature selection algorithms and the proposed updating mechanism has a high impact on the algorithm performance when tackling FS problems. • This paper proposes an improved Binary Dragonfly Algorithm (BDA) for Feature Selection. • Three versions of BDA are proposed: LBDA, QBDA, and SBDA. • Eighteen datasets taken from UCI machine learning repository are used for evaluation process. • SBDA excels LBDA and QBDA in terms of accuracy, number of features, and fitness function. • SBDA excels the nine comparative methods in 12 out of 18 dataset in terms of accuracy. [ABSTRACT FROM AUTHOR]

Feature selection is an important preprocessing technique in the fields of data mining and machine learning. With the promotion of privacy protection awareness, recently it becomes a very practical and challenging issue to select high-quality feature subsets while ensuring the privacy of all participants. However, there is a lack of research results on this issue, i.e., feature selection under privacy protection. Aiming at the issue, this paper proposes a federated feature selection framework for the first time. In the framework, inspiring by the idea of federated learning, a credible third participant is introduced to process and integrate optimal feature subsets from multiple participants. On the basis of the framework, a federated evolutionary feature selection algorithm based on particle swarm optimization is proposed to effectively solve feature selection problems with multiple participants under privacy protection. Two new operators satisfying the requirement of privacy protection, i.e., the feature assembling strategy with multi-participant cooperation and the swarm initialization strategy guided by assembling solution, are designed to improve the ability of the proposed algorithm. Compared with several typical assembling feature selection algorithms on 15 data sets, experimental results show that the proposed algorithm can significantly improve the classification accuracy of the feature subset selected by each participant, while protecting the privacy of data. [ABSTRACT FROM AUTHOR]

• A hybrid Machine Learning (ML) approach for an efficient wrapper feature selection is proposed for enhancing the performance and reducing the computation time of IDS while increasing the accuracy to detect the intruder in IoT networks. • The hybrid ML method consists of two techniques: Differential Evolution (DE) and Extreme Learning Machine (ELM). The DE is responsible for choosing the beneficial features while the ELM is employed to assess the selected features. • Extensive experiments are implemented to evaluate the proposed approach using different performance metrics. In addition, to verify the efficiency of proposed model, the comparison is achieved with some related existing methods such as CFS + SVM [20] , CFS + ANN [19] , GA + Bagging [18] , CFS + RF [30] , and Chi-squared + RF [30]. The intrusion detection system (IDS) has gained a rapid increase of interest due to its widely recognized potential in various security fields, however, it suffers from several challenges. Different network datasets have several redundant and irrelevant features that affect the decision of the IDS classifier. Therefore, it is essential to decrease these features to improve the system performance. In this paper, an efficient wrapper feature selection method is proposed for improving the performance and decreasing the processing time of the IDS. The proposed approach employs a differential evaluation algorithm to select the useful features whilst the extreme learning machine classifier is applied after feature selection to evaluate the selected features. Many experiments are performed using the full NSL-KDD dataset to evaluate the performance of the proposed method. The results prove that the proposed approach can efficiently reduce the features, increase the accuracy, reduce the false alarm rates, and improve the processing time of the IDS in comparison to other recent related works. [ABSTRACT FROM AUTHOR]

With the advent of Big Data, data is being collected at an unprecedented fast pace, and it needs to be processed in a short time. To deal with data streams that flow continuously, classical batch learning algorithms cannot be applied and it is necessary to employ online approaches. Online learning consists of continuously revising and refining a model by incorporating new data as they arrive, and it allows important problems such as concept drift or management of extremely high-dimensional datasets to be solved. In this paper, we present a unified pipeline for online learning which covers online discretization, feature selection and classification. Three classical methods—the k -means discretizer, the χ 2 filter and a one-layer artificial neural network—have been reimplemented to be able to tackle online data, showing promising results on both synthetic and real datasets. [ABSTRACT FROM AUTHOR]

In recent years, there has been a great interest in prostate segmentation, which is an important and challenging task for CT image guided radiotherapy. In this paper, a learning-based segmentation method via joint transductive feature selection and transductive regression is presented, which incorporates the physician׳s simple manual specification (only taking a few seconds), to aid accurate segmentation, especially for the case with large irregular prostate motion. More specifically, for the current treatment image, experienced physician is first allowed to manually assign the labels for a small subset of prostate and non-prostate voxels, especially in the first and last slices of the prostate regions. Then, the proposed method follows the two step: in prostate-likelihood estimation step, two novel algorithms, tLasso and wLapRLS, will be sequentially employed for transductive feature selection and transductive regression, respectively, aiming to generate the prostate-likelihood map. In multi-atlases based label fusion step, the final segmentation result will be obtained according to the corresponding prostate-likelihood map and the previous images of the same patient. The proposed method has been substantially evaluated on a real prostate CT dataset including 24 patients with 330 CT images, and compared with several state-of-the-art methods. Experimental results show that the proposed method outperforms the state-of-the-arts in terms of higher Dice ratio, higher true positive fraction, and lower centroid distances. Also, the results demonstrate that simple manual specification can help improve the segmentation performance, which is clinically feasible in real practice. [ABSTRACT FROM AUTHOR]

Instance selection aims at filtering out noisy data (or outliers) from a given training set, which not only reduces the need for storage space, but can also ensure that the classifier trained by the reduced set provides similar or better performance than the baseline classifier trained by the original set. However, since there are numerous instance selection algorithms, there is no concrete winner that is the best for various problem domain datasets. In other words, the instance selection performance is algorithm and dataset dependent. One main reason for this is because it is very hard to define what the outliers are over different datasets. It should be noted that, using a specific instance selection algorithm, over-selection may occur by filtering out too many ‘good’ data samples, which leads to the classifier providing worse performance than the baseline. In this paper, we introduce a dual classification (DuC) approach, which aims to deal with the potential drawback of over-selection. Specifically, performing instance selection over a given training set, two classifiers are trained using both a ‘good’ and ‘noisy’ sets respectively identified by the instance selection algorithm. Then, a test sample is used to compare the similarities between the data in the good and noisy sets. This comparison guides the input of the test sample to one of the two classifiers. The experiments are conducted using 50 small scale and 4 large scale datasets and the results demonstrate the superior performance of the proposed DuC approach over the baseline instance selection approach. [ABSTRACT FROM AUTHOR]

• 3 layer neural networks outperform other ML algorithms in automated assessments. • In subjective assignments, deep neural networks can improve score prediction. • Word embeddings and linguistic feature results in more precise text mining models. • Extraction and selection of appropriate features can improve results. Machine learning (ML) has recently gained popularity in a variety of domains for automating tasks. This has also been used in the evaluation of student responses/ assignments. This paper presents a qualitatively enhanced methodology for automated score prediction of subjective assignments. To consider all the major aspects of a human grader, 21 linguistic features related to syntactical, grammatical, sentimental, and readability are analyzed from the assignments. This study makes use of the ASAP competition dataset from Kaggle. The evaluation metric used is Quadratic Weighted Kappa (QWK), which measures the agreement between the human graded score and the predicted score. The effect of appropriate feature selection has been observed using Mutual Information Regression. Four ML algorithms are investigated on identified linguistic features. 3 Layer Neural Network with feature selection performed well among all chosen ML algorithms with average QWK of 0.678. To include the benefits of deep learning, a new hybrid model (LF-BiLSTM-att-FS) is proposed that combines a higher level deep neural network (DNN) with the selected features. Pre-trained Glove embedding is used to include contextual information of the text. The proposed model results demonstrated an improvement in overall accuracy, with an average QWK value of 0.768. [ABSTRACT FROM AUTHOR]

Reduced computational time and cost, reduced skilled professional resources, and diagnostic accuracy have made medical diagnosis using computer aided systems (CAD) increasingly popular and can be comfortably employed in the diagnosis of many acute and chronic diseases in ophthalmology, cardiology, cancer detection, etc. There seems to be a growing necessity for the computational algorithms to be robust enough, to identify the abnormality in each of the cases, aiding in early diagnosis. In this paper, a glaucoma diagnostic approach based on the wrapper method employing bio-inspired algorithms, and a Kernel-Extreme Learning Machine (KELM) classifier is proposed. The bio-inspired algorithms are deployed to select feature sub-sets, generating three feature sub-sets from the pre-processed fundus images by adopting a correlation-based feature selection (CFS) approach. The selected features are utilized to train the salp-swarm optimization based KELM, which finds the optimal parameters of the KELM classifier network. The proposed methodology is evaluated on the public and private retinal fundus datasets containing 7280 images. The experimental outcome revealed that the system is able to attain a maximum overall accuracy of 99.61% with 99.89% sensitivity and 100% specificity. A 5-fold cross validation showed 98.78% accuracy ensuring a bias-free classification. Further, by experimenting on degraded images (Gaussian, salt-pepper noise images) of the original dataset, the model achieved extreme robustness with 99.3% accuracy. The proposed method is compared with other similar methods, which showed the efficiency of our method. The framework proposed can aid in making clinical decisions for various pathologies like lung infection, diabetic retinopathy, etc. • A predictive framework involving bio-inspired algorithms and Kernel-ELM is presented. • Feature Selection based on Correlation (CFS) is employed to find the optimal feature set. • KELM parameters are optimized by bio-inspired evolutionary algorithm, salp-swarm optimization algorithm (SSA) for improved diagnosis. • The proposed method works well on degraded (gaussian and salt-pepper noise) images showing the robustness of the method.. • Superior performance with statistically significant improvements on varied datasets compared to competitive algorithms. [ABSTRACT FROM AUTHOR]

Background and objective Understanding the causes of disagreement among experts in clinical decision making has been a challenge for decades. In particular, a high amount of variability exists in diagnosis of retinopathy of prematurity (ROP), which is a disease affecting low birth weight infants and a major cause of childhood blindness. A possible cause of variability, that has been mostly neglected in the literature, is related to discrepancies in the sets of important features considered by different experts. In this paper we propose a methodology which makes use of machine learning techniques to understand the underlying causes of inter-expert variability. Methods The experiments are carried out on a dataset consisting of 34 retinal images, each with diagnoses provided by 22 independent experts. Feature selection techniques are applied to discover the most important features considered by a given expert. Those features selected by each expert are then compared to the features selected by other experts by applying similarity measures. Finally, an automated diagnosis system is built in order to check if this approach can be helpful in solving the problem of understanding high inter-rater variability. Results The experimental results reveal that some features are mostly selected by the feature selection methods regardless the considered expert. Moreover, for pairs of experts with high percentage agreement among them, the feature selection algorithms also select similar features. By using the relevant selected features, the classification performance of the automatic system was improved or maintained. Conclusions The proposed methodology provides a handy framework to identify important features for experts and check whether the selected features reflect the pairwise agreements/disagreements. These findings may lead to improved diagnostic accuracy and standardization among clinicians, and pave the way for the application of this methodology to other problems which present inter-expert variability. [ABSTRACT FROM AUTHOR]

With the advent of high-dimensional stored big data and streaming data, suddenly machine learning on a very large scale has become a critical need. Such machine learning should be extremely fast, should scale up easily with volume and dimension, should be able to learn from streaming data, should automatically perform dimension reduction for high-dimensional data, and should be deployable on hardware. Neural networks are well positioned to address these challenges of large scale machine learning. In this paper, we present a method that can effectively handle large scale, high-dimensional data. It is an online method that can be used for both streaming and large volumes of stored big data. It primarily uses Kohonen nets, although only a few selected neurons (nodes) from multiple Kohonen nets are actually retained in the end; we discard all Kohonen nets after training. We use Kohonen nets both for dimensionality reduction through feature selection and for building an ensemble of classifiers using single Kohonen neurons. The method is meant to exploit massive parallelism and should be easily deployable on hardware that implements Kohonen nets. Some initial computational results are presented. [ABSTRACT FROM AUTHOR]

In the past two decades, the dimensionality of datasets involved in machine learning and data mining applications has increased explosively. Therefore, feature selection has become a necessary step to make the analysis more manageable and to extract useful knowledge about a given domain. A large variety of feature selection techniques are available in literature, and their comparative analysis is a very difficult task. So far, few studies have investigated, from a theoretical and/or experimental point of view, the degree of similarity/dissimilarity among the available techniques, namely the extent to which they tend to produce similar results within specific application contexts. This kind of similarity analysis is of crucial importance when two or more methods are combined in an ensemble fashion: indeed the ensemble paradigm is beneficial only if the involved methods are capable of giving different and complementary representations of the considered domain. This paper gives a contribution in this direction by proposing an empirical approach to evaluate the degree of consistency among the outputs of different selection algorithms in the context of high dimensional classification tasks. Leveraging on a proper similarity index, we systematically compared the feature subsets selected by eight popular selection methods, representatives of different selection approaches, and derived a similarity trend for feature subsets of increasing size. Through an extensive experimentation involving sixteen datasets from three challenging domains (Internet advertisements, text categorization and micro-array data classification), we obtained useful insight into the pattern of agreement of the considered methods. In particular, our results revealed how multivariate selection approaches systematically produce feature subsets that overlap to a small extent with those selected by the other methods. [ABSTRACT FROM AUTHOR]

Bayesian Ying–Yang (BYY) harmony learning system is a powerful tool for statistical learning. Via the BYY harmony leaning of finite mixtures, model selection, i.e., the selection of an appropriate number of components for the mixture, can be made automatically during parameter learning on a given dataset. In this paper, an adaptive gradient BYY harmony learning algorithm is proposed for log-normal mixtures to implement parameter learning with automated model selection. It is demonstrated by the experiments on both synthetic and real-world datasets that the proposed BYY harmony learning algorithm not only has the ability of automated model selection, but also leads to a rather good estimation of the parameters in the original log-normal mixture. [ABSTRACT FROM AUTHOR]

Cancer is curable if it can be detected early. One way to detect cancer is by analyzing the change in expression of genes in the suspected tissue. Serial analysis of gene expression (SAGE) is a sequencing technique used for measuring the expression levels of genes. Cancer detection problem can be posed as binary classification problem like whether a tissue is cancerous or normal. SAGE libraries contain expression levels of thousands of genes which are the features. It is impossible to consider all these features for classification and also the general feature selection algorithms are not efficient with this data. In this paper, closed frequent itemset mining is proposed as a feature selection technique for identifying a small set of features which can distinguish the two classes efficiently. The performance of the proposed technique is evaluated on SAGE data related to breast tissue and a group of 26 genes are selected as best features. Two well known classifiers, extreme learning machine (ELM) and support vector machine (SVM), are used to evaluate the effectiveness of the selected features in classification and found that the proposed method works well with these classifiers. [ABSTRACT FROM AUTHOR]