Search

This paper presents a new contribution to the recommendation system following E-commerce reviews. Here, it implements a dictionary-based sentiment analysis based on four main steps (i) Pre-processing (ii) Semantic word extraction (iii) Feature Extraction (iv) Classification. Initially, the tweets or reviews extracted from the database are subjected to pre-processing, which involves three processes: stop word removal, stemming, and blank space removal. In the semantic word extraction process, the semantic words from the dictionary will be extracted by matching with the extracted keywords. The next process is the feature extraction, which extracts joint holoentropy and cross holoentropy of all keywords, and further proceeds with conditional holoentropy-based feature selection process. Finally, the selected features are subjected to a classification process using Deep Belief Network (DBN), which is a well-performing deep learning algorithm. Moreover, the main contribution of the paper relies on the improvement of DBN architecture, where the number of hidden neurons and activation function is optimally tuned by a hybrid optimization algorithm termed as Grey linked Chick Update-based Chicken Swarm Optimization (GCU-CSO). At last, the comparative analysis of proposed and conventional algorithms in terms of various performance metrics proves the efficacy of the proposed E-commerce recommendation system. [ABSTRACT FROM AUTHOR]

• Newbie study using standard ML techniques with HAR Application and Discussions. • Activities found in Literature with the corresponding reference. • Co-occurrences between activities and sensors with the corresponding reference. • Summary and comparison among the different datasets found in Literature. • Summary of the experimentation settings with performance scores found in Literature. Human Activity Recognition (HAR) is an essential area of research related to the ability of smartphones to retrieve information through embedded sensors and recognize the activity that humans are performing. Researchers have recognized people's activities by processing the data received from the sensors with Machine Learning Models. This work is intended to be a hands-on survey with practical's tables capable of guiding the reader through the sensors used in modern smartphones and highly cited developed machine learning models that perform human activity recognition. Several papers in the literature have been studied, paying attention to the preprocessing, feature extraction, feature selection, and classification techniques of the HAR system. In addition, several summary tables illustrating HAR approaches have been provided: most popular human activities in the literature with paper references, the most popular datasets available for download (Analyzing their characteristics, such as the number of subjects involved, the activities recorded, and the sensors with online-availability), co-occurrences between activities and sensors, and a summary table showing the performance obtained by researchers. =The paper's goal is to recommend, through the discussion phase and thanks to the tables, the current state of the art on this topic. [ABSTRACT FROM AUTHOR]

This paper proposes an improved bubble entropy algorithm called Improved Hierarchical Refined Composite Multiscale Multichannel Bubble Entropy (IHRCMMCBE) to characterize the fault characteristics of rotating machinery. By introducing the refined composite multiscale analysis algorithm, the improved hierarchical decomposition algorithm, and the multi-channel data analysis method, the bubble entropy algorithm can more fully characterize the fault characteristics. Then, this method is combined with the machine learning algorithm to realize automatic fault diagnosis of rotating machinery. Multi-Cluster Feature Selection (MCFS) is used for feature selection to screen out useless features and improve feature extraction efficiency and feature quality. The MCFS method has excellent performance and does not need parameters, so it is suitable for adaptive fault diagnosis. Meanwhile, the EigenClass classifier using the concept of generalized eigenvalue is used to realize fault recognition. Besides, the recently proposed Gorilla Troops Optimizer (GTO) algorithm is used to improve EigenClass, and a GTO-optimized EigenClass (GTO-EigenClass) classifier is proposed to adaptively select parameters and achieve adaptive fault identification. The proposed fault diagnosis method of rotating machinery based on the improved bubble entropy is independent of data and application scenarios and is hardly affected by subjective factors. The ablation experiment proves the effectiveness of each improvement of bubble entropy. Also, the performance verification experiments and the comparative experiments prove that the method can effectively characterize different types of rotating machinery, different fault types, fault degrees, and composite faults, and it can achieve higher calculation efficiency and classification accuracy. When the sample length is 1024, the average accuracy of bearing and gearbox fault identification can reach 97.85 % and 97.57 % respectively, and when the sample length is 2048, the identification accuracy can reach 100 %. [ABSTRACT FROM AUTHOR]

• Pinch force strength is accurately predicted using sEMG. • Different pinch movements are accurately classified using sEMG. • Similarity of EMG features is evaluated. • An EMG feature-to-image method is proposed. The movement of intelligent EMG-driven prosthesis mainly relies on the synergy of different fingers to achieve function of grasping objects. The paper proposes a novel scheme for force prediction and movement classification about pinch between different fingers based on surface electromyography (sEMG) using machine learning. The pinch force and sEMG signals are recorded synchronously by a data acquisition device. Eight features are extracted, which are proven to have better performance in the estimation of sEMG-to-force. We present a novel feature selection method that uses the one-dimensional time series similarity assessment based on Manhattan distance to eliminate the repetitive information between features. Three optimal features carrying less repetitive information are retained. Seven machine learning algorithms are used to predict force strength. The results show that the Long Short Term Memory (LSTM) has the best performance of force prediction, achieving a R 2 of 0.9517 and RMSE of 3.2723. The paper proposes a novel method of converting EMG feature sequences to the normalized gray image in order to classify the finger movement. Five classifiers based on image feature extraction and the Convolutional Neural Network (CNN) are developed respectively. The experimental results indicate that the CNN performs best, achieving an accuracy of 97.66%. In this way, it not only realizes the accurate force prediction, but also realizes the movement classification between different fingers. The proposed methodology has the potential to realize simultaneous force and movement control of prosthetic hand. [ABSTRACT FROM AUTHOR]

• Brain network features were extracted by a Granger causal analysis. • Brain functional connectivity contributes to improving MI task classification. • An effective swarm optimization algorithm are used for feature selection. In the study of motor imagery (MI) brain-computer interfaces (BCIs), how to improve task classification accuracy has been always one of major challenges in the applications of MI-BCIs. As a type of crucial temporal and spatial feature, nonlinear Granger Causality (NGC) analysis was applied to feature extraction of MI-electroencephalogram (EEG) signals because the constructed brain network features can reflect the causal relationship between different channels in various brain regions. However, the MI-BCI task recognition often suffer from the information redundancy of NGC features, and these redundant features will increase the complexity of the machine learning models and accordingly reduce the prediction accuracy of the classification algorithms. To address this problem, this paper proposes a step-by-step tent chaos simulated annealing salp swarm feature selection (STCSA_SaSFS) algorithm to select an optimal set of features in a wrapper feature selection model. Then, the effectiveness of this feature selection method is verified using a support vector machine (SVM) classifier. Through the study of task related MI-BCI EEG data from ten subjects, the experiments showed that the highest classification accuracy of NGC feature extraction plus STCSA_SaSFS reached 97.19%, and the average classification accuracy was 89.57%. This average classification accuracy was 20.07% higher than that of NGC feature extraction without any feature selection, and it is also 2.96% higher than that of NGC feature extraction plus a traditional SaSFS algorithm. The effectiveness of STCSA_SaSFS was also compared with that of other smart swarm optimization algorithms, such as the sparrow search feature selection algorithm (SpSFS). STCSA_SaSFS outperforms SpSFS with an average classification accuracy of 8.07%. The algorithm was validated using a public dataset validation consisting of 10 subjects, which ultimately showed that the feature selection method proposed in this paper (STSA_SaSAFS) has a large advantage in the classification performance of motor imagery brain-computer interface tasks. [ABSTRACT FROM AUTHOR]

Semiconductor technologies have gained wide-spread attention across industries due to its use in varied digital signal processing units, healthcare-devices, industrial communication, science and technologies, entertainment, smart devices etc. However, high design-complexities, untrusted foundries, third-party processors etc., have raised the concern of intrusion, particularly Hardware Trojans. Hardware Trojan, a specifically designed micro-intruder that can cause data breaches, system malfunctions etc. And hence may result disastrous consequences. To alleviate such problems, though different methods have been suggested for Trojan detection; however, pre-manufacturing methods can't guarantee Trojan-free design due to non-trusted Foundries and Third-party IPs. Similarly, post-manufacturing methods like side channel analysis or gate -level netlist are limited due to highly complex design, large structure and unavailability of universal golden design structures. Considering it as motivation, this paper proposes a robust register transfer logic (RTL) analysis assisted multi-parameter-based machine learning-driven Hardware Trojan detection system. We applied delay, power consumption and resource utilization profiles to perform Trojan detection over AES-256 circuits. Aforesaid features were extracted using VIVADO tool, which were further employed for resampling process using SMOTE, SMOTE-ENN, SMOTE-Borderline algorithms. Here, the use of resampling intended to alleviate data-imbalance or skewness. Moreover, it helped retaining sufficiently large features for machine learning based training. The resampled features were processed for feature selection using Significant Predictor Test, Cross-Correlation Test and Principal Component Analysis (PCA), distinctly. The selected features were trained using a novel heterogenous ensemble learning model encompassing a total of 11 base-classifiers belonging to Naïve Bayes, Neuro-computing, Decision Tree, Random Forest, Extra Tree classifier and Bagging-AdaBoost algorithms. Applying the maximum voting ensemble, the proposed model performed path-level Trojan detection, where the simulated results affirmed accuracy, F-Measure and Area Under Curve (AUC) of almost 100%, which is higher than any known approaches towards Hardware Trojan detection. • Detection of hardware Trojan in register transfer level of the hardware circuit. • Hardware Trojan detection by machine learning classifier. • The class imbalance problem is reduced by the resampling techniques. • Heterogeneous Ensemble Machine learning algorithm is used for Trojan detection along with the maximum voting ensemble method. [ABSTRACT FROM AUTHOR]

Machine learning (ML) and deep learning (DL) models have been extensively explored for the early diagnosis of various cancer diseases, including Leukemia, with many of them achieving significant performance improvements comparable to those of human experts. However, challenges like limited image data, inaccurate annotations, and prediction reliability still hinder their broad implementation to establish a trustworthy computer-aided diagnosis (CAD) system. This paper introduces a novel weighted-average ensemble model for classifying Acute Lymphoblastic Leukemia, along with a reliable Computer-Aided Diagnosis (CAD) system that combines the strengths of both ML and DL approaches. Initially, a variety of filtering methods are extensively analyzed to determine the most suitable image representation, with subsequent data augmentation techniques to expand the training data. Second, a modified VGG-19 model was proposed with fine-tuning that was utilized as a feature extractor to extract meaningful features from the training samples. Third, A small-loss approach and probabilistic local outlier factor (pLOF) have been developed on the extracted features to address the label noise issue. Fourth, we proposed an weighted-average ensemble model based on the top five models as base learners, with weights calculated based on their model uncertainty to ensure reliable predictions. Fifth, we calculated Shapley values based on cooperative game theory and performed feature selection with different feature combinations to determine the optimal number of features using SHAP. Finally, we integrate these strategies to develop an interpretable CAD system. This system not only predicts the disease but also generates Grad-CAM images to visualize potential affected areas, enhancing both clarity and diagnostic insight. All of our code is provided in the following repository: https://github.com/taareek/leukemia-classification [ABSTRACT FROM AUTHOR]

Big data scenarios are characterized by multiple devices, massive traffic, and high data dimensionality. In the process of attack identification, the selection of features from massive data directly affects the attack detection effect and has become a key issue that constrains attack identification. Therefore, this paper proposes an evolutionary computation-based machine learning approach for detecting network attacks in big data traffic. First, the RandomSample-SMOTE (Synthetic Minority Over-sampling Technique) method is designed to perform class imbalance processing on network attack traffic collected from big data traffic; second, the feature importance of the attack traffic in different classification layers is calculated and ranked separately using the LightGBM (Light Gradient Boosting Machine) model, and the optimal feature values are selected through retraining; finally, the obtained feature values are used for model training and the most optimal model is obtained by optimizing the hyperparameters with TuneGridSearchCV (Tune's Grid Search Cross Validation). The results of simulation experiments show that the method in this paper can effectively extract features from big data traffic. It can effectively reduce feature dimensionality, significantly improve detection accuracy and save about 40% of computation time compared with existing methods. • A machine learning method based on evolutionary computation is proposed to detect network attacks in big data traffic. • A data pre-processing method is proposed to balance the data set. • A hierarchical feature selection method is proposed to extract features and reduce dimensions. • A hierarchical attack identification model is proposed with high detection accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Many researchers analyze malware through static analysis and dynamic analysis technology, and combine it with excellent deep learning algorithm, which has achieved good results in malware classification. However, many researches only use the. ASM file generated by decompiler or. Bytes file represented by hexadecimal for feature extraction. This paper fully integrates the features of these two files, and uses word frequency and two deep learning algorithms to extract 184 opcode features and 16 probability features from ASM file and section file of Kaggle dataset respectively. Then, double byte feature coding method is used to fuse the features of the two files. Finally, convolution neural network is used to classify the fused samples. The experimental results show that the accuracy is 98.68% and the logarithm loss is 0.022. [ABSTRACT FROM AUTHOR]

• A Vision Transformer-based architecture for lower limb motion recognition from MMG and kinematic signals was proposed. • The model can recognize 8 motions with an accuracy of 94.62%, much higher than machine learning methods and CNN. • Fusion signals of MMG and kinematic data improve the classification accuracies of the model. • The proposed ViT-LLMR maintains high recognition accuracy when undersampling and only use part of signals. One of the key problems in lower limb-based human–computer interaction (HCI) technology is to use wearable devices to recognize the wearer's lower limb motions. The information commonly used to discriminate human motion mainly includes biological and kinematic signals. Considering that unimodal signals do not provide enough information to recognize lower limb movements, in this paper, we proposed a Vision Transformer (ViT)-based architecture for lower limb motion recognition from multichannel Mechanomyography (MMG) signals and kinematic data. Firstly, we applied the self-attention mechanism to enhance each input channel signal. Then the data was fed into ViT model. Vision Transformer-based Lower Limb Motion Recognition (ViT - LLMR) architecture proposed in this paper can avoid the model training problems such as autonomous feature extraction and feature selection for machine learning, and the model can recognize eight lower limb motions containing six subjects with an accuracy of 94.62%. In addition, we analyzed the generalization ability of the model when undersampling and only collecting fragment signals. In conclusion, the proposed ViT - LLMR architecture could provide a basis for practical applications in different HCI fields. [ABSTRACT FROM AUTHOR]

[Display omitted] • Using ORACM as a fast and precise segmentation method for skin lesion segmentation. • Combining the nonlinear indices, GLCM, and texture features for feature extraction. • Investigating the chaotic nature of tumors by utilizing several complexity measures. • Applying NSGA II as a multi-objective metaheuristic approach for feature selection. • Obtaining high classification accuracy of 99.24% based on the proposed method. Skin cancer affects people of all skin tones, including those with darker complexions. Melanomas are known as malignant tumors of skin cancer, resulting in an adverse prognosis, responsible for most deaths relating to skin cancer. Early diagnosis and treatment of skin cancer from dermoscopic images can significantly reduce mortality and save lives. In this paper, Computer-Aided Diagnosis (CAD) system based on machine learning algorithms is provided to classify various skin cancer types. The proposed method uses the Online Region-based Active Contour Model (ORACM) with a new binary level set equation and regularization operation to extract skin lesions' Region of Interest (ROI). Additionally, various combinations of different textures and nonlinear features are extracted for the ROI to show the multiple aspects of skin lesions. Several metaheuristic optimization algorithms are used to remove redundant or irrelevant features and reduce the feature space dimension. These are applied to the combination of the extracted features in which, the Non-dominated Sorting Genetic Algorithm (NSGA II) as a multi-objective optimization algorithm has the best performance. Furthermore, various machine learning algorithms including K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Fitting neural network (Fit net), Feed-Forward neural network (FF net), and Pattern recognition network (Pat net) are employed for the classification. In this paper, the best-obtained evaluation parameters based on fivefold cross-validation belong to the combination of selected nonlinear indices, texture, and GLCM features with an accuracy: 92.24 %, specificity: 100 %, and sensitivity: 100 %, resulting through NSGA II and applying the pattern net classifier. Besides, the comparison between this paper's experimental results and other similar works demonstrates the proposed method's efficiency. [ABSTRACT FROM AUTHOR]

• A sensitive fault feature library based on high frequency components is constructed. • A feature selection method based on XGBoost can remove redundant features. • A reliable HIF identification scheme based on KELM can identify fault phase selection accurately. In distribution networks, high-impedance faults (HIFs) occur frequently and have a harmful impact on the distribution network. However, fault detection and fault phase selection of HIFs are challenging due to weak fault characteristics. Therefore, this paper proposes an improved HIF identification scheme based on a kernel extreme learning machine (KELM) that can sensitively identify HIFs and select the fault phase by adaptively extracting the weak fault characteristics. First, a fault feature extraction strategy based on discrete wavelet decomposition (DWT) and the Hilbert–Huang transform (HHT) is proposed to obtain multiple features that describe the weak fault characteristics of HIFs. Second, an XGBoost-based fault feature selection scheme is proposed for screening with sensitive characterization of HIFs. Next, a sensitive and accurate HIF identification scheme based on the improved learning algorithm (KELM) is proposed to enable accurate and sensitive HIF detection and phase selection. Finally, numerical simulations based on PSCAD/EMTDC and MATLAB were carried out, which reveals the effectiveness and accuracy of the proposed HIF identification scheme. Compared with the traditional HIF identification scheme, the proposed method exhibits conciseness and correctness. [ABSTRACT FROM AUTHOR]

This paper proposes a data-driven approach to estimating geometric track irregularities from instrumented railway vehicle (IRV) data. Machine learning is used to find the nonlinear mapping between IRV data and track irregularities. A dynamic model of the BRA1 railway vehicle was used to generate an artificial dataset that contains variables that are measured by the real BRA1 IRV and other variables measured by IRVs found in the literature. An extensive data analysis step was done to verify if the current instrumentation of the BRA1 IRV is sufficient for obtaining both lateral and vertical track irregularities. Feature engineering based on wagon movements, signal integration and time domain statistical metrics were applied to extract features and then the best features were selected using a wrapper method. Eight different regression ML models were trained and optimized after the feature selection using Optuna. The results show that, with the current instrumentation of the BRA1 IRV, obtaining lateral track irregularities is unlikely due to low correlation, however, vertical irregularities can be obtained with a root mean squared error (RMSE) of 0.556 mm. With postprocessing, the RMSE was further reduced to 0.410 mm. [ABSTRACT FROM AUTHOR]

Due to the presence of redundant and irrelevant features in large-dimensional biomedical datasets, the prediction accuracy of disease diagnosis can often be decreased. Therefore, it is important to adopt feature extraction methodologies that can deal with problem structures and identify underlying data patterns. In this paper, we propose a novel approach called the Anti Coronavirus Optimized Kernel-based Softplus Extreme Learning Machine (ACO-KSELM) to accurately predict different types of skin cancer by analyzing high-dimensional datasets. To evaluate the proposed ACO-KSELM method, we used four different skin cancer image datasets: ISIC 2016, ACS, HAM10000, and PAD-UFES-20. These dermoscopic image datasets were preprocessed using Gaussian filters to remove noise and artifacts, and relevant features based on color, texture, and shape were extracted using color histogram, Haralick texture, and Hu moment extraction approaches, respectively. Finally, the proposed ACO-KSELM method accurately predicted and classified the extracted features into Basal Cell Carcinoma (BCC), Squamous Cell Carcinoma (SCC), Actinic Keratosis (ACK), Seborrheic Keratosis (SEK), Bowen's disease (BOD), Melanoma (MEL), and Nevus (NEV) categories. The analytical results showed that the proposed method achieved a higher rate of prediction accuracy of about 98.9%, 98.7%, 98.6%, and 97.9% for the ISIC 2016, ACS, HAM10000, and PAD-UFES-20 datasets, respectively. [ABSTRACT FROM AUTHOR]

The complex internal structure and heterogeneous composition material of aerospace equipment make it difficult to apply traditional acoustic emission source localization methods to the loose particle localization research. In previous studies, the authors transformed the loose particle localization problem into the multi-classification problem in machine learning, trained the loose particle localization model, but there were problems of low classification accuracy and low practicality. In this paper, the authors first introduced the ensemble learning idea into the loose particle detection field, designed a complete loose particle localization scheme, analyzed the above problems from multiple perspectives, and proposed new algorithms or strategies to enhance the superiority and practicability of the loose particle localization model. Specifically, in view of the low classification accuracy, the authors carried out the research on pulse preprocessing, feature engineering and model training from the four perspectives of signal, feature, data set and classifier, respectively. The zero-pulse-filling pulse matching algorithm and the channel-weighting-based feature selection method was newly proposed, Mel-Frequency Cepstral Coefficients features was newly extracted, feature optimization scheme was designed, and XGBoost ensemble classifier was trained. Thus, high-quality loose particle signals, high-quality localization data sets and high-performance loose particle localization models were obtained, respectively. Test results show that, the classification accuracy achieved by the new loose particle localization model was 96.80%, which was a significant improvement compared to the 83.53% achieved in previous studies. In view of the low practicality, the authors built the loose particle localization experimental system, gave the specific implementation steps of the loose particle localization method, as well as the general procedures for applying the loose particle localization model for physical testing. Meanwhile, taking into account the requirements of aerospace engineering applications, the authors added the majority voting strategy to convert classification results into localization results, thus newly proposed the definition of equipment-level loose particle localization accuracy. Several physical testing results show that, the loose particle localization model achieved a localization accuracy of 90.91%, which effectively verified the feasibility and stability of the proposed method. This study is an important supplement to the loose particle detection research, and of great significance in improving the reliability of aerospace systems. [ABSTRACT FROM AUTHOR]

The internet plays an important role in all areas of society from the economy to the government. Security means permitting things you do want, while preventing things you don’t want from performing. Computer security means storing the information on a computer in a secure manner. Nowadays Computer Security is pretentious by malicious data. Malware represents a serious threat to the security of computer systems. Traditional malware detection techniques like signature-based, heuristic-based, Specification-based detection are used to detect the known malware. These techniques detect the known malware accurately, but unable to detect the new, unknown malware. This paper presents a malware detection system based on the data mining and machine learning technique. The proposed method consists of disassemble process, feature extraction process and feature selection process. Three classification algorithms are employed on dataset to generate and train the classifiers named as Ripper, C4.5, IBk. The ensemble method i.e. Voting method is used to improve the accuracy of results. Here majority voting and veto voting are implemented the expected output is decided on the basis of majority and veto voting. The decision strategy of veto is improved by introducing the trust-based veto voting. This paper efficiently compared the results of majority voting, veto voting and trust-based veto voting. [ABSTRACT FROM AUTHOR]

Abstract This paper presents the Modified Grey Wolf Optimization (MGWO) algorithm which helps with the identification of the symptoms of Parkinson's disease at a premature stage. Parkinson disease is kind of a movement malady, which if not cured timely can prove to be fatal. Thus it becomes significant to identify Parkinson's disease at its premature phase so proper medications can provide longevity to patient by controlling the symptoms. In this work, a new model named Modified Grey Wolf Optimization (MGWO) has been proposed grounded on the traditional Grey Wolf Optimizer (GWO), which acts as a search strategy for feature selection. GWO is a meta-heuristic algorithm which is enthused by hunt down behavior of wolves. Random forest, k-nearest neighbor classifier and decision tree espy on selected features. The proposed model is evaluated using various types of datasets of voice, handwriting (spiral and meander) and speech. The put forward algorithm helps in the prediction of Parkinson disease with an estimated accuracy of 94.83%, detection rate of 98.28%, false alarm rate of 16.03% and further aid the individuals to receive a functional treatment at an early stage. The proposed bio-inspired algorithm is stable enough to find out the optimal subset of features. At last the results derived from the evaluation of proposed algorithm on datasets are compared with the results of Optimized Cuttlefish Algorithm (OCFA). The experimental results depict that the proposed algorithm helps in maximizing the accurateness and minimizing the number of features selected. [ABSTRACT FROM AUTHOR]

Web applications are paramount tools for facilitating services providing in the modern world. Unfortunately, the tremendous growth in the web application usage has resulted in a rise in cyberattacks. Cross-site scripting (XSS) is one of the most frequent cyber security attack vectors that threaten the end user as well as the service provider with the same degree of severity. Recently, an obvious increase of the Machine learning and deep learning ML/DL techniques adoption in XSS attack detection. The goal of this review is to come with a special attention and highlight of Machine learning and deep learning approaches. Thus, in this paper, we present a review of recent advances applied in ML/DL for XSS attack detection and classification. The existing proposed ML/DL approaches for XSS attack detection are analyzed and taxonomized comprehensively in terms of domain areas, data preprocessing, feature extraction, feature selection, dimensionality reduction, Data imbalance, performance metrics, datasets, and data types. Our analysis reveals that the way of how the XSS data is preprocessed considerably impacts the performance and the attack detection models. Proposing a full preprocessing cycle reveals how various ML/DL approaches for XSS attacks detection take advantage of different input data preprocessing techniques. The most used ML/DL and preprocessing stages have also been identified. The limitations of existing ML/DL-based XSS attack detection mechanisms are highlighted to identify the potential gaps and future trends. [ABSTRACT FROM AUTHOR]

Learning useful representations from original features is a key issue in classification tasks. Stacked autoencoders (SAEs) are easy to understand and realize, and they are powerful tools that learn deep features from original features, so they are popular for classification problems. The deep features can further combine the original features to construct more representative features for classification. However, existing SAEs do not consider the original features within the network structure and during training, so the deep features have low complementarity with the original features. To solve the problem, this paper proposes an embedded stacked group sparse autoencoder (ESGSAE) for more effective feature learning. Different from traditional stacked autoencoders, the ESGSAE model considers the complementarity between the original feature and the hidden outputs by embedding the original features into hidden layers. To alleviate the impact of the small sample problem on the generalization of the proposed ESGSAE model, an L 1 regularization-based feature selection strategy is designed to further improve the feature quality. After that, an ensemble model with support vector machine (SVM) and weighted local discriminant preservation projection (w_LPPD) is designed to further enhance the feature quality. Based on the designs above, an embedded stacked group sparse autoencoder ensemble with L 1 regularization and manifold reduction is proposed to obtain deep features with high complementarity in the context of the small sample problem. At the end of this paper, several representative public datasets are used for verification of the proposed algorithm. The results demonstrate that the ESGSAE ensemble model with L 1 regularization and manifold reduction yields superior performance compared to other existing and state-of-the-art feature learning algorithms, including some representative deep stacked autoencoder methods. Specifically, compared with the original features, the representative feature extraction algorithms and the improved autoencoders, the algorithm proposed in this paper can improve the classification accuracy by up to 13.33%, 7.33%, and 9.55%, respectively. The data and codes can be found in: https://share.weiyun.com/Jt7qeORm • A hybrid feature is embedded into the training process to construct a novel deep model. • A group sparsity constraint is introduced to obtain the sparse representations. • The ESGSAE ensemble model is constructed to obtain high complementary features. • A three-step feature learning mechanism is realized. [ABSTRACT FROM AUTHOR]

The curse of dimensionality is a common problem in classification tasks. However, feature selection is an exciting approach to deal with this type of problem by searching for a suboptimal feature set, either by eliminating irrelevant attributes, those with redundant information, or even both. Furthermore, although the use of machine learning techniques for the classification of drum cymbals is found in the literature, little attention has been paid to approaches that focus on classifying these instruments according to their bronze alloys. This paper aims to explore and evaluate the temporal information retrieved from audios, using TSFEL (Time Series Feature Extraction Library) as a tool to extract 18 temporal attributes, three feature selection approaches to assess these features, and logistic regression as a classifier. To this end, 276 audios referring to four drum cymbals of three different bronze alloys were captured in the studio through a sound acquisition procedure that took into account environment and microphone variations. Hence, one expects to find an optimal subset of features that contains enough information from the audios to achieve the best classification performance in the proposed problem. The experimental results show that a feature selection approach sequentially based on L 1 Regularization and Correlation Analysis was able to find a subset consisting of 5 attributes that achieved an average accuracy of 97.08%. • The computational model can identify the cymbals' bronze alloys through their sound. • A standardized impact procedure is developed for the audio signals acquisition. • Feature selection could reduce the dimensionality up to 70%, improving accuracy. • Logistic Regression with feature selection achieved high material classification rate. • High accuracy is attained even considering different sound capture perspectives. [ABSTRACT FROM AUTHOR]

The goal of few-shot learning is to use limited labeled samples to complete independent classification tasks. The feature extractor of few-shot learning needs to have a stronger feature expression ability to generalize in unseen novel classes. To further enhance the expressive ability, in this paper, we propose an inherited feature extraction method, named Base and Meta Feature Extraction (BMFE). Base feature represents the task-irrelevant classification information of each sample. Meta feature obtained by the proposed Triplet Meta-train Mechanism (TMM) inherits the classification information and also contains the task-related meta information of each sample. We concatenate both the base and meta features to complementarily express the rich information of each sample. Besides, instead of relying on limited support samples to obtain the prototype, we propose a novel unsupervised prototype correction module, named Prototype Self-updating (PSU). All unlabeled query samples in a few-shot test task participate in the iterative updating of each prototype in the task without training. Extensive experiments prove that our overall method can obtain richer features by BMFE and more accurate prototypes by PSU. Our overall method outperforms state-of-the-art methods on miniImageNet and tiredImageNet datasets, and especially under the 1-shot case we obtains 78.45% and 81.21% classification accuracy respectively. • BMFE can obtain the complementary features representation of FSL. • TMM mechanism can better perform meta-training of BMFE. • PSU can correct prototypes without training process. • The proposed CFPSU gets outstanding results on two standard datasets. [ABSTRACT FROM AUTHOR]

Disease diagnosis and prediction methods in biotechnology and medicine have significantly advanced over time. Consequently, analyzing raw gene expression is crucial for identifying diseases such as cancer. Interestingly, microarrays are a tool that records gene expression from deoxyribonucleic acid (DNA) or ribonucleic acid. This technique exhibits intriguing characteristics, such as generating high-dimensional data with a small sample size. However, in the case of such dataset, the classification model is prone to overfitting. This limitation can be overcome by reducing the dimensions of the microarray datasets to a reasonable number. Machine learning (ML)-based data reduction has recently achieved considerable attention in genomic research. Therefore, this review examines recent studies that present state-of-the-art data reduction and classification algorithms for microarray gene expression data to diagnose tumors and analyzes their performance. To the best of our knowledge, this is the first review that provides a comprehensive view of data preprocessing, dimensionality reduction, including feature (i.e., gene) selection, feature extraction, and their hybrid, and ML algorithms. The paper is structured as follows. First, this review summarizes several data preprocessing methods applied to gene expression datasets. Then, a detailed review of various ML-based feature selection algorithms, including filter, wrapper, embedded, ensemble, and hybrid algorithms, is discussed. These algorithms are examined under three main classes—supervised, unsupervised, and semisupervised ML. Next, the feature extraction and hybrid of feature extraction and selection algorithms are thoroughly reviewed. Furthermore, a detailed review of broadly applied ML algorithms to simplify tumor and nontumor classification using microarray datasets is presented. Finally, the challenges and open questions related to gene expression datasets for accurate cancer classification and detection are highlighted. [Display omitted] • A comprehensive review on microarray gene expression reduction is proposed. • A novel taxonomy of up-to-date gene reduction algorithms is presented. • The strength and limitations of each type of gene reduction algorithms are provided. • Provide an outline of the accurately used type of machine learning algorithms. • Analyze the performance of the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Machine-learning indirect test relies on powerful statistical algorithms to build prediction models that relate cheap measurements to costly performance metrics. Though many works in the past have been focused on proposing different models or on ways to improve the reliability of the results, it appears that the main bottleneck of the approach is the definition of an information-rich input space. Finding the appropriate measurements that are both cheap and meaningful is a task that has not yet been automated. In this framework, feature selection is a necessary tool to explore possible candidates. In this paper a hybrid method is proposed that lay between filtering and wrapper-based methods, trying to strike the right balance between accuracy and speed for the particular case of Alternate Test. [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]

Hemoglobin can be measured normally after the analysis of the blood sample taken from the body and this measurement is named as invasive. Hemoglobin must continuously be measured to control the disease and its progression in people who go through hemodialysis and have diseases such as oligocythemia and anemia. This gives a perpetual feeling of pain to the people. This paper proposes a non-invasive method for the prediction of the hemoglobin using the characteristic features of the PPG signals and different machine learning algorithms. In this work, PPG signals from 33 people were included in 10 periods and 40 characteristic features were extracted from them. In addition to these features, gender information (male or female), height (as cm), weight (as kg) and age of each subjects were also considered as the features. Blood count and hemoglobin level were measured simultaneously by using the “Hemocue Hb-201TM” device. Using the different machine learning regression techniques (classification and regression trees – CART, least squares regression – LSR, generalized linear regression – GLR, multivariate linear regression – MVLR, partial least squares regression – PLSR, generalized regression neural network – GRNN, MLP – multilayer perceptron, and support vector regression – SVR). RELIEFF feature selection (RFS) and correlation-based feature selection (CFS) were used to select the best features. Original features and selected features using RFS (10 features) and CFS (11 features) were used to predict the hemoglobin level using the different machine learning techniques. To evaluate the performance of the machine learning techniques, different performance measures such as mean absolute error – MAE, mean square error – MSE, R 2 (coefficient of determination), root mean square error – RMSE, Mean Absolute Percentage Error (MAPE) and Index of Agreement – IA were used. The promising results were obtained (MSE-0.0027) using the selected features by RFS and SVR. Hence, the proposed method may clinically be used to predict the hemoglobin level of human being clinically without taking and analyzing blood samples. [ABSTRACT FROM AUTHOR]

[Display omitted] • Multimodal occupancy detection framework for residential buildings. • Two variants of Occ-STPN models with feature level fusion and decision level fusion. • Feature extraction and occupancy detection models for various data modalities. • Transferability of a trained occupancy detection model from one building to another. • Occupancy prediction performance evaluation on three open-sourced datasets. For several years now, smart building energy systems have been a research area of intensive activity. In light of the increasing need for sustainable buildings and energy systems, this trend motivates an increasing need for a solution to reduce carbon dioxide emissions and improve energy efficiency. This paper proposes a high-performing and transferable occupancy detection framework that combines sensor data from different data modalities, including time series environmental data (temperature, humidity, and illuminance), image data, and acoustic energy data using ensemble method. To draw out the best prediction performance in each modality, the proposed framework was developed, including various models that were designed to learn the occupancy patterns reflected in the physical data streams. To tackle the time series environmental data, we designed two variants of an occupancy detection spatiotemporal pattern network (Occ-STPN) that performs both feature level and decision level fusion, respectively. We also propose a new metric; the fading memory mean square error (FMMSE), that provides a fair evaluation and penalization of delayed occupancy predictions. Multiple open-sourced datasets, including the Electricity Consumption and Occupancy and the University of California, Irvine's (UCI) building occupancy detection dataset, along with our own real data collected from six different houses, were used to validate the algorithms' performance. The experimental results presented herein break down the performance for each sensing modality, and a detailed analysis of the performance is also discussed. [ABSTRACT FROM AUTHOR]

In this paper, we develop a feature extraction model using the amalgamation of machine and deep learning techniques for Industrial Internet of Thing (IIoT). We train the model with the most effective feature set evaluated using machine learning algorithms and deep learning feature extraction methods. We test these features with deep learning based network models for validation. We consider error metrics and accuracy as the major factors for combining the machine learning and deep learning techniques. The error rates are analysed using mean square error which show low error rate for the subset than the model tested for full dataset. Further, mean square error, accuracy rate and false values are analysed to test the performance of the proposed model. The comparative analysis with IIoT dataset and existing methods show that our approach is 25% better than the others. [Display omitted] • We address the multi-dimensional problem of the IIoT features to make an adequate feature extraction method. • We use machine learning classifier and wrapper subset evaluation technique with Naive-Bayes and meta bagging methods for feature selection. Deep learning auto encoders are implemented as a part of dimensionality reduction techniques and provides a amalgamation for the appropriateness of the features. We verify them with Principal Component Analysis (PCA) and ranking. Thus, only relevant features are extracted for analysis. • We create network models for analysing detection using deep learning Feed forward neural network and Elman Back propagation techniques for classification and detection with various training methods. [ABSTRACT FROM AUTHOR]

Accurate hourly PM 2.5 concentration prediction plays a key role in air quality monitoring and controlling system, especially when severe haze occurs frequently. A PM 2.5 hourly prediction system is developed in this paper, based on an advanced data processing strategy, an effective feature selection technology and a novel optimization algorithm. First, the collected original sequence is decomposed into a group of filters with different wave frequencies and each filter is weighted and reconstructed to mitigate the negative impact of noisy fluctuations. Then mRMR is introduced for extracting interaction information between pollutants, further determining the input of artificial intelligence models. Whereafter, a five-component combined system is taken shape, in which BPNN, ELM, GRNN and BiLSTM are employed as foundation models while Multi-objective Water Cycle Algorithm (MOWCA) is the weight optimization model. The results of hourly PM 2.5 concentration prediction simulation experiment in the Bei–Shang–Guang–Shen area make clear that the developed system with minimum forecasting error, excellent generalization capability and robust prediction performance shows a definite latent capacity and future to deal with early warning problems and to design suitable abatement strategies. • Develop an hourly PM 2.5 concentration prediction system to support air pollution management. • Employ feature extraction technique to identify the optimal model's input. • Propose a new data preprocessing strategy to fetch the key information. • Integrate several forecasting models through optimization algorithm MOWCA. • Estimate the model's capability in a more professional and overall way. [ABSTRACT FROM AUTHOR]

Feature selection in real-world data mining problems is essential to make the learning task efficient and more accurate. Identifying the best feature selection algorithm, among the many available, is a complex activity that still relies heavily on human experts or some random trial-and-error procedure. Thus, the automated machine learning community has taken some steps towards the automation of this process. In this paper, we address the metalearning challenge of recommending feature selection algorithms by proposing a novel meta-feature engineering model. Our model considers a broad collection of meta-features that enable the study of the relationship between the dataset properties and the feature selection algorithm performance in terms of several criteria. We arrange the input meta-features into eight categories: (i) simple, (ii) statistical, (iii) information-theoretical, (iv) complexity, (v) landmarking, (vi) based on symbolic models, (vii) based on images, and (viii) based on complex networks (graphs). The target meta-features emerge from a multi-criteria performance measure, based on five individual performance indexes, that assesses feature selection methods grounded in information, distance, dependence, consistency, and precision measures. We evaluate our proposal using a recently developed framework that extracts the input meta-features from 213 benchmark datasets, and ranks the assessed feature selection algorithms, to fill in the target meta-features in meta-bases. This evaluation uses five state-of-the-art classification methods to induce recommendation models from meta-bases: C4.5, Random Forest, XGBoost, ANN, and SVM. The results showed that it is possible to reach an average accuracy of up to 90% applying our meta-feature engineering model. This work is the first to use an extensive empirical evaluation to provide a careful discussion of the strengths and limitations of more than 160 meta-features. These meta-features, while designed to aid the task of feature selection algorithm recommendation, can readily be employed in other metalearning scenarios. Therefore, we believe our findings are a valuable contribution to the fields of automated machine learning and data mining, as well as to the feature extraction and pattern recognition communities. • A novel meta-feature engineering model recommends feature selection algorithms. • The proposal obtains promising results from 213 datasets with hit rates of up to 90%. • Some simple, landmarking, image, and graph-based input meta-features highlighted. • A multi-criteria performance measure rigorously assesses candidate algorithms. • Chains of binary or multiclass classifiers can efficiently rank candidate algorithms. [ABSTRACT FROM AUTHOR]

• Optimal feature extraction from ultrasonic waveform essential for machine learning. • Convolutional neural networks pre-trained on auxiliary task. • Principal component analysis applied to CNN extracted features. • CNN feature extraction produced more informative features. • CNN method best for 65% of process monitoring tasks evaluated. Ultrasonic sensors are a low-cost and in-line technique and can be combined with machine learning for industrial process monitoring. However, training accurate machine learning models for process monitoring using sensor data is dependant on the feature selection methodology. This paper compares a convolutional feature extraction method to a traditional, coarse feature engineering approach. The convolutional method uses filter weights pre-trained on an auxiliary task to classify ultrasonic waveform dataset membership using previously obtained sensor data. The filter weights are used to extract features from the ultrasonic waveform. Principal component analysis is then applied to produce five principal components to be input into long short-term memory neural networks. The two approaches are compared on fermentation, mixing and cleaning datasets monitored using ultrasonic sensors. Overall, the convolutional feature method produced more informative waveform features than the coarse feature engineering approach, achieving higher model accuracy for datasets requiring substantial waveform information and for 65% of tasks overall. Multi-task learning also improved feature trajectory learning but led to reduced model accuracy for data points far from the classification decision boundaries. This can be overcome by further optimisation of neural network hyperparameters, though at increased model development time. Once trained, the convolutional feature extraction approach is a fast and convenient way of producing high quality features from ultrasonic waveforms using convolutional neural networks with little training data. [Display omitted] [ABSTRACT FROM AUTHOR]

Random Vector Functional Link (RVFL) is popular among researchers in many areas of machine learning. RVFL is preferred by many researchers as RVFL can produce good performance with relatively little training time. Recent works extend RVFL into deep and ensemble versions. However, RVFL does not have effective feature extraction methods commonly used in time series classification. This results in poor performance of RVFL in time series classification tasks. Also, deep RVFL is a relatively new and evolving area of research. In this paper, we present a framework that extracts features from Residual Networks (Resnet) and trains Ensemble Deep Random Vector Functional Link (edRVFL). We use features extracted from every residual block to train an ensemble of edRVFLs. We propose the following enhancements to edRVFL. Firstly, we diversity the structure of edRVFL and the direct link features to encourage diversity. Secondly, we built an ensemble of edRVFLs with the top two activation functions. Thirdly, we use two-stage tuning to save computational costs. Lastly, we perform a weighted average of all decisions made by every edRVFL. Experiments on the 55 largest UCR datasets show that using features extracted from all Residual blocks improves performance. All our proposed enhancements help improve classification accuracy or computational effort. Consequently, our proposed framework outperforms all traditional and deep learning-based time series classification methods. • Time series features extracted by ResNet are classified using edRVFL. • Using features extracted from every Residual Block improves edRVFL's performance. • 2-Stage tuning drastically reduces computational effort during tuning. • Diverse edRVFLs, feature selection in direct links and weighting of classifiers improve performance. • edRVFL incorporated with all proposed enhancements outperforms other time series classification algorithms. [ABSTRACT FROM AUTHOR]

Undoubtedly, social media, such as Facebook and Twitter, constitute a major part of our everyday life due to the incredible possibilities they offer to their users. However, Twitter and generally online social networks (OSNs) are increasingly used by automated accounts, widely known as bots, due to their immense popularity across a wide range of user categories. Their main purpose is the dissemination of fake news, the promotion of specific ideas and products, the manipulation of the stock market and even the diffusion of sexually explicit material. Therefore, the early detection of bots in social media is quite essential. In this paper, two methods are introduced targeting this that are mainly based on Natural Language Processing (NLP) to distinguish legitimate users from bots. In the first method, a feature extraction approach is proposed for identifying accounts posting automated messages. After applying feature selection techniques and dealing with imbalanced datasets, the subset of features selected is fed in machine learning algorithms. In the second method, a deep learning architecture is proposed to identify whether tweets have been posted by real users or generated by bots. To the best of the authors' knowledge, there is no prior work on using an attention mechanism for identifying bots. The introduced approaches have been evaluated over a series of experiments using two large real Twitter datasets and demonstrate valuable advantages over other existing techniques targeting the identification of malicious users in social media. [ABSTRACT FROM AUTHOR]

• Most effective temporal features from smart meter data for building characterization are found. • State-of-the-art feature selection methods and a proposed customized approach are utilized. • Importance-in-prediction based features that improve building classification are determined. • Number of utilized features is substantially reduced while the accuracy is also enhanced. The present paper aims at determining the most influential features to be extracted from smart meter data to facilitate machine learning-based classification of non-residential buildings. Smart meter-driven remote estimation of the chosen characteristics (the buildings' performance class, use type, and operation group) is significantly helpful in buildings' commissioning, benchmarking, and diagnostics applications. As the first step, state-of-the-art feature selection methods and a proposed customized approach are utilized for determining the most influential parameters in the pool of temporal features, proposed in a previous study. Next, importance-in-prediction based features, generated from an hour-ahead load prediction pipeline, that can improve the classification accuracy are proposed and added as additional input parameters. Finally, interpretations about some of the most influential features for different classification targets are provided. The obtained results demonstrate that, while aiming at estimating the buildings' use type, through performing feature selection and adding importance-in-prediction based features, the number of utilized features is reduced from 290 (initial pool of features proposed in a previous study) to 29, while also increasing the accuracy from 71% to 74%. Similarly, number of employed features for estimating the performance class is decreased from 224 to 17 and the achieved accuracy is improved from 56% to 62%. Finally, using only 6 selected features, compared to 287 features in the initial set, the obtained accuracy for the classification of operation group is increased from 98% to 100%. It is thus demonstrated that the proposed methodology, through selecting and utilizing notably fewer features, results in a notable simplification of the feature extraction procedures, improves the achieved accuracy, and facilitates providing interpretations about the reason behind the influence of some of the most important features. [ABSTRACT FROM AUTHOR]

In the context of natural language processing, accuracy of intention detection is the basis for subsequent research on human-machine speech interaction. However, the problem of ambiguity in word vectors reduces the accuracy of intent detection. Meantime, there is a disconnection between local features and global features as well, resulting in text feature extraction that cannot fully reflect semantic information. These issues are all barriers of intention detection. Therefore, this paper proposes an attention-based convolutional neural network for self-media data learning (called A-CNN) for marketing intention. We cascade the traditional CNN with the self-attention model in the Attention networks to form a new network structure called A-CNN, and put forward a fast feature extraction method based on skip-gram-based learning called FSLText, to represent the high-dimension word vectors in the A-CNN. On the premise of maintaining the advantages of the CNN, A-CNN can not only solve the problem of local and global features disconnection caused by the CNN pooling layer, but also avoid the increase of algorithm complexity. The Self-Attention mechanism in the Attention model can effectively optimize the weight of local features of the information in global features, and retain local features that are more useful for intention detection. A fast feature extraction method which is based on Skip-gram can retain the semantic and word order information of the text. The method is beneficial to the marketing intention detection. According to the experiment, our A-CNN, compared with traditional machine learning methods, can improve 12.32% accuracy. Contrast to the dual-channel CNN, the accuracy rate is improved by 9.68%, and compared with the ATT-CNN, it is improved by 9.97%. On the F1 score, the A-CNN can improve the F1 score by about 9.37% in comparison with the traditional machine learning methods, the accuracy rate is increased by 9.68% compared with the dual-channel CNN, and 9.68% in contrast with ATT-CNN. It illustrates that our A-CNN can effectively address semantic and feature selection for marketing intention detection. [ABSTRACT FROM AUTHOR]

• The main objective of this work is to design and develop a classifier for oral squamous cell carcinoma (OSCC) originating in stratified squamous epithelial tissue of oral cavity. The classifier is named as Stratified Squamous Epithelium Biopsy Image Classifier (SSE-BIC). This algorithm automatize the detection and or classification process which is still a tedious work being done manually via visual inspection by experts. • A total 676 images have been used to design, train and test the classifier. As many as 305 features have been computed from H&E-stained microscopic images of oral mucosa which include colour features, textural features, gradient features, shape features and tamura features to design the SSE-BIC. We used, unsupervised data mining technique to extract cellular regions from the images as a part of shape feature extraction process. • The accuracy observed as an exhaustive simulation is found to be 95.56% using H&E-stained microscopic colour images which has very complicated structure as compared to cytology images. Squamous cell carcinoma (SCC) of oral cavity is the most common among oral cancer patients. In this paper, we have developed machine learning based automatic oral squamous cell carcinoma (OSCC) classifier named as Stratified Squamous Epithelial Biopsy Image Classifier (SSE-BIC) to categorize H&E-stained microscopic images of squamous epithelial layer in four different classes: normal, well-differentiated, moderately-differentiated and poorly-differentiated. Five classifiers are used to perform the classification by maximum voting method. Total 305 features are extracted from the images of oral mucosa which include color features, textural features, gradient features, geometrical features and tamura features. Unsupervised data mining is used for segmenting the cellular area to compute geometrical features of the cells retaining color details of the images. Feature selection has been performed by neighborhood component feature selection (NCFS) technique. Total 676 images have been used to design, train and test the classifier. A detailed performance analysis is presented with individual feature sets and hybrid feature sets with feature selection applied using individual classifiers as well as proposed classifier. The proposed classifier achieves overall accuracy of 95.56%. This can account for first level of automatic screening of the biopsy images. [ABSTRACT FROM AUTHOR]