Combination Of Feature Extraction Methods Research Articles

A Combination of Transfer Learning and Support Vector Machine for Robust Classification on Small Weed and Potato Datasets

Agriculture is the primary sector in Indonesia for meeting people's daily food demands. One of the agricultural commodities that replace rice is potatoes. Potato growth needs to be protected from weeds that compete for nutrients. Spraying using pesticides can cause environmental pollution, affecting cultivated plants. Currently, agricultural technology is being developed using an Artificial Intelligence (AI) approach to classifying crops. The classification process using AI depends on the number of datasets obtained. The number of datasets obtained in this research is not too large, so it requires a particular approach regarding the AI method used. This research aims to use a combination of feature extraction methods with local and deep feature approaches with supervised machine learning to classify of small datasets. The local feature method used in this research is Local Binary Pattern (LBP) and Histogram of Oriented Gradients (HOG), while the deep feature method used is MobileNet and MobileNetV2. The famous Support Vector Machine (SVM) uses the classification method to separate two data classes. The experimental results showed that the local feature HOG method was the fastest in the training process. However, the most accurate result was using the MobileNetV2 deep feature method with an accuracy of 98%. Deep features produced the best accuracy because the feature extraction process went through many neural network layers. This research can provide insight on how to analyze a small number of datasets by combining several strategies

Development and Validation of Machine Models Using Natural Language Processing to Classify Substances Involved in Overdose Deaths

Overdose is one of the leading causes of death in the US; however, surveillance data lag considerably from medical examiner determination of the death to reporting in national surveillance reports. To automate the classification of deaths related to substances in medical examiner data using natural language processing (NLP) and machine learning (ML). Diagnostic study comparing different natural language processing and machine learning algorithms to identify substances related to overdose in 10 health jurisdictions in the US from January 1, 2020, to December 31, 2020. Unstructured text from 35 433 medical examiner and coroners' death records was examined. Text from each case was manually classified to a substance that was related to the death. Three feature representation methods were used and compared: text frequency-inverse document frequency (TF-IDF), global vectors for word representations (GloVe), and concept unique identifier (CUI) embeddings. Several ML algorithms were trained and best models were selected based on F-scores. The best models were tested on a hold-out test set and results were reported with 95% CIs. Text data from death certificates were classified as any opioid, fentanyl, alcohol, cocaine, methamphetamine, heroin, prescription opioid, and an aggregate of other substances. Diagnostic metrics and 95% CIs were calculated for each combination of feature extraction method and machine learning classifier. Of 35 433 death records analyzed (decedent median age, 58 years [IQR, 41-72 years]; 24 449 [69%] were male), the most common substances related to deaths included any opioid (5739 [16%]), fentanyl (4758 [13%]), alcohol (2866 [8%]), cocaine (2247 [6%]), methamphetamine (1876 [5%]), heroin (1613 [5%]), prescription opioids (1197 [3%]), and any benzodiazepine (1076 [3%]). The CUI embeddings had similar or better diagnostic metrics compared with word embeddings and TF-IDF for all substances except alcohol. ML classifiers had perfect or near perfect performance in classifying deaths related to any opioids, heroin, fentanyl, prescription opioids, methamphetamine, cocaine, and alcohol. Classification of benzodiazepines was suboptimal using all 3 feature extraction methods. In this diagnostic study, NLP/ML algorithms demonstrated excellent diagnostic performance at classifying substances related to overdoses. These algorithms should be integrated into workflows to decrease the lag time in reporting overdose surveillance data.

Recent Advances in Classification of Brain Tumor from MR Images - State of the Art Review from 2017 to 2021.

The task of identifying a tumor in the brain is a complex problem that requires sophisticated skills and inference mechanisms to accurately locate the tumor region. The complex nature of the brain tissue makes the problem of locating, segmenting, and ultimately classifying Magnetic Resonance (MR) images a complex problem. The aim of this review paper is to consolidate the details of the most relevant and recent approaches proposed in this domain for the binary and multi-class classification of brain tumors using brain MR images. In this review paper, a detailed summary of the latest techniques used for brain MR image feature extraction and classification is presented. A lot of research papers have been published recently with various techniques proposed for identifying an efficient method for the correct recognition and diagnosis of brain MR images. The review paper allows researchers in the field to familiarize themselves with the latest developments and be able to propose novel techniques that have not yet been explored in this research domain. In addition, the review paper will facilitate researchers who are new to machine learning algorithms for brain tumor recognition to understand the basics of the field and pave the way for them to be able to contribute to this vital field of medical research. In this paper, the review is performed for all recently proposed methods for both feature extraction and classification. It also identifies the combination of feature extraction methods and classification methods that, when combined, would be the most efficient technique for the recognition and diagnosis of brain tumor from MR images. In addition, the paper presents the performance metrics, particularly the recognition accuracy, of selected research published between 2017-2021.

Machine Learning and Radiomic Features to Predict Overall Survival Time for Glioblastoma Patients.

Glioblastoma is an aggressive brain tumor with a low survival rate. Understanding tumor behavior by predicting prognosis outcomes is a crucial factor in deciding a proper treatment plan. In this paper, an automatic overall survival time prediction system (OST) for glioblastoma patients is developed on the basis of radiomic features and machine learning (ML). This system is designed to predict prognosis outcomes by classifying a glioblastoma patient into one of three survival groups: short-term, mid-term, and long-term. To develop the prediction system, a medical dataset based on imaging information from magnetic resonance imaging (MRI) and non-imaging information is used. A novel radiomic feature extraction method is proposed and developed on the basis of volumetric and location information of brain tumor subregions extracted from MRI scans. This method is based on calculating the volumetric features from two brain sub-volumes obtained from the whole brain volume in MRI images using brain sectional planes (sagittal, coronal, and horizontal). Many experiments are conducted on the basis of various ML methods and combinations of feature extraction methods to develop the best OST system. In addition, the feature fusions of both radiomic and non-imaging features are examined to improve the accuracy of the prediction system. The best performance was achieved by the neural network and feature fusions.

A combination of feature extraction methods and deep learning for brain tumour classification

This paper presents a method for categorizing tumour disease from magnetic resonance imaging images using a convolutional neural network. The proposed technique consists of three major phases, including feature extraction, feature selection, and combination. The authors considered the classification method using convolutional neural network without any pre-processing on input images as the original method. The original method is then improved in some sequential phases when convolutional neural network uses features generated by feature extraction methods. Many popular feature extraction methods in generating the input features of the network are examined. After examining the results, a set of feature extraction schemes with appropriate performance are selected for the next phase. Also, the authors assigned weight factors to each of the selected methods, according to their accuracy. By assigning these weight factors to the methods, the network's accuracy increased to an acceptable level compared to the first and second observations. An accuracy above 99.76% was achieved, which is a 2% improvement using these feature extraction methods compared to the original method. The authors also added a third class named ‘I do not know’ to increase the tumour detection problem's reliability in the last phase. The authors have successfully introduced a convolutional neural network architecture with high accuracy and reliability.

South China Sea Conflicts Classification Using Named Entity Recognition (NER) and Part-of-Speech (POS) Tagging

Internet connects everyone to everything globally. The existence of Internet eases people in completing daily tasks. Thanks to Internet, information is being digitalized and spread openly to the public. Online news articles not only provide us with useful and reliable information and reports, it also eases information extraction and gathering for research purposes especially in Natural Language Processing (NLP) and Machine Learning (ML). The topics regarding the South China Sea have been popular lately due to the rise of conflicts between several countries claim on the islands in the sea. Gathering data through Internet and online sources proves to be easy, but to process a huge amount data and to identify only useful information manually takes a longer time to complete. Extracting important features from a text document can be done by using one or a combination of feature extraction methods. Relevant information and the classification of news articles in relation to the conflicts in South China Sea need to be done. In this paper, a model is proposed to use Named Entity Recognition (NER) that search for and classifies important information regarding to the conflicts. In order to do that, a combination of Part-of-Speech (POS) and NER are needed to extract type of conflicts from the news. This study also claims to classify news by using Conditional Random Field (CRF) algorithm and Multinomial Naïve Bayes (MNB) as classification methods by training and testing the data.

Identification of SNARE Proteins Through a Novel Hybrid Model

SNARE proteins are a large family of membrane fusion proteins. As a lot of human diseases are related to the SNARE proteins, they have attracted people to study them. Traditionally, the SNARE proteins can be identified through bioinformatics techniques, which are expensive and time-consuming. Some researchers attempt to identify the SNARE proteins by the machine learning algorithms. A deep learning model called SNARE-CNN is proposed to predict SNARE proteins. A 2D convolutional neural network is constructed and the Position-Specific Scoring Matrix (PSSM) profile is used to distinguish the SNARE proteins from the other kinds of proteins. Although the SNARE-CNN can achieve high accuracy, the performance of the model still has room to improve. In this paper, a novel hybrid model, that combines the random forest algorithm with the oversampling filter and 188D feature extraction method, is proposed. By trying different combinations of feature extraction methods, filtering methods and classification algorithms, the hybrid model, we proposed, can achieve the best performance among all combinations. Experiments show that the performance of our hybrid model is better than that of the SNARE-CNN model.

Improving identification performance in iris recognition systems through combined feature extraction based on binary genetics

The process of iris recognition is composed of several stages. The most important stage is feature extraction. The majority of systems use a single method of feature extraction. We used a binary genetic algorithm with a new fitness criterion to find an efficient combination of feature extraction methods with the aim of improving the performance of iris recognition systems. The proposed method makes use of a large number of filters and transforms which are widely used to extract iris features. The best combination of these techniques is found over iterations of the proposed algorithm. The final combination, which is argued to be the most efficient method of feature extraction, consists of a number of wavelet transforms, Gabor filter, and Fourier transform. ROC graphs are used to show the experimental superiority of the performance of this method over methods that use a single filter. According to our findings, the proposed method shows better performance in most situations. An FAR value of 0 and an FRR value of 0.092 was obtained by this method.

Human Activity Recognition Based on Weighted Sum Method and Combination of Feature Extraction Methods

Human Activity Recognition (HAR) is one of the most important areas of computer vision research. The biggest difficulty for HAR system is that the camera could only film in one direction, leading to a shortage of data and low recognition results. This paper focuses on researching and building new models of HAR, including Principal Components Analysis (PCA), Linear discriminant Analysis (LDA) is to reduce the dimensionality and size of data, contributing to high recognition accuracy. First, from the 3D motion data, we conducted a pretreatment and feature extraction of objects. Next, we built a recognition model corresponding to each feature extraction method and we used Support Vector Machine (SVM) model to train. Finally, we used weighted methods to combine the results of the model to train and give the final results. The paper experiment on CMU MOCAP database and the percentage receiving proposed method is higher than that from the previous method.

Index Finger Motion Recognition Using Self-Advise Support Vector Machine

abstract Because of the functionality of an index finger, the disability of its motion in the modern age can decrease the person’s quality of life. As a part of rehabilitation therapy, the recognition of the index finger motion for rehabilitation purposes should be done properly. This paper proposes a novel recognition system of the index finger motion suing a cutting-edge method and its improvements. The proposed system consists of combination of feature extraction method, a dimensionality reduction and well-known classifier, Support Vector Machine (SVM). An improvement of SVM, Self-advise SVM (SA-SVM), is tested to evaluate and compare its performance with the original one. The experimental result shows that SA-SVM improves the classification performance by on average 0.63 %.

A Hybrid Feature Extraction Method for Accuracy Improvement in “Aksara Lontara” Translation

An Optical Character Recognition (OCR) of “Aksara Lontara” has been constructed using a novel combination of feature extraction methods in this study. The ancient font of “Lontara” is then translated into Bahasa Indonesia to help non-native language to learn this language. Two powerful extraction feature methods, i.e., Modified Direction Feature (MDF) and Fourier Descriptor (FD) are stages combined to deal with two dominant phases of the Lontara font. The classification process is conducted using Support Vector Machine (SVM) as a fast and straightforward learning method deal with 23 fonts in image containing of 150×120 pixels. In this research, 50 verbs were used for training and 30 verbs for validating the system. The results show that system can reach 96% accuracy using this hybrid in extraction feature with kernel variable of C = 3 and σ = 8.

Telugu character recognition for degraded palm leaf documents using optimal feature selection techniques - a 3D approach

Palm leaves were used as a medium of recording information about 700 years ago. This work deals with the recognition of Telugu palm leaf characters by acquiring 3D data using a contact-type profiler. A novel concept of using a 3D inherent feature, i.e. depth of incision is proposed to eliminate noise. With the help of this 3D feature, improved recognition accuracy is also reported for various features extracted from the palm leaf characters. Experiments are conducted by implementing optimisation techniques, such as differential evolution and particle swarm optimisation, to find the optimum number of features to reduce the memory needed. With varying feature dimensions, average classification accuracies are reported for combination of feature extraction methods and optimisation techniques.

Fuzzy clustering-based feature extraction method for mental task classification

A brain computer interface (BCI) is a communication system by which a person can send messages or requests for basic necessities without using peripheral nerves and muscles. Response to mental task-based BCI is one of the privileged areas of investigation. Electroencephalography (EEG) signals are used to represent the brain activities in the BCI domain. For any mental task classification model, the performance of the learning model depends on the extraction of features from EEG signal. In literature, wavelet transform and empirical mode decomposition are two popular feature extraction methods used to analyze a signal having non-linear and non-stationary property. By adopting the virtue of both techniques, a theoretical adaptive filter-based method to decompose non-linear and non-stationary signal has been proposed known as empirical wavelet transform (EWT) in recent past. EWT does not work well for the signals having overlapped in frequency and time domain and failed to provide good features for further classification. In this work, Fuzzy c-means algorithm is utilized along with EWT to handle this problem. It has been observed from the experimental results that EWT along with fuzzy clustering outperforms in comparison to EWT for the EEG-based response to mental task problem. Further, in case of mental task classification, the ratio of samples to features is very small. To handle the problem of small ratio of samples to features, in this paper, we have also utilized three well-known multivariate feature selection methods viz. Bhattacharyya distance (BD), ratio of scatter matrices (SR), and linear regression (LR). The results of experiment demonstrate that the performance of mental task classification has improved considerably by aforesaid methods. Ranking method and Friedman’s statistical test are also performed to rank and compare different combinations of feature extraction methods and feature selection methods which endorse the efficacy of the proposed approach.

Survival analysis for high-dimensional, heterogeneous medical data: Exploring feature extraction as an alternative to feature selection.

In clinical research, the primary interest is often the time until occurrence of an adverse event, i.e., survival analysis. Its application to electronic health records is challenging for two main reasons: (1) patient records are comprised of high-dimensional feature vectors, and (2) feature vectors are a mix of categorical and real-valued features, which implies varying statistical properties among features. To learn from high-dimensional data, researchers can choose from a wide range of methods in the fields of feature selection and feature extraction. Whereas feature selection is well studied, little work focused on utilizing feature extraction techniques for survival analysis. We investigate how well feature extraction methods can deal with features having varying statistical properties. In particular, we consider multiview spectral embedding algorithms, which specifically have been developed for these situations. We propose to use random survival forests to accurately determine local neighborhood relations from right censored survival data. We evaluated 10 combinations of feature extraction methods and 6 survival models with and without intrinsic feature selection in the context of survival analysis on 3 clinical datasets. Our results demonstrate that for small sample sizes - less than 500 patients - models with built-in feature selection (Cox model with ℓ1 penalty, random survival forest, and gradient boosted models) outperform feature extraction methods by a median margin of 6.3% in concordance index (inter-quartile range: [-1.2%;14.6%]). If the number of samples is insufficient, feature extraction methods are unable to reliably identify the underlying manifold, which makes them of limited use in these situations. For large sample sizes - in our experiments, 2500 samples or more - feature extraction methods perform as well as feature selection methods.

A Combination of Feature Extraction Methods with an Ensemble of Support Vector Machines for Bacterial Virulent Proteins Prediction

A Combination of Feature Extraction Methods with an Ensemble of Support Vector Machines for Bacterial Virulent Proteins Prediction

Face Recognition using Two-dimensional Subspace Analysis and PNN

paper, we present an new approach to face recognition based on the combination of feature extraction methods, such as two-dimensional DWT-2DPCA and DWT-2DLDA, with a probabilistic neural networks. This later is used to classify the features matrix extracts for space data created by Two- dimensional Subspace Analysis .The technique 2D-DWT is used to eliminate the illumination ,noise and redundancy of face in order to reduce calculations of the probabilistic neural network operations ,and improve a face recognition system in accuracy and computation time. The proposed approach is tested on ORL and FEI face databases. Experimental results on this databases demonstrated the effectiveness of the proposed approach for face recognition with high accuracy compared with previous methods..

A Combination of Feature Extraction Methods with an Ensemble of Different Classifiers for Protein Structural Class Prediction Problem

Better understanding of structural class of a given protein reveals important information about its overall folding type and its domain. It can also be directly used to provide critical information on general tertiary structure of a protein which has a profound impact on protein function determination and drug design. Despite tremendous enhancements made by pattern recognition-based approaches to solve this problem, it still remains as an unsolved issue for bioinformatics that demands more attention and exploration. In this study, we propose a novel feature extraction model that incorporates physicochemical and evolutionary-based information simultaneously. We also propose overlapped segmented distribution and autocorrelation-based feature extraction methods to provide more local and global discriminatory information. The proposed feature extraction methods are explored for 15 most promising attributes that are selected from a wide range of physicochemical-based attributes. Finally, by applying an ensemble of different classifiers namely, Adaboost.M1, LogitBoost, naive Bayes, multilayer perceptron (MLP), and support vector machine (SVM) we show enhancement of the protein structural class prediction accuracy for four popular benchmarks.

Combination of Feature Extraction Methods for SVM Pedestrian Detection

This paper describes a comprehensive combination of feature extraction methods for vision-based pedestrian detection in Intelligent Transportation Systems. The basic components of pedestrians are first located in the image and then combined with a support-vector-machine-based classifier. This poses the problem of pedestrian detection in real cluttered road images. Candidate pedestrians are located using a subtractive clustering attention mechanism based on stereo vision. A components-based learning approach is proposed in order to better deal with pedestrian variability, illumination conditions, partial occlusions, and rotations. Extensive comparisons have been carried out using different feature extraction methods as a key to image understanding in real traffic conditions. A database containing thousands of pedestrian samples extracted from real traffic images has been created for learning purposes at either daytime or nighttime. The results achieved to date show interesting conclusions that suggest a combination of feature extraction methods as an essential clue for enhanced detection performance