Learning-based Classification Models Research Articles

Intelligent system for assembly-line worker’s fatigue recognition and facilitation

The fourth industrial revolution, or Industry 4.0 paradigm, gives scope to resolve the challenges of 21st-century manufacturing. Physical stress or worker’s work-related musculoskeletal disorders (WMSDs) remain a challenge to the manual assembly line production system. This research provides a new intelligent framework to recognize and facilitate an assembly line fatigue worker. The intelligent framework provides a real-time fatigue status of each assembly worker on the manager’s work screen. Furthermore, the intelligent system provides a suitable solution to facilitate the fatigue worker, without disturbing the line balance. A novel approach is given to recognize the assembly line workers’ physical stress or work fatigue using the Knowledge-based Intelligent System (KBIS) methodology. Similarly, intelligent work rotation algorithms are proposed to find the solutions to facilitate the fatigue worker. Details of intelligent fatigue recognition, making of a learning-based fatigue recognition model, Industrial Internet of Things (IIoT) setup for worker’s real-time status monitoring, and Pseudocodes of intelligent work rotation algorithms are discussed in detail. Furthermore, an illustrative example is given to demonstrate the scope of the proposed framework for a manual assembly line system. Four worker assembly line is considered during the illustration. Details of intelligent fatigue worker recognition with their current-status is explained. The intelligent algorithm has recognised that the worker 4 is under fatigue condition. Furthermore, the intelligent facilitation algorithm found the worker 6 as alternative worker to facilitate the fatigue worker. Learning-based classification models are used to make fatigue recognition model. However, Random Forest model has shown better performance in terms of prediction accuracy while recognising the worker’s fatigue status. Similarly, worker’s work capability models are used to find a suitable alternative worker to facilitate the fatigue worker.

Machine Learning Based Classification Model for Prediction of Bank Loan Approval

Abstract: In this paper we creating and testing a smart loan prediction system using a special type of classifier called gradient boosting. Our methodology is meticulously laid out, covering a spectrum of essential stages including data pre-processing methodologies, in-depth exploratory analysis, and rigorous model training leveraging a dataset meticulously curated from Kaggle. Our model, with 300 estimators, a learning rate of 0.05, and a maximum of six features, achieves an astounding 98.03% accuracy. We utilize the pickle module to serialize our trained model, consequently deploying it as a web application. This application, which is powered by React for the front end and Flask for the back end, offers users a simple interface for entering loan application details and receiving real-time forecasts on loan acceptance or denial. Our research shows that gradient boosting classified the well in predicting the loans to the users.

Design of Barnacle Mating Optimizer with Deep Learning Based Classification Model for Medical X-Ray Images

Design of Barnacle Mating Optimizer with Deep Learning Based Classification Model for Medical X-Ray Images

Learning adaptive coarse basis functions of FETI-DP

Domain decomposition methods are successful and highly parallel scalable iterative solution methods for discretized partial differential equations. Nevertheless, for many classes of problems, for example, elliptic partial differential equations with arbitrary coefficient distributions, adaptive coarse spaces are necessary to obtain robustness or, in other words, to guarantee a reliable and fast convergence. Adaptive coarse spaces are usually computed by solving many localized eigenvalue problems related to edges or faces of the domain decomposition. This results in a computationally expensive setup of the domain decomposition preconditioner or system operator. In earlier work, to which the authors have contributed, a deep learning based classification model was used to classify the critical edges or faces where eigenvalue problems have to be solved. In the present paper, we suggest to directly learn the adaptive constraints using a deep feedforward neural network regression model and thus completely skip the computationally most expensive part of the setup, i.e., the solution of local eigenvalue problems. We consider a specific adaptive FETI-DP (Finite Tearing and Interconnecting – Dual Primal) approach and concentrate on stationary diffusion problems in two dimensions with arbitrary coefficient functions with large jumps. As an input for the neural network, we use an image representation of the coefficient function which resolves the structure of the coefficient distribution but is not necessarily identical to the discretization of the partial differential equation. Therefore, our approach is independent of the finite element mesh and can, in principle, be easily extended to other adaptive coarse spaces, problems, and domain decomposition methods. We show the robustness of our method for different problems and the generalization property of our trained neural networks by considering different coefficient distributions not contained in the training set. We also combine the learned constraints with computationally cheap frugal constraints to further improve our approach.

An Automatic Non-Destructive External and Internal Quality Evaluation of Mango Fruits based on Color and X-ray Imaging with Machine Learning and Deep Learning Based Classification Models

Quality evaluation of food products, agricultural produce to be specific, has gained momentum from past few decades due to the increased awareness among consumers across the world. This has resulted in the increased emphasis on the development and use of quality assessment techniques in food industry. Moreover, there is a need to automate the quality monitoring of agricultural produce like fruits and vegetables which is otherwise done manually in developing countries hence labor intensive, time consuming and subjective in nature. This paper presents an empirical analysis to build a rapid, robust, real-time, non-destructive computer vision based quality assessment model for mango fruits. The work employs the automatic disease classification of mango fruits based on machine and deep learning models. Firstly, the dataset of colored mango fruits images with 2279 images falling into three classes and another dataset of soft X-ray images of mango fruits with 572 images belonging to two quality classes are developed for detecting external and internal defects, respectively. The multilayer perceptron neural network (MLP NN) with two hidden layers, which may be considered as the starting point for deep learning technique, is proposed as machine learning model to classify the color images of mango fruits into one of three external quality classes with 95.1% accuracy and also to classify the soft X-ray images into two internal quality classes with 97.5% accuracy. In order to step out of feature engineering, actual deep learning convolutional neural network (CNN) models, a customized CNN model and pre-trained CNN models, VGGNet (VGG16) and DenseNet121 were also explored for mango disease classification. The maximum validation accuracy of custom CNN was found to be with 91.52% and 98.7% for color and augmented X-ray images, respectively. The classification accuracy of pre-trained models were found to be reasonably good for the color images but exhibited high variability in results and made it difficult to draw a general conclusion for the proposed datasets. However, the proposed MLP NN model based on few basic intensity and geometric features and also the proposed customized CNN model were found to be the best models and they outperform the state of the art reported in the literature.

Machine Learning Based Classification Model for Network Traffic Anomaly Detection

In current days, cloud environments are facing a huge challenge from the attackers in terms of various attacks thrown to the cloud service providers. In both industry and academics, the problem of detection and mitigation of DDoS attacks is now a challenging issue. Detecting Distributed Denial of Service (DDos) threats is mainly a classification problem that can be addressed using data mining, machine learning and deep learning techniques. DDoS attacks can occur in any of the seven-layer OSI model's network. Hence, detecting the DDoS attacks is an important task for cloud service providers to overcome dangerous attacks and loss incurred to stake holders and also the provider.

Different Deep Learning Based Classification Models for COVID-19 CT-Scans and Lesion Segmentation Through the cGAN-UNet Hybrid Method

Different Deep Learning Based Classification Models for COVID-19 CT-Scans and Lesion Segmentation Through the cGAN-UNet Hybrid Method

Neurocybernetic Assistive Technologies to Enhance Robotic Wheelchair Navigation

Millions of individuals worldwide face mobility impairments and rely on wheelchairs for assistance. Yet a significant portion of individuals with severe motor disabilities or insufficient familiarity with conventional interfaces cannot consider powered wheelchairs as a viable solution. Neurocybernetics present a promising solution to this challenge. Neurocybernetics involves exploring how the nervous system interacts with systems, like computers and robots. Through the use of interfaces users can control wheelchairs using neural signals from their brains making navigation more intuitive and effective. Assistive technologies, especially robotic wheelchairs powered by sensors and control systems, are essential for those disabled people, with mobility issues, to live more independently and better lives in their daily life even without the assistance of others. The primary objective of this study is to improve the quality of life for individuals with disabilities who depend on such assistance for their mobility requirements. This article examines some of the cutting-edge neurocybernetic devices that assist people with disabilities, such as brain-computer interfaces (BCIs), algorithms to understand brain signals, and adaptive control methods for wheelchairs. This study proposes a robotic wheelchair which utilizes the human brain signal and undergoes machine learning based classification model with neuromorphic adaptive control. It discusses the key considerations, difficulties, and potential uses of neurocybernetic robotic wheelchairs in real life situations. Additionally, it addresses the ethical, legal, and societal implications of using these technologies and suggests future research areas to advance their creation and use.

Explainability of a Deep Learning Based Classification Model for ANCA-Associated Glomerulonephritis

Explainability of a Deep Learning Based Classification Model for ANCA-Associated Glomerulonephritis

Alzheimer disease classification using tawny flamingo based deep convolutional neural networks via federated learning

ABSTRACT Smart health-care is the recent technology, which ensures the early diagnosis and prevention to the patients in the remote area. Particularly, for the serious illness, like cardiac problems, brain abnormalities require immediate attention. Moreover, owing to the commencement of the smart systems, the data from distributed sources is bulky, which imposes the complexity to handle and degrades the diagnosis accuracy. Hence, this research proposes an innovative distributed learning based classification model named federated learning dependent Tawny flamingo-based deep CNN classifier for the disease classification, which handles the clinical data from the distributed sources and ensures the disease classification with good accuracy. In this research, a tawny flamingo-based deep CNN is proposed to detect the Alzheimer’s disease, where the parameter of the classifier is tuned by the tawny flamingo algorithm. The tawny flamingo optimization reveals the maximal accuracy of 98.252% for K-fold and 97.995% for training percentage dependent analysis.

Machine Learning–Driven Model to Analyze Particular Conditions of Contracts: A Multifunctional and Risk Perspective

Contracts can serve three functions—Control, Coordination, and Adaptation—to mitigate different risks. This multifunctional perspective based on the development of contract theory has been introduced in construction industries to analyze how general conditions of contracts deal with risks. However, the particular conditions of contracts present specific requirements closely related to projects, which can generate additional risks for both parties. To fill in the gap, this study analyzes nine particular conditions from nine projects with good performance respectively based on the Fédération Internationale Des Ingénieurs-Conseils (FIDIC) 1999 Silver Book and explores what aspects good particular conditions focus more on to avert risks when modifying the general conditions. We built machine learning–based classification models to conduct a content analysis of complete contracts (containing both general conditions and particular conditions) to measure the functions of contracts. The optimal model was proved to be acceptable and then used for efficiently coding more particular conditions of the FIDIC 1999 Silver Book. The results indicate that good particular conditions roughly follow the functional distribution of general conditions and no significant difference is found in functional distribution between a whole complete contract and its underlying general conditions. It also suggests minor revisions should be made to the Adaptation function, and particular conditions should concentrate more on explicitly illustrating both parties’ obligations and tasks prudently to mitigate potential risks. These findings can practically assist contract managers to analyze particular conditions and identify the potential risks.

A machine learning based classification models for plastic recycling using different wavelength range spectrums

Currently, the use and production of plastic are on the rising trend, leading to the increase in waste generation and consumption of raw materials, making it one of the most significant issues facing densely populated cities. As a result, plastic waste management is becoming one of society's primary concerns, as it has direct effects on the environment and sustainability of urban areas. However, the identification and classification of the different types of plastic remains a challenge, as current techniques still face limitations. Unfortunately, identification technologies cannot classify many types of plastic, and often a particular technology is used to classify certain types of plastic. In this study, we propose a classification methodology that analyses the use of different machine learning algorithms based on the infrared spectrum of polymers. This proposed classification methodology will be able to identify spectrums obtained from different wavelength ranges, including near-infrared spectroscopy (NIR) and mid-infrared spectroscopy (MWIR). Moreover, this study proposes a solution to the classification of black plastic, as well as common recycled plastics such as Polypropylene (PP), Polystyrene (PS), Polyethylene (PE), Polyvinyl chloride (PVC), and Polyethylene terephthalate (PET). Experiments were conducted using eleven machine learning classification algorithms, and a comparative analysis of their performance was presented. The results indicate that five out of eleven classifiers achieved over 95% on the four metrics analyzed (accuracy, precision, recall and f1-score), with the Multi-layer Perceptron (MLP) Classifier achieving 99.72% accuracy, 99.35% precision, 99.82% recall, and 99.58% f1-score.

Cross-spatiotemporal land-cover classification from VHR remote sensing images with deep learning based domain adaptation

Automatic land use/land cover (LULC) classification from very high resolution (VHR) remote sensing images can provide us with rapid, large-scale, and fine-grained understanding of the urbanization and ecosystem processes on the Earth’s surface. Although the burgeoning use of deep learning technology has boosted land cover classification from VHR images, one of the key challenges has not been investigated in depth, i.e., the current data-driven deep learning models are heavily reliant on the high similarity between the distributions of the labeled training data, i.e., the source data, and the unlabeled data, i.e., the target data. However, in practice, this condition is rarely met, and preparing labels manually every time for the target data is unrealistic. In this paper, we comprehensively evaluate the domain adaptation methods for modern deep learning based classification models. Domain adaptation is aimed at narrowing the domain shift between the labeled source data and the unlabeled target data, providing a practical way for a deep learning based model to fully utilize historical training data and get rid of the need for continual manual work. Furthermore, we propose a novel two-stage Domain Adaptation method for Cross-Spatio-Temporal classification called the DACST method with the inputs of labeled source data and the unlabeled target data. It consists of an image-level adaptation stage that aligns the appearance of the source and target data and produces the target-stylized source images, and a feature-level adaptation stage that further narrows the domain shift in the deep feature space. DACST significantly improves the spatiotemporal transferability of the classification model, which is embedded in the feature-level adaptation stage, to output a satisfactory classification map. In this study, we also conducted a comprehensive performance evaluation of the conventional and deep learning based image-level and feature-level domain adaptation methods for VHR LULC classification. Both binary and multi-class classification was conducted in cross-temporal and cross-spatiotemporal scenes in five large-scale datasets from around the world. The very high performance and the best robustness of the proposed method suggests that a new baseline of cross-domain VHR land cover classification in the deep learning age is being witnessed. The experiments also indicate that both the conventional and deep learning based image-level domain adaptation methods function in various situations, but almost all the feature-level methods are highly unstable on different data.

An Intelligent system for the categorization of question time official documents of the Italian Chamber of Deputies

ABSTRACT In this work, we present an intelligent system for the automatic categorization of political documents, specifically the documents containing the parliamentary questions collected during the weekly Question Times at the Chamber of Deputies of the Italian Republic. The proposed intelligent system leverages text classification models to perform the document categorization. The system is aimed at supporting and facilitating the research activities of political science scholars, who deal with comparative and longitudinal analysis of thousands of documents. To select the best classification models for our specific task, several classical machine learning and deep learning-based text classification models have been experimentally compared.

Mining social media text for disaster resource management using a feature selection based on forest optimization

Resource management is an essential task that needs to be performed by the government or any disaster management agency during natural disasters. During these critical circumstances, people mostly depend upon a social media platform to share and collect information about the situation of the affected localities. The huge volume of real-time data can be useful in disaster assessment, response, and relief activities. We have presented a system which analyzes tweets during natural disasters and categorizes them according to the availability or need for general or medical resources along with their location information (if any) mentioned in the tweets. Several statistical classifiers are applied to show their usefulness for a better solution. Optimal feature representation is the heart of any machine learning based classification model. Here, we have applied a forest optimization-based wrapper feature selection algorithm to improve the classification accuracy. FIRE, SMERP, and CrisisLex dataset are used to evaluate our system and its effectiveness is demonstrated for smooth management of the resources. From the experimentation, it is found that forest optimization algorithm (FOA) wrapped multinomial naive bayes classifier gives an accuracy of 91.41 percent and f-measure of 88.33 percent on the FIRE dataset. The execution time of the model is quite less which will be very helpful for this challenging task.

BACS: blockchain and AutoML-based technology for efficient credit scoring classification.

Credit evaluation is of high scientific significance and practical use, especially in today’s plight of the world suffering from the COVID-19 epidemic. However, due to the difficulties inherent in credit scoring model building which involves a large number of data mining steps and requires a lot of time to process the data and build the model, efficient and accurate credit scoring methods are are urgently required. Aiming to solve this problem, we propose BACS, an blockchain and automated machine learning based classification model using credit dataset so that the credit modelling processes are performed in the pipeline in an automated manner to eventually obtain the classification results of credit scoring. BACS scheme consists of credit data storage to blockchain, feature extraction, feature selection, modelling algorithm and hyperparameter optimization, and model evaluation. Firstly, we propose a mechanism for credit data management and storage using blockchain to ensure that the entire credit scoring system is traceable and that the information of each scoring candidate is securely, efficiently and tamper-proofly stored on the blockchain nodes. Next, we design a pipeline using a random forest model to effectively integrate the key steps of credit data feature extraction, feature selection, credit model construction, and model evaluation. The experimental results demonstrate that our proposed automated machine learning-based credit scoring classification scheme BACS can assess the credit condition efficiently and accurately.

Histopathologic and Molecular Biomarkers of PD-1/PD-L1 Inhibitor Treatment Response among Patients with Microsatellite Instability‒High Colon Cancer

PurposeRecent clinical trials have reported response rates < 50% among patients treated with programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitors for microsatellite instability–high (MSI-H) colorectal cancer (CRC), and factors predicting treatment response have not been fully identified. This study aimed to identify potential biomarkers of PD-1/PD-L1 inhibitor treatment response among patients with MSI-H CRC.Materials and MethodsMSI-H CRC patients enrolled in three clinical trials of PD-1/PD-L1 blockade at Asan Medical Center (Seoul, Republic of Korea) were screened and classified into two groups according to treatment response. Their histopathologic features and expression of 730 immune-related genes from the NanoString platform were evaluated, and a machine learning–based classification model was built to predict treatment response among MSI-H CRCs patients.ResultsA total of 27 patients (15 responders, 12 non-responders) were included. A high degree of lymphocytic/neutrophilic infiltration and an expansile tumor border were associated with treatment response and prolonged progression-free survival (PFS), while mucinous/signet-ring cell carcinoma was associated with a lack of treatment response and short PFS. Gene expression profiles revealed that the interferon-γ response pathway was enriched in the responder group. Of the top eight differentially expressed immune-related genes, PRAME had the highest fold change in the responder group. Higher expression of PRAME was independently associated with better PFS along with histologic subtypes in the multivariate analysis. The classification model using these genes showed good performance for predicting treatment response.ConclusionWe identified histologic and immune-related gene expression characteristics associated with treatment response in MSI-H CRC, which may contribute to optimal patient stratification.

Design of robust deep learning-based object detection and classification model for autonomous driving applications

Design of robust deep learning-based object detection and classification model for autonomous driving applications

MANAGING IRRIGATION NEEDS BASED ON SMART DECISIONS USING MACHINE LEARNING

Optimized utilization of water for agriculture is a big challenge in today’s world. Internet of Things (IoT) based solutions along with machine learning techniques help in achieving effective utilization of waters in farming landspace. This paper presents sensor-based acquisition of soil moisture, temperature and humidity from the farm. Data are then stored in the server and clustered into two groups. Next machine learning based classification models like Naïve Bayes (NB), K-Nearest Neighbor (K-NN) and Support Vector Machines (SVM) are applied to decide irrigation need. The performance measures of the classification models show that K-NN classifier performs better than the other two classification models considered in this study.

A Deep Learning Based Classification Model for the Detection of Brain Tumor using MRI

The diagnosis of a brain tumor requires high accuracy, as even small errors in judgment can lead to critical problems. For this reason, brain tumor segmentation is an important challenge for medical purposes. The wrong classification can lead to worse consequences. Therefore, these must be properly divided into many classes or levels, and this is where multiclass classification comes into play. The latest development of image classification technology has made great progress, and the most popular and better method is considered to be the best in this area is CNN, so this paper uses CNN for the brain tumor classification problem. The proposed model successfully classifies brain images into two distinct categories, namely the absence of tumors indicating that a given brain MRI is free of tumors or the Brain contains Tumor. This model produces an accuracy based on the results of a study that was conducted on a group of volunteers.