Incorporating Machine Learning Research Articles

Monitoring of Water Quality using Machine Learning - A Review

Abstract: The incorporation of machine learning (ML) can improve the monitoring of water quality. It improves the sustainability and security of water resources by enabling forecasting and early detection of possible water contamination. A real-time system that collects information on numerous water parameters, including pH level, dissolved oxygen, turbidity, temperature, and conductivity, is made possible by the combination of IoT technology with ML algorithms. The system models complex links between water quality parameters and looks for variations from typical patterns using a mix of supervised and unsupervised learning and anomaly detection methods. Advantage of this method include cost-effectiveness, scalability, remote accessible, and real-time monitoring

Accommodating Machine Learning Algorithms in Professional Service Firms

Machine learning algorithms, as one form of artificial intelligence, are significant for professional work because they create the possibility for some predictions, interpretations and judgements that inform decision-making to be made by algorithms. However, little is known about whether it is possible to transform professional work to incorporate machine learning while also addressing negative responses from professionals whose work is changed by inscrutable algorithms. Through original empirical analysis of the effects of machine learning algorithms on the work of accountants and lawyers, this paper identifies the role of accommodating machine learning algorithms in professional service firms. Accommodating machine learning algorithms involves strategic responses that both justify adoption in the context of the possibilities and new contributions of machine learning algorithms and respond to the algorithms’ limitations and opaque and inscrutable nature. The analysis advances understanding of the processes that enable or inhibit the cooperative adoption of artificial intelligence in professional service firms and develops insights relevant when examining the long-term impacts of machine learning algorithms as they become ever more sophisticated.

A machine learning based approach for generating point sketch maps from qualitative directional information

People often use qualitative relations to describe locations or directional information, especially in written communication, such as ‘the restaurant is located at the southeast corner of the square’. However, when a large number of spatial entities are involved, qualitative relations alone are not intuitive enough for people to understand a spatial configuration. In fact, many applications, e.g. pertaining to sharing travel experiences, use sketch maps, i.e. maps focusing on the main features of an area whilst abstracting exact scale measurements, to help demonstrate abstract qualitative relations with more intuitive geometric points. Current approaches for generating point sketch maps from qualitative spatial relations require a high level of expertise, face inherent difficulties with efficiently processing large-scale data in bulk, and are vulnerable to inaccurate or conflicting information contained in qualitative data. To address these limitations, by incorporating machine learning techniques, we propose to translate the problem into an optimization problem of data reconstruction, enabling a novel end-to-end approach for generating point sketch maps from qualitative directional relations in bulk. Experiments on real-world datasets show that the proposed approach has very high accuracy and is robust even with a large portion of inaccurate or incomplete information.

From Connectivity to Controllability: Unraveling the Brain Biomarkers of Major Depressive Disorder

Major Depressive Disorder (MDD) is a significant neurological condition associated with aberrations in brain functional networks. Traditional studies have predominantly analyzed these from a network topology perspective. However, given the brain’s dynamic and complex nature, exploring its mechanisms from a network control standpoint provides a fresh and insightful framework. This research investigates the integration of network controllability and machine learning to pinpoint essential biomarkers for MDD using functional magnetic resonance imaging (fMRI) data. By employing network controllability methods, we identify crucial brain regions that are instrumental in facilitating transitions between brain states. These regions demonstrate the brain’s ability to navigate various functional states, emphasizing the utility of network controllability metrics as potential biomarkers. Furthermore, these metrics elucidate the complex dynamics of MDD and support the development of precision medicine strategies that incorporate machine learning to improve the precision of diagnostics and the efficacy of treatments. This study underscores the value of merging machine learning with network neuroscience to craft personalized interventions that align with the unique pathological profiles of individuals, ultimately enhancing the management and treatment of MDD.

Advanced Traffic Clearance System for Emergency Vehicles

This paper elaborates development of Traffic Clearance System using camera and mobile Net. In urban areas, navigating through traffic congestion poses a significant challenge for emergency vehicles, often leading to delays that can impact response times and potentially endanger lives. This paper presents an innovative traffic control system leveraging camera technology to prioritize the passage of emergency vehicles. The proposed system utilizes a network of cameras strategically placed at intersections to detect emergency vehicle sirens and flashing lights. Upon detection, the system dynamically adjusts traffic signal timing to create a green corridor, facilitating the swift and safe passage of emergency vehicles. Additionally, the system incorporates machine learning algorithms to optimize traffic flow, minimize congestion, and ensure efficient utilization of road space. Through simulation studies and real-world testing, the effectiveness and reliability of the proposed system are evaluated, demonstrating its potential to enhance emergency response capabilities and improve overall traffic management in urban environments.

Prospects for the computational humanization of antibodies and nanobodies

To be viable therapeutics, antibodies must be tolerated by the human immune system. Rational approaches to reduce the risk of unwanted immunogenicity involve maximizing the ‘humanness’ of the candidate drug. However, despite the emergence of new discovery technologies, many of which start from entirely human gene fragments, most antibody therapeutics continue to be derived from non-human sources with concomitant humanization to increase their human compatibility. Early experimental humanization strategies that focus on CDR loop grafting onto human frameworks have been critical to the dominance of this discovery route but do not consider the context of each antibody sequence, impacting their success rate. Other challenges include the simultaneous optimization of other drug-like properties alongside humanness and the humanization of fundamentally non-human modalities such as nanobodies. Significant efforts have been made to develop in silico methodologies able to address these issues, most recently incorporating machine learning techniques. Here, we outline these recent advancements in antibody and nanobody humanization, focusing on computational strategies that make use of the increasing volume of sequence and structural data available and the validation of these tools. We highlight that structural distinctions between antibodies and nanobodies make the application of antibody-focused in silico tools to nanobody humanization non-trivial. Furthermore, we discuss the effects of humanizing mutations on other essential drug-like properties such as binding affinity and developability, and methods that aim to tackle this multi-parameter optimization problem.

Intelligent Environment using Extended Reality and Machine Learning to Improve Healthcare

Intelligent Environments (IEs) represent a paradigm shift in the way we interact with our surroundings, leveraging Ambient Intelligence (AmI) to create responsive and adaptive spaces that seamlessly integrate technology into the built environment. By harnessing sensors, actuators, and artificial intelligence, IEs have the potential to enhance human experience, improve efficiency, and promote sustainability across various domains, including smart homes, workplaces, cities, and healthcare facilities. This paper provides an overview of IE concepts, discusses key technologies and applications, explores design considerations, and outlines future directions for research and development in this dynamic field. This paper focuses on the study of the potential use of Extended Reality technologies in order to promote Green Health as well as their use in Smart Hospitals, one of the variants of Smart Environments, incorporating Machine Learning and Computer Vision, as a tool to support and improve healthcare, both from the point of view of the health professional and the patient, through a literature review and analysis of the current situation. Resulting in the elaboration of a conceptual model with the suggestion of technologies that can be used to achieve this scenario selected for their potential, and then the development of prototypes of parts of the conceptual model for Extended Reality Headsets as concept validation.

Predicting metabolic modules in incomplete bacterial genomes with MetaPathPredict.

The reconstruction of complete microbial metabolic pathways using 'omics data from environmental samples remains challenging. Computational pipelines for pathway reconstruction that utilize machine learning methods to predict the presence or absence of KEGG modules in incomplete genomes are lacking. Here, we present MetaPathPredict, a software tool that incorporates machine learning models to predict the presence of complete KEGG modules within bacterial genomic datasets. Using gene annotation data and information from the KEGG module database, MetaPathPredict employs deep learning models to predict the presence of KEGG modules in a genome. MetaPathPredict can be used as a command line tool or as a Python module, and both options are designed to be run locally or on a compute cluster. Benchmarks show that MetaPathPredict makes robust predictions of KEGG module presence within highly incomplete genomes.

Comparison of traditional copromicroscopy with image analysis devices for detection of gastrointestinal nematode infection in sheep

Sustainable parasite control practices are necessary to combat the negative effects of gastrointestinal nematodes on animal health and production while reducing the selection pressure for anthelmintic resistance. Parasite diagnostic tests can inform treatment decisions, the timing and effectiveness of treatment and enable livestock breeding programmes. In recent years new diagnostic methods have been developed, some incorporating machine learning (ML), to facilitate the detection and enumeration of parasite eggs. It is important to understand the technical characteristics and performance of such new methods compared to long standing and commonly utilised methods before they are widely implemented. The aim of the present study was to trial three new diagnostic tools relying on image analysis (FECPAKG2, Micron and OvaCyte) and to compare them to traditional manual devices (McMaster and Mini-FLOTAC). Faecal samples were obtained from 41 lambs naturally infected with gastrointestinal nematodes. Samples were mixed and separated into 2 aliquots for examination by each of the 5 methods: McMaster, Mini-FLOTAC, FECPAKG2, Micron and OvaCyte. The techniques were performed according to their respective standard protocols and results were collected by trained staff (McMaster and Mini-FLOTAC) or by the device (FECPAKG2, Micron and OvaCyte). Regarding strongyle worm egg count, McMaster values varied from 0 to 9,000 eggs per gram (EPG). When comparing replicate aliquots, both the Mini-FLOTAC and Micron methods displayed similar repeatability to McMaster. However, we found FECPAKG2 and OvaCyte significantly less precise than McMaster. When comparing parasite egg enumeration, significant positive linear correlations were established between McMaster and all other methods. No difference was observed in EPG between McMaster and Mini-FLOTAC or FECPAKG2; however, Micron and OvaCyte returned significantly higher and lower EPG, respectively, compared to McMaster. The number of eggs ascribed to other parasite species was not sufficient for performing a robust statistical comparison between all methods. However, it was noted that FECPAKG2 generally did not detect Strongyloides papillosus eggs, despite these being detected by other methods. In addition, Moniezia spp and Trichuris spp eggs were detected by OvaCyte and Mini-FLOTAC, respectively, but not by other methods. The observed variation between traditional and new methods for parasite diagnostics highlights the need for continued training and enhancing of ML models used and the importance of developing clear guidelines for validation of newly developed methods.

A generalized detection framework for covert timing channels based on perceptual hashing

AbstractNetwork covert channels use network resources to transmit data covertly, and their existence will seriously threaten network security. Therefore, an effective method is needed to prevent and detect them. Current network covert timing channel detection methods often incorporate machine learning methods in order to achieve generalized detection, but they consume a large amount of computational resources. In this paper, we propose a generalized detection framework for covert channels based on perceptual hashing without relying on machine learning methods. And we propose a one‐dimensional data feature descriptor for feature extraction of perceptual hash for the data characteristics of covert timing channels. We first generate the hash sequence of the corresponding channel to get the average hash, which is used for comparison in the test phase. The experimental results show that the feature descriptor can capture the feature differences of one‐dimensional data well. When compared to machine learning methods, this perceptual hashing algorithms enable faster traffic detection. Meanwhile, our method is able to detect the effectiveness with the smallest coverage window compared with the latest solutions. Moreover, it exhibits robustness in jitter network environment.

Optimization of English Classroom Interaction Models Incorporating Machine Learning

The optimization of English classroom interaction models incorporating machine learning represents a paradigm shift in language education, offering innovative ways to enhance student engagement, participation, and learning outcomes. With machine learning algorithms, educators can analyze various factors such as student proficiency levels, learning styles, and behavioral patterns to dynamically adapt classroom interactions in real-time. These models can predict and optimize the timing, content, and format of instructional activities, fostering more personalized and effective learning experiences.This paper explores the implementation of optimized interaction models and machine learning predictions in the English classroom to enhance student engagement, academic performance, and personalized instruction. Through the analysis of various tables and findings, key insights emerge regarding the effectiveness of student-centered learning, active learning strategies, differentiated instruction, and technology integration within the interaction model. Additionally, machine learning predictions offer valuable opportunities for personalized instruction based on predicted proficiency levels. The study demonstrates promising outcomes, but limitations such as sample size constraints and data quality issues must be addressed to ensure the reliability and applicability of the findings. Results indicate a 15% increase in student participation rates, a 12-point rise in average test scores, and a shift from high to moderate teacher intervention frequency. Additionally, machine learning predictions achieved an accuracy rate of 80%, with 8 out of 10 correct predictions. Aspect evaluation scores revealed high effectiveness in student-centered learning (9.2), formative assessment (9.3), and teacher facilitation (9.0). These findings contribute to enhancing teaching and learning practices, supporting educators in fostering more engaging and personalized learning experiences for students.

RadPhysBio: A Radiobiological Database for the Prediction of Cell Survival upon Exposure to Ionizing Radiation.

Based on the need for radiobiological databases, in this work, we mined experimental ionizing radiation data of human cells treated with X-rays, γ-rays, carbon ions, protons and α-particles, by manually searching the relevant literature in PubMed from 1980 until 2024. In order to calculate normal and tumor cell survival α and β coefficients of the linear quadratic (LQ) established model, as well as the initial values of the double-strand breaks (DSBs) in DNA, we used WebPlotDigitizer and Python programming language. We also produced complex DNA damage results through the fast Monte Carlo code MCDS in order to complete any missing data. The calculated α/β values are in good agreement with those valued reported in the literature, where α shows a relatively good association with linear energy transfer (LET), but not β. In general, a positive correlation between DSBs and LET was observed as far as the experimental values are concerned. Furthermore, we developed a biophysical prediction model by using machine learning, which showed a good performance for α, while it underscored LET as the most important feature for its prediction. In this study, we designed and developed the novel radiobiological 'RadPhysBio' database for the prediction of irradiated cell survival (α and β coefficients of the LQ model). The incorporation of machine learning and repair models increases the applicability of our results and the spectrum of potential users.

Monitoring the Wear Trends in Wind Turbines by Tracking Fourier Vibration Spectra and Density Based Support Vector Machines

To make wind power more competitive, it is necessary to reduce turbine downtime and reduce costs associated with wind turbine operation and maintenance (O&M). Incorporating machine learning in the development of condition-based predictive maintenance methodologies for wind turbines can enhance their efficiency and reliability. This paper presents a monitoring method that utilizes Density Based Support Vector Machines (DBSVM) and the evolutionary Fourier spectra of vibrations. This method allows for the smart monitoring of the function evolution of the turbine. A complex optimal function (FO) for 5-degree order has been developed that will be the boundary function of the DBSVM to be timely determined from the Fourier spectrum through the magnitude–frequency and place of the failure occurring in the wind turbine drivetrains. The trend of the failure was constructed with the maximal values of the optimal frequency function for both yesthe cases of the upwind and downwind parts of the gearbox.

Optimizing green supply chain circular economy in smart cities with integrated machine learning technology

This paper explores methodologies to enhance the integration of a green supply chain circular economy within smart cities by incorporating machine learning technology. To refine the precision and effectiveness of the prediction model, the gravitational algorithm is introduced to optimize parameter selection in the support vector machine model. A nationwide prediction model for green supply chain economic development efficiency is meticulously constructed by leveraging public economic, environmental, and demographic data. A comprehensive empirical analysis follows, revealing a noteworthy reduction in mean squared error and root mean squared error with increasing iterations, reaching a minimum of 0.007 and 0.103, respectively—figures that are the lowest among all considered machine learning models. Moreover, the mean absolute percentage error value is remarkably low at 0.0923. The data illustrate a gradual decline in average prediction error and standard deviation throughout the model optimization process, indicative of both model convergence and heightened prediction accuracy. These results underscore the significant potential of machine learning technology in optimizing supply chain and circular economy management. The paper provides valuable insights for decision-makers and researchers navigating the landscape of sustainable development.

Empowering Structural Integrity: YOLO-Based Crack Detection and ML-Driven Concrete Strength Prediction of Critical Infrastructure Caused due to Mining Operation.

In this paper, YOLO algorithm is used for detecting crack by incorporating machine learning approach utilized for predicting concrete strength and possible crack in critical infrastructure caused during mining operation. Over time, those fissures become more noticeable until they finally collapse. Big data is accumulating huge amount of data by utilizing remote sensing, earth penetrating RADAR, LiDAR, GPS, and satellite image data for large-scale operations. The performance of the proposed approach is compared with other state-of-the-art algorithms, such as, YOLO V3 and YOLO V8. To investigate the performance, four performances metric like mean absolute error (MAE), root mean square error (RMSE), relative standard error (RSE) and root relative squared error (RRSE) has considered. It is found that our approach is giving superior results in term of the compared performance metrics. Moreover, non-parametric statistical testing, such as, signed test, Wilcoxon Rank test and Friedman's test has been conducted to verify the dominance of proposed approach over others.

Droplet-based microfluidic platform for detecting agonistic peptides that are self-secreted by yeast expressing a G-protein-coupled receptor

BackgroundSingle-cell droplet microfluidics is an important platform for high-throughput analyses and screening because it provides an independent and compartmentalized microenvironment for reaction or cultivation by coencapsulating individual cells with various molecules in monodisperse microdroplets. In combination with microbial biosensors, this technology becomes a potent tool for the screening of mutant strains. In this study, we demonstrated that a genetically engineered yeast strain that can fluorescently sense agonist ligands via the heterologous expression of a human G-protein-coupled receptor (GPCR) and concurrently secrete candidate peptides is highly compatible with single-cell droplet microfluidic technology for the high-throughput screening of new agonistically active peptides.ResultsThe water-in-oil microdroplets were generated using a flow-focusing microfluidic chip to encapsulate engineered yeast cells coexpressing a human GPCR [i.e., angiotensin II receptor type 1 (AGTR1)] and a secretory agonistic peptide [i.e., angiotensin II (Ang II)]. The single yeast cells cultured in the droplets were then observed under a microscope and analyzed using image processing incorporating machine learning techniques. The AGTR1-mediated signal transduction elicited by the self-secreted Ang II peptide was successfully detected via the expression of a fluorescent reporter in single-cell yeast droplet cultures. The system could also distinguish Ang II analog peptides with different agonistic activities. Notably, we further demonstrated that the microenvironment of the single-cell droplet culture enabled the detection of rarely existing positive (Ang II-secreting) yeast cells in the model mixed cell library, whereas the conventional batch-culture environment using a shake flask failed to do so. Thus, our approach provided compartmentalized microculture environments, which can prevent the diffusion, dilution, and cross-contamination of peptides secreted from individual single yeast cells for the easy identification of GPCR agonists.ConclusionsWe established a droplet-based microfluidic platform that integrated an engineered yeast biosensor strain that concurrently expressed GPCR and self-secreted the agonistic peptides. This offers individually isolated microenvironments that allow the culture of single yeast cells secreting these peptides and gaging their signaling activities, for the high-throughput screening of agonistic peptides. Our platform base on yeast GPCR biosensors and droplet microfluidics will be widely applicable to metabolic engineering, environmental engineering, and drug discovery.

Challenges and Outcomes of Combining Machine Learning with Software-Defined Networking for Network Security and management Purpose: A Review

Current research in data dissemination in Vehicular Ad Hoc Networks (VANETs) has examined different approaches and frameworks to enhance the effectiveness and dependability of information sharing between vehicles on the road. The integration of Machine Learning (ML) with Software-Defined Networking (SDN) has fundamentally transformed the field of network administration and security. This paper specifically addresses the challenges faced by traditional network architectures in effectively handling the increasing amount of data and complex applications. Software-Defined Networking (SDN), a cutting-edge framework, separates the control of network operations from the actual forwarding of data, offering a versatile and cost-effective solution. The combination of Software-Defined Networking (SDN) and Machine Learning (ML) allows for the extraction of valuable information from network data, leading to enhanced network management and the facilitation of predictive analytics. The aim of this study is to examine the feasibility and challenges of incorporating machine learning into software-defined networking (SDN), focusing particularly on practical applications. Integrating Machine Learning (ML) into Software-Defined Networking (SDN) presents challenges, including the requirement for robust algorithms to detect patterns and ensure security. It is crucial to carry out the tasks of developing and implementing machine learning models for real-time predictions and ensuring the robustness of the system. Research is essential to strike a balance between the transformative abilities of ML-SDN and the practical network environments. This helps to improve the resilience, security, and adaptability of network infrastructures in the digital age.

Efficient Disease Prediction Framework to Suggest Early Treatment Decisions in Healthcare

This study proposes a novel framework for disease prediction intended to aid healthcare providers in making early treatment decisions. In order to efficiently identify early disease indicators and provide prompt suggestions for treatment interventions, the framework incorporates machine learning algorithms, clinical data analysis, and predictive modeling techniques (Smith, Johnson, & Davis, 2023). For the purpose of creating the framework, a comprehensive dataset made up of clinical records, demographic data, and test results from a large patient cohort was gathered. The dataset was analyzed using a variety of machine learning methods, and predictive models for various diseases were created. As evaluation criteria, accuracy, precision, recall, and F1-score were used to gauge how well these models performed. The experimental findings show that the suggested framework has a high degree of disease prediction accuracy. Additionally, by recommending early treatment selections. The approach offers the potential to improve patient outcomes and lower healthcare expenditures based on anticipated illness outcomes. The study benefits healthcare professionals, policymakers, and academics looking to use data-driven strategies for better disease management and patient care by providing an effective method for disease prediction and early treatment decisions.

Question Paper Generator

The automatic question paper generator (AQPG) is a software tool designed to streamline and automate the process of generating examination question papers. It leverages algorithms and databases to create balanced, fair, and diverse sets of questions based on predefined parameters such as topic, difficulty level, and question type. By incorporating machine learning techniques, natural language processing, and data mining, AQPG can analyze vast amounts of educational content to ensure the integrity and quality of the generated questions. This abstract provides an overview of the AQPG system, highlighting its significance in reducing manual effort, enhancing efficiency, and maintaining the integrity of assessment practices in educational institutions.

Construction and evaluation of financial distress early warning model based on machine learning

As the economy of our nation has expanded, the harm resulting from the financial risks associated with listed corporations has intensified, significantly impeding the survival and expansion of businesses. We must create a reliable financial crisis early warning system to prevent financial risks from endangering the business. This article aims to analyse the development of financial early warning models that use machine learning techniques. The model employs the Archimedes optimization method (AOA) method to optimize the parameters of SVM and selects 20 financial risk assessment indices as the input to anticipate the financial crises. The results suggest that the proposed model surpasses current models in terms of both prediction accuracy and resilience. They also highlight the ensemble model's greater predictive capacity when compared to individual algorithms, highlighting its efficacy in spotting early warning signs of financial instability in a variety of market scenarios. Financial organisations and governments may improve their risk management procedures and prevent future crises by incorporating machine learning techniques into the early warning system.