Multiband THz MIMO antenna with regression machine learning techniques for isolation prediction in IoT applications

Regression supervised model techniques THz MIMO antenna for 6G wireless communication and IoT application with isolation prediction

Globally, wind energy if properly harnessed, could serve as a source of energy generation in Africa. This study compared the performance of two Machine Learning (ML) algorithms (Linear regression and Random Forest) in predicting wind speed in five major cities in Africa (Yaoundé, Pretoria, Nairobi, Cairo and Abuja). Wind data were collected between January 1, 2000, and December 31, 2022, using the Solar Radiation Data Archive. The data preprocessing was carried out with 80% of the data used for training and 20% for validation. The performance of these ML algorithms was evaluated using Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and coefficient of determination (R2). The result shows that Nairobi (3.814795 m/s) closely followed by Cairo (3.606453 m/s) has the highest mean wind speed while Yaoundé (1.090512 m/s) has the lowest. Based on the performance metrics used, the two Machine Learning algorithms were competitive. Still, the Linear Regression (LR) algorithm outperformed the Random Forest Algorithm in predicting wind speed in all the selected major African cities. In Yaoundé (RMSE = 0.3892, MAE= 0.3001, MAPE =0.5030), Pretoria (RMSE=1.2339, MAE=0.9480, MAPE=0.7450) Nairobi (RMSE= 0.4223, MAE =0.6499, MAPE =0.1872), Nairobi (RMSE=0.6499, MAE=0.5171, MAPE =0.1872), Cairo (RMSE =1.0909, MAE =0.8544, MAPE =0.3541) and Abuja (RMSE = 0.70245, MAE =0.5441, MAPE= 0.4515) the Linear regression algorithms was found to outperformed Random Forest Regression. Therefore, the Linear regression algorithm is more reliable in predicting wind speed compared with the Random Forest regression.

Performance evaluation of machine learning techniques in predicting cumulative absolute velocity

This article uses machine learning techniques as fuzzy and neuro-fuzzy ANFISs, to develop and compare prediction models capable of relating pregnant women’s exposure to air pollutants, such as Nitrogen Dioxide and Particulate Matter, the mother’s age, and the number of prenatal consultations to the incidence of premature birth. In the current literature, studies can be found that relate prematurity to the exposure of pregnant women to NO2, O3, and PM10; to Toluene and benzene, mainly in the window 5 to 10 days before birth; and to PM10 in the week before birth. Both models used logistic regression to quantify the effects of pollutants as a result of premature birth. Datasets from Brazil—Departamento de Informatica do Sistema Único de Saúde (DATASUS) and Companhia Ambiental do Estado de São Paulo (CETESB)—were used, covering the period from 2016 to 2018 and comprising women living in the city of São José dos Campos (SP), Brazil. In order to evaluate and compare the different techniques used, evaluation metrics were calculated, such as correlation (r), coefficient of determination (R2), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Square Error (MSE), and Mean Absolute Error (MAE). These metrics are widely used in the literature due to their ability to evaluate the robustness and efficiency of prediction models. For the RMSE, MAPE, MSE, and MAE metrics, lower values indicate that prediction errors are smaller, demonstrating better model accuracy and confidence. In the case of (r) and R2, a positive and strong result indicates alignment and better performance between the real and predicted data. The neuro-fuzzy ANFIS model showed superior performance, with a correlation (r) of 0.59, R2 = 0.35, RMSE = 2.83, MAPE = 5.35%, MSE = 8.00, and MAE = 1.70, while the fuzzy model returned results of r = 0.20, R2 = 0.04, RMSE = 3.29, MSE = 10.81, MAPE = 6.67%, and MAE = 2.01. Therefore, the results from the ANFIS neuro-fuzzy system indicate greater prediction capacity and precision in relation to the fuzzy system. This superiority can be explained by integration with neural networks, allowing data learning and, consequently, more efficient modeling. In addition, the findings obtained in this study have potential for the formulation of public health policies aimed at reducing the number of premature births and promoting improvements in maternal and neonatal health.

Demand forecasting is a critical function in Enterprise Resource Planning (ERP) systems, directly influencing inventory management, production planning, and overall operational efficiency. Traditional statistical models often fall short in handling the complexity and variability of modern supply chains. This study investigates the application of Artificial Intelligence (AI), specifically Machine Learning (ML) algorithms, to enhance demand forecasting accuracy within ERP environments. I conduct a comparative analysis of four widely used ML models: Linear Regression, Random Forest, Support Vector Regression (SVR), and Long Short-Term Memory (LSTM) networks. Using real-world ERP datasets, each model is evaluated based on Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and computational performance. The results reveal that while Random Forest and LSTM models outperform others in terms of accuracy, their complexity and training time vary significantly. My findings highlight the trade-offs between model accuracy and computational efficiency, offering practical insights for ERP stakeholders. This study contributes to the growing field of AI-driven enterprise analytics and provides guidance on selecting appropriate ML techniques tailored to specific forecasting needs within ERP systems.

Air quality, both indoors and outdoors, is crucial for public health as it affects respiratory conditions and overall well-being. Machine learning (ML) techniques offer innovative solutions for monitoring and predicting air quality. However, choosing the right ML algorithms and evaluation metrics is essential for creating accurate air quality prediction models, as these choices directly impact the accuracy and reliability of the results. This study aims to explore and summarize the literature on the use of ML techniques in predicting and optimizing air quality, addressing the urgent issue of air pollution. The review analyzed papers from two electronic databases, namely Scopus and Science Direct. Information on ML techniques for predicting air quality and identifying the main sources of pollution was extracted from 26 studies. The study focuses on common ML techniques employed in air quality prediction, including classification, deep learning, regression, ensemble learning, and combinations of regression and deep learning. It also identifies the evaluation metrics used to assess the performance of these models, such as recall, root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2). By synthesizing existing knowledge, this study provides a comprehensive understanding of the metrics researchers use to measure the overall effectiveness of ML algorithms. It serves as a benchmark for future research and guides the selection of appropriate evaluation metrics in the field of air quality prediction.

This study investigates the quality of peak oxygen consumption (VO2peak) prediction based on cardiac and respiratory parameters calculated from warmup and submaximal stages of treadmill cardiopulmonary exercise test (CPET) using machine learning (ML) techniques and assesses the importance of respiratory parameters for the prediction outcome. The database consists of the following parameters: heart rate (HR), respiratory rate (RespRate), pulmonary ventilation (VE), oxygen consumption (VO2) and carbon dioxide production (VCO2) obtained from 369 treadmill CPETs. Combinations of features calculated based on the HR, VE and RespRate time-series from different stages of CPET were used to create 11 datasets for VO2peak prediction. Thirteen ML algorithms were employed, and model performances were evaluated using cross-validation with mean absolute percentage error (MAPE), R2 score, mean absolute error (MAE), and root mean squared error (RMSE) calculated after each iteration of the validation. The results demonstrated that incorporating respiratory-based features improves the prediction of VO2peak. The best results in terms of R2 score (0.47) and RMSE (5.78) were obtained for the dataset which included both cardiac- and respiratory-based features from CPET up to 85% of age-predicted HRmax, while the best results in terms of MAPE (10.5%) and MAE (4.63) were obtained for the dataset containing cardiorespiratory features from the last 30 seconds of warmup. The study showed the potential of using ML models based on cardiorespiratory features from submaximal tests for prediction of VO2peak and highlights the importance of the monitoring of respiratory signals, enabling to include respiratory parameters into the analysis. Presented approach offers a feasible alternative to direct VO2peak measurement, especially when specialized equipment is limited or unavailable.

The ever-growing number of interconnected devices and the emergence of high-bandwidth technologies such as holographic communications in mixed reality are driving an exponential growth in the demand for terabits-per-second (Tbps) data rates with sub-milliseconds latency. For this reason, the sixth-generation (6G) wireless communications are moving towards the terahertz (THz) frequency band because of its wider spectral bandwidth availability to enable higher channel capacity and data rates according to the Shannon-Hartley theorem. However, the efficiency of wireless THz signal transmission is reduced by the increased path loss and atmospheric attenuation due to water vapor absorption at THz frequencies, which significantly decreases the signal-to-noise ratio and the spectral efficiency of THz wireless communications. The realization of THz band communications is further impeded by the limited power output of existing compact THz sources due to the extremely low conversion efficiencies. Therefore, it is crucial to develop broadband THz integrated circuits with waveguiding and beamforming functionalities for supporting high data-rate on-chip communications and to provide sufficient gains to counteract the path loss and limited power of THz wireless transmissions. This thesis aims to improve the spectral efficiency of 6G wireless communications via the spatiotemporal shaping of THz waves using silicon photonics. Spatiotemporal light shaping is the basis of nearly all THz photonic devices, ranging from THz wavefront manipulation using metasurfaces to THz planar waveguiding using photonic crystal waveguides. Silicon photonics is a promising material platform for THz wave shaping due to the low-loss propagation of THz waves in high-resistivity silicon and the prospect of a monolithic integrated circuit due to the complementary metal-oxide semiconductor (CMOS) fabrication process compatibility. Silicon metasurfaces allow the wavefront modulation of incident THz radiation by exploiting the generalized Snell’s law of refraction and reflection to beamform highly directional THz waves in a dynamic multi-user communication environment for enhancing the spectral efficiency. Likewise, broadband silicon photonic crystal waveguides allow the in-plane and lossless transport of THz waves to support on-chip high data-rate THz communications that can also seamlessly integrate multiple THz photonic devices in a THz integrated circuit. In addition, silicon electron emitters and silicon gratings potentially allow the on-chip generation of highly intense and broadband THz radiation to counteract the significant THz attenuation. In this thesis, spatiotemporal wave shaping techniques using silicon photonics for enhancing the spectral efficiency of THz band communications is presented. First, a self-adaptive and deep reinforcement learning model for THz beamforming is proposed, which can beamform multiple highly directional THz radiation beam in real-time to serve a dynamically changing multi-user communication environment and improve the signal-to-noise ratio of THz wireless transmission while potentially reducing the communication latency. Second, a broadband THz topological silicon waveguide designed based on valley photonic crystals is proposed for the robust and lossless transport of broadband THz waves for the construction of THz integrated circuits with broadband channel capacity for enabling higher data-rate on-chip communication. Last, the versatility of free-electron radiation from silicon gratings is investigated for the generation of highly intense spatiotemporal wave packets with user-designed group velocities and tunable intensity profile, with the potential to realize integrated high-power THz sources. The deep learning beamforming model, the broadband topological waveguide designs, and the free-electron radiation intensity shaping techniques presented in this thesis extends beyond the THz regime and are scalable across the entire electromagnetic spectrum. The CMOS process compatibility of silicon metasurfaces, silicon planar waveguides, along with silicon electron emitters and silicon gratings offers the opportunity to realize a broadband THz photonic integrated circuit with waveguiding, wave generation and beamforming functionalities for enhancing the spectral efficiency of THz communications in 6G wireless networks.

Accurate prediction of Rate of Penetration (ROP) is critical for optimizing drilling efficiency and reducing costs in hydrocarbon exploration. Traditional ROP models often overlook formation temperature, despite its significant influence on rock mechanics and drilling fluid rheology, particularly in high-temperature sandstone reservoirs like those in the Niger Delta. This study employs three machine learning (ML) algorithms—Support Vector Regression (SVR), Random Forest (RF), and Artificial Neural Networks (ANN)—to evaluate the contribution of formation temperature to ROP prediction. A dataset of 1,200 drilling records from Niger Delta wells was used, incorporating parameters such as weight-on-bit (WOB), rotary speed (RPM), pump pressure, and formation temperature. Model performance was assessed using R², Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), with ablation studies to isolate temperature’s impact. Inclusion of formation temperature improved ROP prediction accuracy across all models. The ANN achieved the highest performance (R² = 89.3%, RMSE = 0.387, MAE = 0.141), followed by RF (R² = 90.5%, RMSE = 2.737, MAE = 0.900) and SVR (R² = 87.8%, RMSE = 0.553, MAE = 0.169). Temperature omission led to significant performance degradation (R² reductions of 7–13%). Sensitivity analysis ranked temperature among the top three influential features. Formation temperature is a critical but underutilized parameter in ROP modeling. ML techniques, particularly ANN, demonstrate superior capability in capturing nonlinear temperature-dependent effects, offering actionable insights for real-time drilling optimization in thermally complex formations. This study provides a framework for integrating thermal data into predictive models to enhance drilling efficiency in the Niger Delta and analogous basins.

The term Internet of Things (IoT) refers to a technological revolution that allows nearly all environmental devices to connect to the internet and share data to create services and applications that improve our standard of living. As IoT systems generate huge amounts of data, new IoT applications and facilities arise. Machine learning has made significant progress in several fields and is now being used to solve a variety of IoT network problems. Machine learning has recently become more popular as a means of improving IoT applications. Users can find deep analytics and develop efficient intelligent IoT applications using machine learning for IoT. This chapter focuses on offering an overview of machine learning techniques in the IoT platform. The chapter begins with an overview of the IoT, including its history, architecture, and its applications. Then, many machine learning algorithms, various types, and their applications, with examples, are discussed.

This study aimed to model and predict iron concentrations in groundwater within Yenagoa, Bayelsa State, Nigeria, using machine learning techniques. It focused on evaluating spatial variability and determining the most influential predictors to support groundwater quality management. A total of 50 groundwater samples were collected from spatially distributed boreholes across multiple towns in Yenagoa. Geolocation data and iron concentrations were recorded. Two supervised machine learning models Multiple Linear Regression (MLR) and Random Forest Regression (RFR) were implemented. One-hot encoding was applied to categorical town data, and models were evaluated using R², MAE, and Root Mean Square Error (RMSE) metrics. Feature importance was assessed to identify key predictors. A geospatial heatmap was developed using Inverse Distance Weighting (IDW) to visualize spatial trends. The MLR model slightly outperformed the RFR, achieving an Coefficient of Determination (R²) of 0.92, Mean Absolute Error (MAE) of 0.13 mg/L, and RMSE of 0.15 mg/L. Longitude and specific towns (notably Beta and Opolo) emerged as dominant predictors, confirming spatial clustering of high iron concentrations in the eastern region of the study area. Cross-validation confirmed the models’ robustness. The findings support the use of machine learning (ML) techniques for cost-effective water quality prediction and spatial monitoring. This study introduces a hybrid geo-categorical modeling approach, integrating both spatial coordinates and administrative town identifiers into ML frameworks. It demonstrates the feasibility of lightweight, interpretable models like MLR for real-time deployment in low-resource settings, offering a replicable solution for groundwater quality assessment in data-scarce regions. Future research should expand datasets and explore additional hydrogeological variables to enhance model robustness.

This paper analyses water pollution data focusing on Biochemical Oxygen Demand (BOD), Chemical Oxygen Demand (COD), pH, and turbidity. Each parameter has specific acceptable ranges to ensure the water is safe and suitable for consumption. BOD, COD, pH and turbidity levels using a Comma separated value (CSV) dataset. Initially, the data is pre-processed by cleaning it to handle missing values and outliers, ensuring consistency and accuracy. The dataset was then divided into training and testing subsets. For modelling, K-Nearest Neighbor(KNN) algorithm of machine learning will be employed. The model was trained on the processed data and evaluated using metrics such as the mean squared error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) to assess pollutants label of BOD, COD, pH and turbidity of Damodar at Asansol, Raniganj, Andal and Durgapur with some predefined threshold values based on past experience. Finally, the above analysis will be interpreted to understand the patterns and factors influencing pollution levels of BOD, COD, pH and turbidity and the results are found to be in very good agreement.

A machine-learning ensemble model for predicting energy consumption in smart homes

The road infrastructure serves as a linchpin for both developed and developing nations, requiring efficient management strategies to uphold this vital mode of transportation. Long-term pavement management agencies are tasked with continuously monitoring and maintaining road pavements, relying on databases to secure critical data. Pavement condition index (PCI) is a pivotal metric for assessing pavement health, quantifying distress severity and structural integrity. Manual PCI assessment is laborious and error-prone, especially for extensive road networks. Recent technological advancements, particularly in Artificial Intelligence (AI), have revolutionised pavement management through predictive modelling. In this study, supervised machine learning algorithms are leveraged, including Random Forest Regressor, M5 Prime Model Tree Algorithm, Decision Tree, Multiple Linear Regression (MLR), Gradient Boosting Regressor, K-Nearest Neighbours (KNN) and Support Vector Machine Regressor (SVMR) to predict PCI, optimise rehabilitation strategies and prioritise maintenance efforts. The dataset utilised in this study comprises 237 roads with 9576 sample units covering various distress types, extents and severity levels crucial for PCI determination. The data undergoes rigorous training, testing and validation across all the selected algorithms. Model performance is evaluated using key metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Coefficient of Determination (R 2) and Mean Absolute Percentage Error (MAPE). Additionally, the Kruskal–Wallis test and t-test are employed to assess the statistical significance of the results. Our findings highlight the superior performance of the Random Forest Regressor and M5P models, with R 2 values of 0.96 and 0.95, respectively. These models outshine others, showcasing their effectiveness in PCI prediction. Conversely, the results of Gradient Boosting Regressor, KNN and SVMR models exhibit inconsistency and unreliability in PCI prediction. This study underscores the transformative potential of machine learning algorithms in pavement management, emphasising strategies to mitigate the limitations of manual methods in PCI determination.

Estimation of solubility of acid gases in ionic liquids using different machine learning methods