Time series extrinsic regression

The objective of this document is to identify the most accurate machine learning (ML) models for predicting the value of a real estate property, based on a systematic literature review (SLR). It was conducted on research published between 2022 and 2023 that analyzed the accuracy of ML models in real estate valuation. Information was extracted regarding the ML models used, the databases employed, and the models highlighted for their accuracy. A variety of ML models used in real estate valuation were identified, including Random Forest (RF), XGBoost, Gradient Boosting Machine (GBM), Linear Regression (LR), and Lasso Regression. RF and LR models stood out as the most accurate in the analyzed research, showing that the accuracy of ML models varies depending on the database, property characteristics, and valuation context. It was concluded that ML models like RF and LR are promising tools for improving the accuracy of real estate valuation. Additionally, the choice of the best ML model depends on factors such as the database, property characteristics, and specific valuation objectives. Further research is needed on the application of ML in real estate valuation, considering aspects such as model interpretability and the real estate market context.

This paper investigates the importance of explainability versus accuracy of machine learning (ML) models. We propose that greater task uncertainty makes people want to interact more with the ML model, which increases the importance of explainability relative to accuracy. We focus on the use of ML models for product cost estimation during new product development. The paper provides mixed-methods evidence on the trade-off between explainability and accuracy of ML models. Specifically, we find support for an inverse relationship between explainability and accuracy from the perspective of cost experts. We also find that the accurate but complex and less explainable ML model of gradient boosted regression (GBR) was preferred in only a few situations; mostly, the more basic, better explainable models of multiple linear regression (MLR) and case-based reasoning (CBR) were preferred, although these were less accurate. This suggests that lack of explainability can indeed be a major limitation for the application of ML models. Furthermore, we investigate specific characteristics that could increase task uncertainty and the importance of explainability in our context: project unpredictability, product cost granularity, predecessor product availability, target cost gap, and product development phase.

It is necessary to develop general and efficient models for the representation of vapor-liquid equilibria (VLE) for the binary mixtures containing R1234ze(E) which will be new alternatives to conventional refrigerants. This work investigates the applicability of four machine learning models, including K-nearest neighbor (KNN), support vector regression (SVR), random forests (RF), and multi-layer perceptron (MLP), in representing VLE for 10 binary mixtures. The accuracy, stability, computational complexity and extrapolation ability of these machine learning models are analyzed and compared with two thermodynamic models including Soave–Redlich–Kwong (SRK) and Peng–Robinson (PR). The results show the KNN, RF and MLP models are not suitable to accurately represent the experimental pressure and vapor-phase mole fraction. The SVR model is the most accurate machine learning model in representing the experimental pressure with a mean absolute relative error of 0.71%, basically as accurate as the PR and SVR models. Meanwhile, the SVR model describes the experimental vapor-phase mole fraction more accurately with a mean absolute error of 0.0049 compared to the PR and SVR models. In addition, the SVR model has a low running-time cost and a certain extent of extrapolation ability.

Thermodynamic models are usually employed to predict formation condition of hydrates. However, these thermodynamic models usually require a large amount of calculations to approach phase equilibrium. Additionally, parameters included in the thermodynamic model need to be calibrated based on the experimental data, which leads to high uncertainties in the predicted results. With the rapid development of artificial intelligence (AI), machine learning as one of sub-discipline has been developed and been widely applied in various research area. In this work, machine learning was innovatively employed to predict the formation condition of natural gas hydrates to overcome the high computation cost and low accuracy. Three data-driven models, Random Forest (RF), Naive Bayes (NB), Support Vector Regression (SVR) were tentatively used to determine the formation condition of hydrate formed by pure and mixed gases. Experimental data reported in previous work were taken to train and test the machine learning models. As a representative thermodynamic model the Chen–Guo (C-G) model was used to analyze the computational efficiency and accuracy of machine learning models. The comparison of results predicted by C-G model and machine learning models with the experimental data indicated that the RF model performed better than the NB and SVR models on both computation speed and accuracy. According to the experimental data, the average AADP calculated by the C-G model is 7.62 times that calculated by the RF model. Meanwhile, the average time costed by the C-G model is 75.65 times that by the RF model. Compared with the other two machine learning models, the RF model is expected to be used in predicting the formation condition of natural gas hydrate under field conditions.

Development of compressive strength prediction platform for concrete materials based on machine learning techniques

The prediction of wind energy generation is important to enhance the performance and dependability of renewable energy systems due to the rising demand for wind-generated electricity and advancements in wind energy technology competitiveness. This study leverages advanced machine learning (ML) and some other statistical and deep learning based time series forecasting models to enhance the accuracy of wind energy predictions. This comprehensive analysis includes nine ML models—Linear Regression, Random Forests (RF), Gradient Boosting Machines (GBM), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), AdaBoost, XGBoost, Support Vector Regression (SVR), and Neural Networks—as well as Four time-series forecasting models—ARIMA, Temporal Convolutional Networks (TCNs), Long Short-Term Memory (LSTM) networks and GRU. Each ML model underwent rigorous cross-validation to ensure optimal performance. The assessment criteria utilized here comprised the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the R² Score. It was found that among the nine ML models, Random Forests, GBM and KNN consistently provided superior accuracy and robustness, making them the top choices for wind energy prediction whereas the performance of linear regression, SVM and SVR were very poor for the considered dataset. From the experiment, Random Forest, GBM, and KNN showed the best performance with low MSE values of 0.77, 1.95, and 1.51 respectively, while other models had MSEs above 7.5, with AdaBoost reaching 30. Their RMSEs (0.88, 1.40, 1.23) and MAEs (0.093, 0.73, 0.10) also indicate strong predictive accuracy compared to the rest.In this paper, time series forecasting, TCNs, LSTM and GRU networks showed strong capabilities in capturing temporal dependencies and trends within the wind energy data. Visualization techniques were employed to compare model performances comprehensively, providing clear insights into their predictive power. Therefore, this present study offers a robust framework for researchers and practitioners aiming to leverage machine learning and time series forecasting in the realm of renewable energy prediction.

Accurate and fast prediction of sea ice conditions is the foundation of safety guarantee for Arctic navigation. Aiming at the imperious demand of short-term prediction for sea ice, we develop a new data-driven prediction technique for the sea ice concentration (SIC) combined with causal analysis. Through the causal analysis based on kernel Granger causality (KGC) test, key environmental factors affecting SIC are selected. Then multiple popular machine learning (ML) algorithms, namely self-adaptive differential extreme learning machine (SaD-ELM), classification and regression tree (CART), random forest (RF) and support vector regression (SVR), are employed to predict daily SIC, respectively. The experimental results in the Barents-Kara (B-K) sea show: (1) compared with correlation analysis, the input variables of ML models screened out by causal analysis achieve better prediction; (2) when lead time is short (<3 d), the four ML algorithms are all suitable for short-term prediction of daily SIC, while RF and SaD-ELM have better prediction performance with long lead time (>3 d); (3) RF has the best prediction accuracy and generalization ability but hugely time consuming, while SaD-ELM achieves more favorable performance when taking computational complexity into consideration. In summary, ML is applicable to short-term prediction of daily SIC, which develops a new way of sea ice prediction and provides technical support for Arctic navigation.

SummaryAccurate prediction of the solubility of chemical substances in solvents remains a challenge. The sparsity of high-quality solubility data is recognized as the biggest hurdle in the development of robust data-driven methods for practical use. Nonetheless, the effects of the quality and quantity of data on aqueous solubility predictions have not yet been scrutinized. In this study, the roles of the size and the quality of data sets on the performances of the solubility prediction models are unraveled, and the concepts of actual and observed performances are introduced. In an effort to curtail the gap between actual and observed performances, a quality-oriented data selection method, which evaluates the quality of data and extracts the most accurate part of it through statistical validation, is designed. Applying this method on the largest publicly available solubility database and using a consensus machine learning approach, a top-performing solubility prediction model is achieved.

COVID-19 in Bangladesh: A Deeper Outlook into The Forecast with Prediction of Upcoming Per Day Cases Using Time Series.

Accurate leaf area index estimation in sorghum using high-resolution UAV data and machine learning models

A machine-learning-based model for predicting the effective stiffness of precast concrete columns

Classification using machine learning algorithms in physical fitness tests carried out by students in educational centers can help prevent obesity and other related diseases. This research aims to evaluate physical fitness using percentiles of the tests and machine learning algorithms with hyperparameter optimization. The process followed was knowledge discovery in databases (KDD). Data were collected from 1525 students (784 women, 741 men) aged 6 to 17, selected non-probabilistically from five public schools. For the evaluation, anthropometric parameters such as age, weight, height, sitting height, abdominal circumference, relaxed arm circumference, oxygen saturation, resting heart rate, and maximum expiratory flow were considered. Physical Fitness tests included sitting flexibility, kangaroo horizontal jump, and 20-meter fly speed. Within the percentiles observed, we took three cut-off points as a basis for the present research: > P75 (above average), p25 to p75 (average), and < P25 (below average). The following machine learning algorithms were used for classification: Random Forest, Support Vector Machine, Decision tree, Logistic Regression, Naive Bayes, K-nearest neighbor, XGBboost, Neural network, Cat Boost, LGBM, and Gradient Boosting. The algorithms were hyperparameter optimized using GridSearchCV to find the best configurations. In conclusion, the importance of hyperparameter optimization in improving the accuracy of machine learning models is highlighted. Random Forest performs well in classifying the “High” and “Low” categories in most tests but struggles to correctly classify the “Normal” category for both male and female students.

e17607 Background: Gestational Choriocarcinoma (GCC) is a highly aggressive tumor derived from syncytiotrophoblasts. There is currently controversy surrounding the initiation of chemotherapy for patients with GCC. We aimed to investigate the influence of chemotherapy on survival rates and develop machine-learning-based prognostic models for GCC patients to improve treatment decisions and personalized medicine. Methods: Data were obtained from the Surveillance, Epidemiology, and End Results database (2000–2020). Patients who met any of the following criteria were excluded: diagnosis not confirmed by histology, previous history of cancer or with other concurrent malignancies, and unknown data. To identify prognostic variables, we conducted Cox regression analysis and constructed prognostic models using machine learning (ML) algorithms to predict the 5-year survival. Patient records were randomly divided into training (70 %) and validation (30 %) sets. A validation method incorporating the area under the curve (AUC) of the receiver operating characteristic curve was used to validate the accuracy and reliability of the ML models. We also investigated the role of multiple therapeutic options using Kaplan-Meier survival analysis. Results: This study included 732 patients, of whom 283 received chemotherapy, 116 underwent surgery alone, and 333 underwent both surgery and chemotherapy. Most patients (54.5%) were aged 30 years or older, with a median age of 32 years. The largest racial group was white, comprising 66.4% of cases, followed by black (16.7 %). Furthermore, most cases were localized (n=262), and only 13% presented with stage IV disease and 44.3% with metastasis. Survival analysis showed no significant differences in survival between the treatment modalities. Multivariate Cox regression analysis identified older age, unmarried status, and AJCC stage II as significant prognostic factors. The random forest classifier (RFC) and K-Nearest Neighbors (KNN) were the most accurate ML models, with the AJCC stage, race, and SEER stage as key factors. Performance metrics for all ML algorithms are summarized in the Table. Conclusions: Our study provides a novel and practical tool for GCC prognosis and management. Although our tool offers a valuable resource for personalized clinical decisions, further research is needed to better understand the role of chemotherapy in GCC treatment and its impact on patient outcomes. [Table: see text]

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.

This study highlights drought forecasting for understanding the semi-arid area in India, where drought phenomena play vital role in the irrigation, drinking water supplies, and sustaining the ecological with economic balance for every nation. Therefore, drought forecasting is important for the future drought planning based on the machine learning (ML) models. Hence, The Standardized Precipitation Index (SPI) at 3- and 6-month periods have been selected and used for future drought forecasting scenarios in area. The combinations of ten inputs SPI-1- and SPI-10 were used for predicting modeling for SPI-3 and SPI-6 timescales, that modeling developed based on the historical SPI datasets from 1989 to 2019 years. The SPI-3 and SPI-6 maximum and minimum values are shown SPI-3 (2.03 and -5.522) and SPI-6 (1.94 and -6.93). The SPI is a popular method for estimating the drought analysis and has been used everywhere at global level. The developed models have been compared with each other, with the best combination of input variables selected using subset regression models and sensitivity studies. After that, the active input parameters were used for forecasting of SPI-3 and SPI-6 values to understanding of drought in semi-arid area. The finest input variables combination have been used in the Ml models and established the novel five models such as robust linear regression, bagged trees, boosted trees, support vector regression (SVM-Linear), and Matern Gaussian Process Regression (Matern GPR) models. Such kind of models first time has been applied for the forecasting of future drought conditions. Whole models were fine and improved modeling by using hyperparameters tuning, bagging, and boosting models. Entire ML models’ accuracy was compared using different statistical metrics. Compared with five ML models accuracy, we have found that the Matern GPR model better accuracy than other ML models. The best model accuracy is R2 = 0.95 and 0.93, RMSE, MSE, MAE, MARE, and NSE values, respectively, for predicting SPI-3 and SPI-6 values in the area. Therefore, the Matern GPR model was identified as the finest ML algorithm for predicting SPI-3 and SPI-6 associated with other algorithms. This research demonstrates the Matern GPR model's efficacy in predicting multiscale SPI-3 and SPI-6 under climate variations. It can be helpful in soil and water resource conservation planning and management and understanding droughts in the entire basin areas of the country India.