Lowest Mean Absolute Percentage Error Research Articles

A Series of Models based on Long Short Time Memory for Temperature Prediction

Given the increasingly serious climate problems that have been appearing around the world in recent years, as part of the climate research, finding ways of smart, accurate machine learning to carry out the temperature prediction is of great significance both for human activities and the earth's ecosystem. In this study, 3 kinds of the Long Short Time Memory (LSTM) models, which differs on the way how the data is input and output and the network structure, respectively Directly-Multiple-Output (DMO), Single-Step-Scrolling (SSS), and Convolutional Neural Networks Plus Long Short Time Memory (CNN+LSTM) are built and trained by Pytorch on the Jena Climate dataset to compare their predicted performance. The training loss of these 3 models are 0.0053, 0.1568 and 0. 0079; testing loss are 0.0048, 0.1764 and 0.0096; Mean Absolute Percentage Error (MAPE) are 3.42%,6.58% and 3.30%. The result turns out that CNN+LSTM is the best model in comprehensive consideration with the least convergence time, low loss and MAPE, followed by the DMO model with the least loss and low MAPE but longer convergence time. SSS performs worst with increasing high loss. In general, CNN+LSTM is suitable for temperature prediction while DMO is also good when short convergence time is not needed. SSS is not recommended for temperature prediction.

Forecasting of GPU Prices Using Transformer Method

GPU or VGA (graphic processing unit) is a vital component of computers and laptops, used for tasks such as rendering videos, creating game environments, and compiling large amounts of code. The price of GPU/VGA has fluctuated significantly since the start of the COVID-19 pandemic in 2020, due in part to the increased demand for GPUs for remote work and online activities. Furthermore, accurate GPU price forecasting can have broader implications beyond the computer hardware industry, with potential applications in investment decision-making, production planning, and pricing strategies for manufacturers. This research aims to forecast future GPU prices using deep learning-based time series forecasting using the Transformer model. We use daily prices of NVIDIA RTX 3090 Founder Edition as a test case. We use historical GPU prices to forecast 8, 16, and 30 days. Moreover, Transformer we compare the results of the Transformer model with two other models, RNN and LSTM. We found that to forecast 30 days; the Transformer model gets a higher coefficient of correlation (CC) of 0.8743, a lower root mean squared error (RMSE) value of 34.68, and a lower mean absolute percentage error (MAPE) of 0.82 compared to the RNN and LSTM model. These results suggest that the model is an effective and efficient method for predicting GPU prices.

Analysis and Forecasting of International Airport Traffic Volume

Globalization has resulted in increases in air transportation demand and air passenger traffic. With the increases in air traffic, airports face challenges related to infrastructure, air services, and future development. Air traffic forecasting is essential to ensuring appropriate investment in airports. In this study, we combined fuzzy theory with support vector regression (SVR) to develop a fuzzy SVR (FSVR) model for forecasting international airport traffic. This model was used to predict the air traffic volumes at the world’s 10 busiest airports in terms of air traffic in 2018. The predictions were made for the period from August 2014 to December 2019. For fuzzy time series, the developed FSVR model can consider historical air traffic changes. The FSVR model can suitably divide air traffic changes into appropriate fuzzy sets, generate membership function values, and establish fuzzy relations to produce fuzzy interpolated values with minimal errors. Thus, in the prediction of continuous data, the fuzzy data with the smallest errors can be subjected to SVR to find the optimal hyperplane model with the minimum distance to the appropriate support vector sample points. The performance of the proposed model was compared with those of five other models. Of the compared models, the FSVR model exhibited the lowest mean absolute percentage error (MAPE), mean absolute error, and root mean square error for all types of traffic at all of the airports analyzed; all of the MAPE values were below 2.5. The FSVR model can predict future growth trends in air traffic, air passenger flows, aircraft flows, and logistics. An airport authority can use this model to analyze the existing operational facilities and service capacity, find bottlenecks in airport operations, and create a blueprint for future development. The findings revealed that implementing a hybrid modeling approach, specifically the FSVR model, can significantly enhance the performance of the SVR model. The FSVR model allows airlines to predict traffic growth patterns, identify viable new destinations, optimize their schedules or fleet, make accurate marketing decisions, and plan traffic effectively. The FSVR model can guide the timely construction of appropriate airport facilities with accurate predictions. Rapid, cost-effective, efficient, and balanced transportation planning enables the provision of fast, cost-effective, comfortable, safe, and convenient passenger and cargo services while ensuring the proper planning of the airport’s capacity for land-side transportation connections.

An approach based on wavelets and machine learning to build a prediction model for SF6 filling pressure of high-voltage circuit breakers

This paper presents the development of a filling pressure prediction model for sulfur hexafluoride (SF6) insulated high-voltage circuit breakers (HVCBs) for predictive maintenance, using actual data from existing HVCBs in operation and machine learning techniques. It is the result of a research project developed with a power transmission utility in Latin America to improve the maintenance procedures while reducing the environmental impact of SF6 leakages. Firstly, the authors use principal component analysis (PCA), neighborhood component analysis (NCA), and Pearson correlation (PC) to indicate the relationship of filling pressure with eleven variables, all collected from fifty HVCBs in use. Secondly, the authors employ discrete wavelet transform (DWT) Daubechies (db), symlets (sym), and coiflets (coif) on the essential variables to create smoothed signals with them. Finally, the authors apply long short-term memory (LSTM), multilayer perceptron (MLP) and extreme technical gradient boosting (XGBoost) algorithms to predict two days and one month of filling pressure and compare the results to the actual filling pressure data. The results indicate that selecting the most relevant variables and setting accordingly the wavelets generate high precision prediction models reaching low mean absolute percentage error (MAPE). The paper also proposes a mathematical approach to building efficient prediction models, which researchers can use in other predictive maintenance applications.

Rapid forecasting of hydrogen concentration based on a multilayer CNN-LSTM network

Gas sensors with rapid response are desirable in many safety applications. Reducing the response time of gas sensors is a challenging task. Computing a part of the initial temporal signals of gas sensors based on neural networks is an effective and powerful method for forecasting sensors’ output. To rapidly and robust forecasting hydrogen concentration, a sensor array is composed of a temperature and humidity sensor, and two hydrogen sensors. A neural network combined with convolutional neural networks and long-short-term memory networks is proposed to fuse temporal signals of the sensor array to forecast hydrogen concentrations. The structure of the neural network is optimized by increasing its depth. For the optimal neural network, the lowest mean absolute percent error is about 12.8% by computing initial 30 s of transient signals within 300–400 s response curves, the predicted mean absolute error is 1158 ppm in the testing range of 18 000 ppm. When the time span of initial transient signals of the sensor array increase to 150 s for the computing, the mean absolute percent error decreases to 5.7%. This study verifies the potential and effectiveness of the neural network for concentration forecasting by computing the temporal signals of the sensors.

Machine learning approach to residential valuation: a convolutional neural network model for geographic variation

Geographic location and neighbourhood attributes are major contributors to residential property values. Automated valuation models (AVM) often use hedonic pricing with location and neighbourhood attributes in the form of numeric and categorical variables. This paper proposed a novel approach to automated property valuation using a machine learning model with a convolutional neural network (CNN), fully connected neural network layers with numeric and categorical variables. In this study we compare the results of a fused model, which treat geographical data as an input with the performance of the baseline neural network model with only numerically or categorically represented data. Furthermore, the residential valuation by the proposed fused model was tested with actual sold price data in Greater Sydney, Australia. The study found that the fused model produced valuations with a significantly lower mean absolute percentage error (MAPE) (8.71%) than the MAPE of the baseline model (11.59%). The results show that the fused model with CNN significantly improves the accuracy for residential valuation, reducing spatial information loss by data manipulation and distance calibration.

Assessing the algal population dynamics using multiple machine learning approaches: Application to Macao reservoirs

The quality of reservoir water is important to the health and wellbeing of human and animals. Eutrophication is one of the most serious problems threatening the safety of reservoir water resource. Machine learning (ML) approaches are effective tools to understand and evaluate various environmental processes of concern, such as eutrophication. However, limited studies have compared the performances of different ML models to reveal algal dynamics using time-series data of redundant variables. In this study, the water quality data from two reservoirs in Macao were analyzed by adopting various ML approaches, including stepwise multiple linear regression (LR), principal component (PC)-LR, PC-artificial neuron network (ANN) and genetic algorithm (GA)-ANN-connective weight (CW) models. The influence of water quality parameters on algal growth and proliferation in two reservoirs was systematically investigated. The GA-ANN-CW model demonstrated the best performance in reducing the size of data and interpreting the algal population dynamics data, which displayed higher R-squared, lower mean absolute percentage error and lower root mean squared error values. Moreover, the variable contribution based on ML approaches suggest that water quality parameters, such as silica, phosphorus, nitrogen, and suspended solid have a direct impact on algal metabolisms in two reservoirs’ water systems. This study can expand our capacity in adopting ML models in predicting algal population dynamics based on time-series data of redundant variables.

Forecasting Peak Electricity Consumption Demand in Luzon by utilizing ARIMA Model

This research study focuses on forecasting the future values of peak demand in electricity consumption for Luzon, Philippines based on monthly historical data spanning from 2001 to 2020. The data was obtained from the official website of the Philippines Department of Energy (DOE). The primary objective of this study is to employ the ARIMA (Autoregressive Integrated Moving Average) model-building procedure developed by Box and Jenkins to accomplish accurate peak demand forecasting. The methodology involved conducting various tests and evaluations to identify the ARIMA model with the least Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). After careful analysis, the best-fitting ARIMA model was determined to be ARIMA (11, 1, 12). The findings of this study indicate that, according to the ARIMA (11, 1, 12) model, Luzon's peak demand is projected to reach 10,497.65 megawatts by December 2021. Furthermore, the model predicts that by the end of 2022, 2023, and 2024, Luzon's peak demand will be approximately 10,738.34 MW, 10,953.98 MW, and 11,148.43 MW per electrical grid, respectively. The accuracy of the ARIMA (11, 1, 12) model is found to be satisfactory, with a low MAPE value of 3.639% and the most negligible RMSE value of 517.132. The implications of these forecasted peak demand values are significant for decision-makers in the energy and utilities sector. The accurate predictions provided by the ARIMA model can aid in resource allocation, infrastructure planning, and overall operational strategies to effectively meet the anticipated high-demand periods. In conclusion, this study successfully forecasts Luzon's future values of peak demand in electricity consumption using the ARIMA (11, 1, 12) model. The findings highlight the importance of accurate peak demand forecasting and provide valuable insights for energy industry professionals.

Long-term electricity forecasting for the industrial sector in western China under the carbon peaking and carbon neutral targets

To reflect the future electricity demand variations in the industrial sector of western China under the “carbon peaking” and “carbon neutral” strategies, that the traditional methods of electricity consumption forecasting are no longer effective, this paper proposes a new hierarchy of electricity demand influencing factors based on the transition path of low-carbon and CO2 emission intensity constraints of the industrial sector. Then, the feedback equations of the influencing factors based on CO2 emission intensity is constructed, and a long-term electricity forecast model based on system dynamics is established. Finally, taking Ningxia Province as an example, the electricity demand and carbon emissions of the industrial sector are predicted under different policy constraint scenarios. The results show: (1) In the baseline scenario, electricity demand will tend to saturate with about 140.1TWh in 2030, and the CO2 emission level of electricity demand will peak at 72.52 million tons in 2029. (2) Compared with the system dynamics model without considering CO2 emission intensity, the mean absolute percentage error is reduced about 3.16 % by inputting the influence factors and regression coefficients designed in this paper. (3) Compared with other forecasting methods, the model proposed in this paper has the lowest mean absolute percentage error of 2.75 %. The model improves the accuracy and provides a reference for planning departments to develop energy transition plans, where the prediction results reflect the reality of the industrial sector in western China. Correspondingly, this model has reference value for relevant institutions in forecasting the electricity demand under the low-carbon development path in industrial sectors of different regions.

Multiplex Analysis of Fundamental Surface Water Quality Indicators by Multichannel Electrochemical Device Combined With MFCSC Model

Rapid and accurate detection of water quality is essential for environmental monitoring and management. Among them, permanganate index (CODMn), nitrate–nitrogen (NO3−-N), dissolvable inorganic phosphorus (DIP), and ammonia nitrogen (AN) are four important water quality indicators. A portable multichannel electrochemical device (MCED) combined with a sensor array composed of gold nanoparticles (AuNPs), silver nanodendritic structures (AgNDs), and reduced graphene oxide (rGO)-modified screen-printed electrodes (SPEs) was applied for facile, rapid on-site inspection. The sensor array exhibits good complementarity in voltammetric measurements of water quality indicators and has cross-sensitivity in multiplex measurement. A multiple fully connected shared convolution (MFCSC) model was constructed to process the voltammetric data, extracting unique features to resolve the overlap in voltammetric data of complex solutions for the quantitative determination of target analytes. Accurate predictions were successfully achieved for the four parameters, CODMn, NO3−-N, DIP, and AN in the mixture, where CODMn, NO3−-N, and AN have high coefficients of determination ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${R} ^{{2}}$ </tex-math></inline-formula> ) and low mean absolute percentage errors (MAPE%). Using the migration ability of MFCSC, combined with MCED, it is further applied for real water sample testing, which proves the good prediction of the water quality indicators of real water samples in complex matrices. The results show that the sensing system developed in this work has wide application potential and can be used to rapidly analyze surface water quality and obtain water quality information in a wide range of water bodies for monitoring.

Guidelines for Increasing the Effectiveness of Thailand’s Sustainable Development Policy based on Energy Consumption: Enriching the Path-GARCH Model

The objective of this study is to develop a model for forecasting energy consumption and to increase the effectiveness of Thailand's sustainable development policy based on energy consumption by using the best model, the Path Analysis-Generalized Autoregressive Conditional Heteroscedasticity Model (Path-GARCH model). To improve the effectiveness of sustainability policies, the researcher has envisioned the final energy consumption over a 20-year period (AD 2023–2022) by defining a new scenario policy. Comparing the performance of the Path-GARCH model to other previous models, the Path-GARCH model was found to have the lowest mean absolute percentage error (MAPE) and root mean square error (RMSE) values. In addition, the study found that energy consumption continued to rise to 125,055 ktoe by 2042, with a growth rate of 115.05% between 2042 and 2023, which exceeded the carrying capacity limit of 90,000 ktoe. When a new scenario policy is implemented, however, the final energy consumption continues to rise to 74,091 ktoe (2042). Consequently, defining a new scenario policy is a crucial development guideline for enhancing the effectiveness of Thailand's sustainable development policy.

Supervised learning techniques for dairy cattle body weight prediction from 3D digital images.

Introduction: The use of automation and sensor-based systems in livestock production allows monitoring of individual cows in real-time and provides the possibility of early warning systems to take necessary management actions against possible anomalies. Among the different RT monitoring parameters, body weight (BW) plays an important role in tracking the productivity and health status. Methods: In this study, various supervised learning techniques representing different families of methods in the machine learning space were implemented and compared for performance in the prediction of body weight from 3D image data in dairy cows. A total of 83,011 records of contour data from 3D images and body weight measurements taken from a total of 914 Danish Holstein and Jersey cows from 3 different herds were used for the predictions. Various metrics including Pearson's correlation coefficient (r), the root mean squared error (RMSE), and the mean absolute percentage error (MAPE) were used for robust evaluation of the various supervised techniques and to facilitate comparison with other studies. Prediction was undertaken separately within each breed and subsequently in a combined multi-breed dataset. Results and discussion: Despite differences in predictive performance across the different supervised learning techniques and datasets (breeds), our results indicate reasonable prediction accuracies with mean correlation coefficient (r) as high as 0.94 and MAPE and RMSE as low as 4.0 % and 33.0 (kg), respectively. In comparison to the within-breed analyses (Jersey, Holstein), prediction using the combined multi-breed data set resulted in higher predictive performance in terms of high correlation coefficient and low MAPE. Additional tests showed that the improvement in predictive performance is mainly due to increase in data size from combining data rather than the multi-breed nature of the combined data. Of the different supervised learning techniques implemented, the tree-based group of supervised learning techniques (Catboost, AdaBoost, random forest) resulted in the highest prediction performance in all the metrics used to evaluate technique performance. Reported prediction errors in our study (RMSE and MAPE) are one of the lowest in the literature for prediction of BW using image data in dairy cattle, highlighting the promising predictive value of contour data from 3D images for BW in dairy cows under commercial farm conditions.

Ensemble of Effect Size Methods Based on Meta Fuzzy Functions

Many methods are used in the literature to determine the effect size (ES) for two independent groups. Many of these methods yield consistent results under the assumptions of normality of data and homogeneity of variances. However, not every dataset can provide these assumptions. In order to overcome the limitations mentioned, this study proposes the use of meta fuzzy effect size functions (MFESF). The MFESF weights six ES methods (Cohen’s d, Hedge’s g, and Glass’ delta, Cliff’s delta, Vargha and Delaney A and Glass’ rank-biserial correlation) used for two independent groups according to their performances and provides better outcomes, regardless of assumptions. The MFESF method uses the fuzzy c-means (FCM) clustering algorithm to combine the selected ES methods. In this study, MFESF is evaluated by using generated datasets based on normal and non-normal distribution for six reference values. In addition, the performance of MFESF is evaluated by using real datasets with normal and non-normal distributions. As a result, the MFESF performed the best with the lowest mean absolute percentage error (MAPE) compared to the individual ES methods for all datasets.

Simulation and validation of response of ‘Shalimar wheat 2’ (Triticum aestivum) to sowing dates and nitrogen levels under rainfed conditions using DSSAT-CSM-CERES-wheat model

A-2 year experiment was carried out at Wadura, Kashmir, Jammu and Kashmir during 2015–16 and 2016–17 in a split-plot design, keeping 3 sowing dates (15 October, 30 October and 15 November) in mainplots and 4 nitrogen levels (0 kg, 50, 100 and 150 kg N/ha) in subplots, to simulate the response of Shalimar Wheat-2 var. wheat (Triticum aestivum L.) to sowing dates and nitrogen levels under rainfed conditions using DSSAT-CSM-CERES-Wheat. The pooled results of 2 years revealed that the highest grain yield of 4.46 t/ha, straw yield of 8.30 t/ha and biological yield of 12.8 t/ha were obtained when sowing was done on 15th October. The grain, straw and biological yields decreased by 17.33, 12.58 and 14.25% with delay in sowing from 15 October to 15 November, respectively. Genetic coefficients of newly evolved wheat cultivar ‘Shalimar Wheat 2’ were generated for calibration and validation of model CERES-wheat Ver.4.6 (DSSAT). The DSSAT-CERES-Wheat model performed well, as revealed by high correlation coefficient (r), low root mean square error (RMSE) and low mean absolute percentage error (MAPE) in simulating the days to anthesis (r, 0.99; RMSE, 3.9; MAPE, 1.84), maturity (r, 0.97; RMSE, 3.9; MAPE, 1.51), leaf area index (r, 0.90; RMSE, 0.2; MAPE, 6.22), grain yield (r, 0.92; RMSE, 0.39; MAPE, 17.85), and biological yield (r, 0.89; RMSE, 1.03; MAPE, 7.71). Also, a good line of fit (1:1) was found between the simulated and observed grain yield with R2 of 0.89.

Forecasting the Municipal Solid Waste Using GSO-XGBoost Model

Waste production rises in tandem with population growth and increased utilization. The indecorous disposal of waste paves the way for huge disaster named as climate change. The National Environment Agency (NEA) of Singapore oversees the sustainable management of waste across the country. The three main contributors to the solid waste of Singapore are paper and cardboard (P&C), plastic, and food scraps. Besides, they have a negligible rate of recycling. In this study, Machine Learning techniques were utilized to forecast the amount of garbage also known as waste audits. The waste audit would aid the authorities to plan their waste infrastructure. The applied models were k-nearest neighbors, Support Vector Regressor, ExtraTrees, CatBoost, and XGBoost. The XGBoost model with its default parameters performed better with a lower Mean Absolute Percentage Error (MAPE) of 8.3093 (P&C waste), 8.3217 (plastic waste), and 6.9495 (food waste). However, Grid Search Optimization (GSO) was used to enhance the parameters of the XGBoost model, increasing its effectiveness. Therefore, the optimized XGBoost algorithm performs the best for P&C, plastics, and food waste with MAPE of 4.9349, 6.7967, and 5.9626, respectively. The proposed GSO-XGBoost model yields better results than the other employed models in predicting municipal solid waste.

Forecasting exchange rate dynamics in developing countries

Given that volatility influences decisions about currency rates, monetary policy, and macroeconomic policy, it is crucial to predict and anticipate volatility in emerging economies. The study employed generalized autoregressive conditional heteroskedasticity (GARCH) asymmetric models to estimate and forecast exchange rate dynamics in developing countries. We found that South Africa model had similar variance and covariance proportion of 0.99356 percent and 0.995901 percent respectively and the exchange rate could rise or fall by 2 to 6 units of rand, in exchange for USD. In Kenya, exchange rates continually exhibited steady rise monthly with extremely low mean absolute percentage error of 0.01568 percent and this demonstrates how strongly the model predicts Kenya’s future currency rates while the variance chart supports absence of persistence. In Ghana, exchange rates are projected to increase significantly as 99.5 percent of unsystematic error was un accounted for in the model. Volatility is highly persistent in Nigeria; hence the forecasting model reported a high error rate by taking 1.06 percent of the symmetric error into cognizance. Kenya, Ghana, and Mauritius had asymmetry in currency volatility, revealing turbulence in exchange rates when the bad news hit the market. Hence, local currencies are rendered worthless in the foreign exchange market.

A New Estimation Scheme for Improving the Performance of Shear Wave Elasticity Imaging.

Shear wave velocity (SWV) reconstruction based on time-of-flight (TOF) is widely adopted to realize shear wave elasticity imaging (SWEI). It typically breaks down the reconstruction of a SWV image into many kernels and treats them independently. We hypothesized that information exchange among kernels improves the performance of SWEI. Therefore, we propose the approach of iterative re-weighted least squares based on inter-kernel communication (IKC-IRLS). We also hypothesized that time-to-peak (TTP) is superior to cross-correlation (CC) in visualizing small targets because TTP uses higher shear wave frequencies than CC. To examine the hypotheses, IKC-IRLS was combined with TTP data and compared with four established methods. The five methods were tested by imaging several small-size stiff targets (2.5, 4.0 and 6.4 mm in diameter) using different kernel sizes in the simulation and real experiments. The results indicate that the IKC-IRLS approach can mitigate speckle noise and is robust to TTP outliers. Consequently, the proposed method achieves the highest contrast-to-noise ratio and the lowest mean absolute percentage error of target in almost all tested cases.

Analysis of Ghana’s Gross Domestic Product from 1960 - 2019

The most popular metric for assessing or predicting global economic progress is the GDP. The objective of this report was to analyze the GDP of Ghana from the period 1960 – 2019. Secondary data was obtained, and trend analysis was done using the linear, quadratic, and exponential trend models to determine which model best fits the dataset. Trend analysis is a technique for examining and forecasting the movements of an item based on current and historical data. The results from the trend analysis showed that the exponential trend model had the lowest Mean Absolute Percent Error (MAPE), Mean Absolute Deviation (MAD), and Mean Square Deviation (MSD). When compared to the other models, the exponential trend model fits the dataset better, which is why it was chosen to forecast Ghana's GDP. The forecast showed that Ghana's GDP is expected to grow in the coming years to about $77 billion by 2026. Agriculture is considered as the backbone of Ghana and the country imports majority if its fertilizer from Russia. The challenges the Ghanaian economy is currently facing due to inflation, the global pandemic (COVID-19), and the Russia- Ukraine conflict could have an impact on the country’ economic growth. Moreover, a change in leadership in the coming 2024 presidential election, could also have an impact on the projection. There are both positive and negative effects due to changes in leadership on economic growth. Keywords: GDP, linear, quadratic, exponential, forecast, economic DOI: 10.7176/JESD/14-2-07 Publication date: January 31 st 2023

Time Series Prediction of Wheat Crop based on FB Prophet Forecast Framework

The production of wheat plays an important role in the Indian economy. Wheat yield prediction is significant in trade, industry, and agriculture to increase profitability and better growth for farmers. We propose a prediction model to classify the wheat yield using time series analysis using the FB Prophet algorithm, which is considered as better than most of the other supervised learning models with respect to accuracy. [1]. The study aims to evaluate the predicted growth of wheat yield for the next five years. The dataset is collected by the government agency of India [2], considering the years 1997 to 2022, seasonal data, Gujarat state with four districts, and analysis is done for the Wheat/ Rabi crop. A total of 589 instances are collected from a dataset. We pre-process the data, train the data, and through the testing result set, the experimental result indicates the model achieves the lowest Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) for the summer wheat prediction (10.03 and 0.39 respectively) when the number of the layer in seasonality is yearly. The study will help the research community and other stakeholders to make plans for the next five years for the sustainable growth of India.

Forecasting the concentration of NO2 using statistical and machine learning methods: A case study in the UAE

Nitrogen dioxide (NO2) is the most active pollutant gas emitted in the industrial era and is highly correlated with human activities. Tracking NO2 emissions and predicting their concentrations represent important steps toward controlling pollution and setting rules to protect people's health indoors, such as in factories, and in outdoor environments. The concentration of NO2 was affected by the COVID-19 lockdown period and decreased because of restrictions on outdoor activities. In this study, the concentration of NO2 was predicted at 14 ground stations in the United Arab Emirates (UAE) during December 2020 based on training over a full time period of two years (2019–2020). Statistical and machine learning models, such as autoregressive integrated moving average (ARIMA), seasonal autoregressive integrated moving average (SARIMA), long short-term memory (LSTM), and nonlinear autoregressive neural network (NAR-NN), are used with both open- and closed-loop architectures. The mean absolute percentage error (MAPE) was used to evaluate the performance of the models, and the results ranged from “very good” (MAPE of 8.64% at the Liwa station with the closed loop) to “acceptable” (MAPE of 42.45% at the Khadejah School station with the open loop). The results show that the predictions based on the open loop are generally better than those based on the closed loop because they yield statistically significantly lower MAPE values. For both loop types, we selected stations exhibiting the lowest, medium, and highest MAPE values as representative cases. In addition, we demonstrated that the MAPE value is highly correlated with the relative standard deviation of NO2 concentration values.