Minimum Root Mean Square Error Research Articles

Performance analysis for a suitable propagation model in outdoor with 2.5 GHz band

As demand for mobile wireless network services continues to rise, network planning and optimization significantly affect development. One of the critical elements in network planning is predicting pathloss. Thus, propagation models predict pathloss in indoor and outdoor environments. Choosing the appropriate propagation model for the area out of existing models is essential for network planning. Selected propagation models suitable with 2.5GHz, such as Friis Free Space Propagation Model (FSPL), Sandford University Interim (SUI), Ericsson, Okumura, and COST-231 HATA models, are utilized for evaluation and compared with empirical data collected from long-term evolution (LTE) networks in urban areas. The best acceptable model is chosen based on statistical results such as mean, standard deviation, and root mean square errors (RMSE). The analytical results show Cost-231 Hata model fits the empirical pathloss with a minimum RMSE of 5.27 dB.

Machine Learning Application for Black Friday Sales Prediction Framework

Abstract: Understanding the purchase behavior of various customers (dependent variable) against different products using their demographic information (IS features where most of the features are self -explanatory. This dataset consist of null values, redundant and unstructured data. Machine learning is the most common applications in the domain retail industry. This concept helps to develop a predictor that has a distinct commercial value to the shop owners as it will help with their inventory management, financial planning, advertising and marketing. This entire process of developing a model includes preprocessing, modelling, training testing and evaluating. Hence, frameworks will be developed to automate few of this process and its complexity will be reduced. The algorithm we proposed was Random Forest regressor that performed an average accuracy of 83.6% and with minimum RMSE (Root Mean Squared Error) value of 2829 on tire Black Friday sales dataset.

Parameter Estimation of a Thermoelectric Generator by Using Salps Search Algorithm

Thermoelectric generators (TEGs) have the potential to convert waste heat into electrical energy, making them attractive for energy harvesting applications. However, accurately estimating TEG parameters from industrial systems is a complex problem due to the mathematical complex non-linearities and numerous variables involved in the TEG modeling. This paper addresses this research gap by presenting a comparative evaluation of three optimization methods, Particle Swarm Optimization (PSO), Salps Search Algorithm (SSA), and Vortex Search Algorithm (VSA), for TEG parameter estimation. The proposed integrated approach is significant as it overcomes the limitations of existing methods and provides a more accurate and rapid estimation of TEG parameters. The performance of each optimization method is evaluated in terms of root mean square error (RMSE), standard deviation, and processing time. The results indicate that all three methods perform similarly, with average RMSE errors ranging from 0.0019 W to 0.0021 W, and minimum RMSE errors ranging from 0.0017 W to 0.0018 W. However, PSO has a higher standard deviation of the RMSE errors compared to the other two methods. In addition, we present the optimized parameters achieved through the proposed optimization methods, which serve as a reference for future research and enable the comparison of various optimization strategies. The disparities observed in the optimized outcomes underscore the intricacy of the issue and underscore the importance of the integrated approach suggested for precise TEG parameter estimation.

Critical measurement parameters estimation in liquid rocket engine using LSTM-based soft sensor

Liquid Rocket engines (LREs) need to be hot tested on the ground, in order to make them flight-worthy. Ground hot tests are critical, expensive, and involve considerable human efforts in acquiring the critical measurement parameters. Measurement sensors are prone to failure due to the hazardous propellant environment and are vulnerable to noise (internal and external) during the LRE ground hot tests. The usage of redundant sensors is not a viable solution for space applications due to cost, power, weight, and space constraints. Data-driven Soft Sensors (SS) are being used in different industrial processes. SS can provide an effective alternative and economical solution for space applications. The objective of the proposed work is to implement a suitable data-driven soft sensor for estimating three critical measurement parameters namely, turbine speed, chamber pressure (Pc), and propellant flow rate of LRE. Long Short-Term Memory (LSTM) networks are capable of learning order dependence in sequence prediction problems and are well-suited to make predictions based on time series data. As the selected LRE test data contain order dependence sequences, LSTM is well suited for the estimation of critical measurement parameters. LSTM-based soft sensor (LSTM-SS) was developed and trained using a large sample set of 59,80,000 data points collected over a period of 12 years of LRE ground hot tests. The maximum root mean square error (RMSE) of 6.06% was obtained for propellant flow-rate estimation, whereas the minimum RMSE of 0.22% for chamber pressure and RMSE of 0.87% for turbine speed were obtained. The obtained results show that the designed LSTM-SS has superior performance and better generalization for practical use.

Predicting the Single Diode Model Parameters using Machine Learning Model

The increasing energy demands, requires the usage of solar photovoltaic (SPV) more largely. The work focuses on the Single Diode Model (SDM) of solar panels with the help of five parameters namely series resistance (Rse), shunt resistance (Rsh), diode ideality factor (A), light generated current (ILG), and diode reverse saturation current (Isat). To identify unknown parameters, modeling is done by deriving the equations and initial values are assumed under Standard Test Conditions (STC). The deviations of Rse and Rsh values are very less when the experimented and analytical values of the AlokP module are compared. The predicted values hold good for the panel and show that the NR (Newton Raphson) method is simple, robust, and computationally less intensive. With the help of NR methodology, the dataset is created for various solar panels. The proposed methodology is checked with various other PV (Photovoltaic) models and the best-fit machine learning model is predicted. The unknown values are predicted under cross-out validation and hold-out validation methods and the best-fit machine learning model is identified by checking the least RMSE (Root Mean Squared Error) value. Nineteen machine learning models are compared and the model with the minimum RMSE value is chosen to the best-fit model.

An assessment of stingless beehive climate impact using multivariate recurrent neural networks

<p>A healthy bee colony depends on various elements, including a stable habitat, a sufficient source of food, and favorable weather. This paper aims to assess the stingless beehive climate data and examine the precise short-term forecast model for hive weight output. The dataset was extracted from a single hive, for approximately 36-hours, at every seven seconds time stamp. The result represents the correlation analysis between all variables. The evaluation of root-mean-square error (RMSE), as well as the RMSE performance from various types of topologies, are tested on four different forecasting window sizes. The proposed forecast model considers seven of input vectors such as hive weight, an inside temperature, inside humidity, outside temperature, outside humidity, the dewpoint, and bee count. The various network architecture examined for minimal RMSE are long short-term memory (LSTM) and gated recurrent units (GRU). The LSTM1X50 topology was found to be the best fit while analyzing several forecasting windows sizes for the beehive weight forecast. The results obtained indicate a significant unusual symptom occurring in the stingless bee colonies, which allow beekeepers to make decisions with the main objective of improving the colony’s health and propagation.</p>

Machine learning accelerated approach to infer nuclear magnetic resonance porosity for a middle eastern carbonate reservoir

Carbonate rocks present a complicated pore system owing to the existence of intra-particle and interparticle porosities. Therefore, characterization of carbonate rocks using petrophysical data is a challenging task. Conventional neutron, sonic, and neutron-density porosities are proven to be less accurate as compared to the NMR porosity. This study aims to predict the NMR porosity by implementing three different machine learning (ML) algorithms using conventional well logs including neutron-porosity, sonic, resistivity, gamma ray, and photoelectric factor. Data, comprising 3500 data points, was acquired from a vast carbonate petroleum reservoir in the Middle East. The input parameters were selected based on their relative importance with respect to output parameter. Three ML techniques such as adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and functional network (FN) were implemented for the development of prediction models. The model’s accuracy was evaluated by correlation coefficient (R), root mean square error (RMSE), and average absolute percentage error (AAPE). The results demonstrated that all three prediction models are reliable and consistent exhibiting low errors and high ‘R’ values for both training and testing prediction when related to actual dataset. However, the performance of ANN model was better as compared to other two studied ML techniques based on minimum AAPE and RMSE errors (5.12 and 0.39) and highest R (0.95) for testing and validation outcome. The AAPE and RMSE for the testing and validation results were found to be 5.38 and 0.41 for ANFIS and 6.06 and 0.48 for FN model, respectively. The ANFIS and FN models exhibited ‘R’ 0.937 and 0.942, for testing and validation dataset, respectively. Based on testing and validation results, ANFIS and FN models have been ranked second and third after ANN. Further, optimized ANN and FN models were used to extract explicit correlations to compute the NMR porosity. Hence, this study reveals the successful applications of ML techniques for the accurate prediction of NMR porosity.

Spatial scale applicability of 4-Scale geometrical optics model canopy reflectance simulation.

Accurately clarifying the applicable spatial scale of 4-Scale model is conducive to improving the accuracy of its application in canopy reflectance simulation of different vegetation types, and to further improving the inversion accuracy of leaf area index, canopy density, and other parameters. Two forest plots (one for broad-leaved forest and one for mixed forest) with each area of 100 m×100 m in Maoershan Experimental Forest Farm, Shangzhi, Heilongjiang, were divided into the spatial scales of 10, 20, 30, 40 and 50 m, respectively. The 4-Scale model was used to simulate forest canopy reflectance. Local mean method, the nearest neighbor method, bilinear interpolation method, and cubic convolution method were used to convert Sentinel-2 images with spatial resolution of 10 m to other scales, with the results being evaluated. The simulated canopy reflectance and remote sensing pixel reflectance were compared and analyzed. The spatial scale of mixed forest and broad-leaved forest suitable for high-precision inversion parameters of 4-Scale model was determined. The results showed that the 4-Scale model underestimated the pixel forest canopy reflectance as a whole. The canopy reflectance of mixed forest and broad-leaved forest had the worst simulation effect at the 20 m scale. Both the root mean square error (RMSE) and the mean absolute error from (MAE) of red and near-infrared band were large. When the scale was >20 m, the simulation effect became better. The applicability of the model was the best when the mixed forest was 40 m and the broad-leaved forest was 30 m. The mean and standard deviation of the reflectance difference between the simulated value and the remote sensing pixel were the minimum in the red and near near-infrared bands, with the minimum RMSE and MAE. The simulation results of mixed forest and broad-leaved forest at 10 m scale were not stable, the rule of mean and standard deviation was inconsistent, and the difference between RMSE and MAE was large under the same band.

Computer vision analysis of sample colors versus quadruple-disk iridium-platinum voltammetric e-tongue for recognition of natural honey adulteration

This work presents a deep learning approach for computer vision-based recognition of natural honey adulteration. The results are compared with the machine learning models using chemical data obtained with an innovative bimetallic voltammetric electrode that acts as an e-tongue. Computer vision is based on the recognition of the color of the sample, which changes with the addition of undesirable synthetic substances. The pictures were taken with a smartphone under controlled conditions using seven different background illumination colors for samples that contained 0 – 50 % foreign additives. In the best case root mean square error (RMSE) in the regression approach was 0.46 % for the external test set at R2 = 0.9993, calculated as a compatibility evaluation between the true values and the prediction result. The minimum RMSE for the voltammetric measurements on the quadruple-disk iridium-platinum electrode and the regression algorithms PCR, PLSR, and SVR was 0.25 % with R2 = 0.9998. It has been shown that when food adulteration affects its color, it is possible to successfully replace the classical e-tongue with computer vision. The convenient, rapid, and nondestructive approach does not require the use of equipment, reagents, or solvents and laboratory work techniques typical for analytical chemistry.

Nesting the SIRV model with NAR, LSTM and statistical methods to fit and predict COVID-19 epidemic trend in Africa

ObjectiveCompared with other regions in the world, the transmission characteristics of the COVID-19 epidemic in Africa are more obvious, has a unique transmission mode in this region; At the same time, the data related to the COVID-19 epidemic in Africa is characterized by low data quality and incomplete data coverage, which makes the prediction method of COVID-19 epidemic suitable for other regions unable to achieve good results in Africa. In order to solve the above problems, this paper proposes a prediction method that nests the in-depth learning method in the mechanism model. From the experimental results, it can better solve the above problems and better adapt to the transmission characteristics of the COVID-19 epidemic in African countries.MethodsBased on the SIRV model, the COVID-19 transmission rate and trend from September 2021 to January 2022 of the top 15 African countries (South Africa, Morocco, Tunisia, Libya, Egypt, Ethiopia, Kenya, Zambia, Algeria, Botswana, Nigeria, Zimbabwe, Mozambique, Uganda, and Ghana) in the accumulative number of COVID-19 confirmed cases was fitted by using the data from Worldometer. Non-autoregressive (NAR), Long-short term memory (LSTM), Autoregressive integrated moving average (ARIMA) models, Gaussian and polynomial functions were used to predict the transmission rate β in the next 7, 14, and 21 days. Then, the predicted transmission rate βs were substituted into the SIRV model to predict the number of the COVID-19 active cases. The error analysis was conducted using root-mean-square error (RMSE) and mean absolute percentage error (MAPE).ResultsThe fitting curves of the 7, 14, and 21 days were consistent with and higher than the original curves of daily active cases (DAC). The MAPE between the fitted and original 7-day DAC was only 1.15% and increased with the longer of predict days. Both the predicted β and DAC of the next 7, 14, and 21 days by NAR and LSTM nested models were closer to the real ones than other three ones. The minimum RMSEs for the predicted number of COVID-19 active cases in the next 7, 14, and 21 days were 12,974, 14,152, and 12,211 people, respectively when the order of magnitude for was 106, with the minimum MAPE being 1.79%, 1.97%, and 1.64%, respectively.ConclusionNesting the SIRV model with NAR, LSTM, ARIMA methods etc. through functionalizing β respectively could obtain more accurate fitting and predicting results than these models/methods alone for the number of confirmed COVID-19 cases in Africa in which nesting with NAR had the highest accuracy for the 14-day and 21-day predictions. The nested model was of high significance for early understanding of the COVID-19 disease burden and preparedness for the response.

Watershed prioritization and decision-making based on weighted sum analysis, feature ranking, and machine learning techniques

Prediction and validation of Compound factors for prioritization of watersheds are an essential application using machine learning (ML) techniques in water resource engineering. The current paper proposes a methodology to derive 14 morphometric and 3 topo-hydrological parameters using remote sensing (RS) and geographical information systems (GIS). Compound factor (CF) values are calculated using weighted sum analysis (WSA), ReliefF, and the Pearson correlation coefficient, and the important parameters are identified. Two machine learning models, multilayer perceptron (MLP) and support vector machine (SVM), are utilized to predict CF values. Predication accuracy of ML models is evaluated with three parameters, mean absolute error (MAE), Pearson correlation coefficient (PCC), and root mean square error (RMSE). It is observed that the maximum value of PCC equal to 1 is achieved through ReliefF and SVM, whereas minimum MAE and RMSE are observed with ReliefF and SVM when Tenfold cross-validation is applied. Since ReliefF shows better results, CF values are calculated and applied to create the watershed. The proposed methodology is helpful for accurately predicting CF values and advantageous to allocating the proper watershed, which will be useful for decision-making and implementation of conservation techniques for soil and water.

Prediction of the Discharge Coefficient in Compound Broad-Crested-Weir Gate by Supervised Data Mining Techniques

The current investigation evaluated the discharge coefficient of a combined compound rectangular broad-crested-weir (BCW) gate (Cdt) using the computational fluid dynamics (CFD) modeling approach and soft computing models. First, CFD was applied to the experimental data and 61 compound BCW gates were numerically simulated by resolving the Reynolds-averaged Navier–Stokes equations and stress turbulence models. Then, six data-driven procedures, including M5P tree, random forest (RF), support vector machine (SVM), Gaussian process (GP), multimode ANN and multilinear regression (MLR) were used for estimating the coefficient of discharge (Cdt) of the weir gates. The results showed the superlative accuracy of the SVM model compared to M5P, RF, GP and MLR in predicting the discharge coefficient. The sensitivity investigation revealed the h1/H as the most effective parameter in predicting the Cdt, followed by the d/p, b/B0, B/B0 and z/p. The multimode ANN model reduced the root mean square error (RMSE) of M5P, RF, GP, SVM and MLR by 37, 13, 6.9, 6.5 and 32%, respectively. The graphical inspection indicated the multimode ANN model as the most suitable for predicting the Cdt of a BCW gate with minimum RMSE and maximum correlation.

A Forecast of Surface Ozone Using Analytical Models

In this study, several analytical models were tested to forecast the surface ozone concentration using Artificial Neural Network (ANN), Multiple Linear Regression (MLR) and Time Series Regression (TSR). Four study areas were selected in this study, namely Seberang Jaya in Penang, Shah Alam in Selangor, Larkin in Johor and Kota Bharu in Kelantan. The main objective of this study is to determine the appropriate analytical models MLR and ANN for surface ozone forecasting in some zones of peninsular Malaysia, to forecast surface ozone concentration with TSR model in several zones of peninsular Malaysia and to compare the performance of each model by the performance index. The performance index that will be shown in this study for the model comparison are root mean square error (RMSE), mean square error (MSE) and determination of coefficient (R2). The ANN model showed better performance compared to the MLR and TSR models in the model comparison in each station. The station in Larkin, Johor provides high accuracy in forecasting surface ozone concentrations for each model with minimum MSE, 0.000009 ppm and RMSE, 0.0042 ppm compared to other stations. The value of R2 is 0.33 which is highest compared to station in Seberang Jaya and Kota Bharu.

HRLT: a high-resolution (1 d, 1 km) and long-term (1961–2019) gridded dataset for surface temperature and precipitation across China

Abstract. Accurate long-term temperature and precipitation estimates at high spatial and temporal resolutions are vital for a wide variety of climatological studies. We have produced a new, publicly available, daily, gridded maximum temperature, minimum temperature, and precipitation dataset for China with a high spatial resolution of 1 km that covers a long-term period (1961 to 2019). It has been named the HRLT, and the dataset is publicly available at https://doi.org/10.1594/PANGAEA.941329 (Qin and Zhang, 2022). In this study, the daily gridded data were interpolated using comprehensive statistical analyses, which included machine learning methods, the generalized additive model, and thin plate splines. It was based on the 0.5∘ × 0.5∘ gridded dataset from the China Meteorological Administration, together with covariates for elevation, aspect, slope, topographic wetness index, latitude, and longitude. The accuracy of the HRLT daily dataset was assessed using observation data from meteorological stations across China. The maximum and minimum temperature estimates were more accurate than the precipitation estimates. For maximum temperature, the mean absolute error (MAE), root mean square error (RMSE), Pearson's correlation coefficient (Cor), coefficient of determination after adjustment (R2), and Nash–Sutcliffe modeling efficiency (NSE) were 1.07 ∘C, 1.62 ∘C, 0.99, 0.98, and 0.98, respectively. For minimum temperature, the MAE, RMSE, Cor, R2, and NSE were 1.08 ∘C, 1.53 ∘C, 0.99, 0.99, and 0.99, respectively. For precipitation, the MAE, RMSE, Cor, R2, and NSE were 1.30 mm, 4.78 mm, 0.84, 0.71, and 0.70, respectively. The accuracy of the HRLT was compared to those of three other existing datasets, and its accuracy was either greater than the others, especially for precipitation, or comparable in accuracy, but with higher spatial resolution or over a longer time period. In summary, the HRLT dataset, which has a high spatial resolution, covers a longer period of time and has reliable accuracy.

Method of Hydrological Time Series Recovery Using Neural Network Technologies

n this work we applied machine learning methods in the field of hydrological measurements to recover (to predict) missing or damaged data. Field measurements of temperature field that were taken as a typical example, were carried out on the Sea of Japan shelf (the Peter the Great Bay) in October 2021 using a vertical moored thermostring. Recording of one of its sensors was partly deleted in manual way. To recover the missing part, we used the method of averaging the readings of the nearest sensors, as well as one-dimensional and two-dimensional Long Short-Term Memory (LSTM) neural network models. The results were compared with real data; recovery quality was assessed using the Root Mean Square Error (RMSE) of the prediction. The study showed that two-dimensional LSTM provides more accurate data recovering, with a minimum RMSE of 0.014 ± 0.006. The simulation results prove that multidimensional recovering can significantly increase the length of the predicted series while maintaining acceptable accuracy and compensate for the error accumulation effect that is observed with one-dimensional recovering. We believe that multidimensional data recovery method based on the LSTM neural network is very promising for non-stationary time series.

Inversion of Soil Organic Matter Content Based on Improved Convolutional Neural Network.

Soil organic matter (SOM) is an important source of nutrients required during crop growth and is an important component of cultivated soil. In this paper, we studied the possibility of using deep learning methods to establish a multi-feature model to predict SOM content. Moreover, using Nong’an County of Changchun City as the study area, Sentinel-2A remote sensing images were taken as the data source to construct the dataset by using field sampling and image processing. The LeNet-5 convolutional neural network model was chosen as the deep learning model, which was improved based on the basic model. The evaluation metrics were selected as the root mean square error (RMSE) and the coefficient of determination R2. Through comparison, the R2 of the improved model was found to be higher than that of the linear regression method, Support Vector Machines (SVM) (RMSE = 2.471, R2 = 0.4035), and Random Forest (RF) (RMSE = 2.577, R2 = 0.4913). The result shows that: (1) It is feasible to use the multispectral data extracted from remote sensing images for soil organic matter content inversion based on the deep learning model with a minimum RMSE of 2.979 and with the R2 reaching 0.89. (2) The choice of features has an impact on the prediction of the model to a certain extent. After ranking the importance of features, selecting the appropriate number of features for inversion provides better results than full feature inversion, and the computational speed is improved.

Testing the equations for wave attenuation over shore platforms

Modelling the geomorphologic evolution of rocky coasts over geological time requires estimating the attenuation of waves approaching cliffs and thus the forces of waves impacting lower cliff surfaces. Sunamura et al. (2014) proposed a set of equations for waves crossing horizontal shore platforms that are potentially useful for geomorphological modelling. Here, we use a compilation of wave measurements to test their equations on dipping platforms and in particular to evaluate the equation parameters. We select only those measurements made where waves approached coastlines nearly orthogonally, where tidal height variations are small and where satellite images show fewer complications from wave refraction, reefs, or coastal geomorphology causing reflected waves. For each dipping rock platform, we forward model the wave heights using the measured wave height at the most seaward location as the boundary condition and apply the Sunamura et al. (2014) equations in a series of small steps across the platform. Using the parameters provided by Sunamura et al. (2014), the model predicts wave heights with a global root-mean square error (RMSE) that is <0.08 m. We repeated the analysis varying parameters by factors of two and four (both smaller and larger), finding that the minimum RMSEs are achieved with exactly the same parameters proposed by Sunamura et al. (2014). This suggests the equations should reliably predict wave attenuation for one-dimensional geomorphologic modelling with those parameters.

Improving aboveground biomass estimation of natural forests on the Tibetan Plateau using spaceborne LiDAR and machine learning algorithms

• ICESat-2 has the potential to improve the estimation accuracy and efficiency of forest aboveground biomass on a large scale. • The Google Earth Engine cloud platform was used for Sentinel-2 images acquisition and composition. • The optimized extreme learning machine achieved the highest estimation accuracy and reasonable forest aboveground biomass spatial distribution. Natural forests have the most complex structure and richest biodiversity among terrestrial ecosystems and are essential for maintaining the carbon balance and stability of the biosphere. Aboveground biomass (AGB) is a primary indicator used to evaluate forests and can directly measure forest growth and the quality of natural forests. Accurate and rapid AGB estimations can significantly improve the efficiency of forest management and deepen the understanding of the forest carbon cycle. The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2), one of the most recently launched spaceborne light detection and ranging (LiDAR) instruments, can penetrate forest canopies and obtain accurate, large-scale forest vertical parameter measurements. However, the discrete footprints offered by ICESat-2 cannot provide comprehensive spatial AGB distribution data. In this study, an optimized extreme learning machine (ELM) method was proposed to estimate the AGB of natural forests on the eastern Qinghai-Xizang Plateau, China. The synthetic Sentinel-2 images acquired from the Google Earth Engine (GEE) were synergized with ICESat-2 to realize continuous AGB mapping. To verify the effectiveness of the optimized method, support vector machine (SVM), k-nearest neighbor (kNN), and backpropagation (BP) neural network models were also established for comparison. The measured AGB data extracted from the forest management inventory (FMI) were used for model accuracy evaluation. The results show that the optimized ELM achieved the best estimation effect among all the analyzed models, with an R 2 value of 0.68 and a root mean square error (RMSE) value of 25.14 mg/ha. The optimized ELM obtained the minimum RMSE and greatly improved the AGB prediction efficiency. These findings prove the ability of ICESat-2 in estimating the AGB of natural forests, which can provide a new way for large-scale forest resources investigation in high-altitude areas with a harsh environment.

Machine learning algorithms for predicting peak yield in buffaloes using linear traits

Various studies have proved that linear traits have strong relationship with milk productivity but no such models are available for selection of animals based on linear traits. The present study conducted during 2020-22, is an attempt to develop an intelligent model using machine learning algorithms to predict peak milk yield based on its linear traits for selection of best dairy animals. A dataset on 14 linear traits of 259 buffalos across 5 lactations with peak milk yield was created and used for developing models. Data was collected from the buffalos having 8 to 26 kg peak milk yield maintained at Animal Farm Section, Central Institute for Research on Buffaloes, Hisar and also from private farms maintained by farmers. Predictive models were developed using various machine learning algorithms (artificial neural network, support vector regression and random forest) along with multi-linear regression executed on WEKA machine learning platform. Performance of these models was evaluated using evaluation metrics root mean squared error (RMSE). Results revealed that the Artificial Neural Network (ANN) model performed best with minimum RMSE 2.0308. Rear udder height and Lactation number emerged as the two most important attributes affecting the peak milk yield. Such model will be useful and handy for the stakeholders in selection of best dairy animals based on linear traits in absence of authentic record of peak milk yield.

Predicting Crude Oil Price in Nigeria with Machine Learning Models

Background: The rise in crude oil prices yields serious consequences for both oil-producing and non-oil-producing countries. The increase in global commodity prices contributes to the financial income and foreign exchange reserves of oil-exporting countries. However, countries such as Nigeria that sell crude oil and purchase refined fuel confront more complicated situations. To this end, there exists a need to obtain a robust prediction model for the crude oil price of Nigeria. Objective: This study is to determine the best model among the machine-learning time series models considered to predict crude oil prices in Nigeria. Methods: The alternative models were the auto-regressive integrated moving average model, Naive Bayes, Holtwinter trend model, exponential smoothing model, and neural network autoregressive (NNETAR) model. The prediction criteria adopted for model screening were the root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). Daily crude oil prices in dollars obtained from the Central Bank of Nigeria were used for analysis spanning from October 1, 2009 to March 22, 2022 with 2836 data points. Results: The NNETAR model showed the minimum RMSE, MAE, and MAPE for cross-validation sets considered. Conclusion: The NNETAR model was recommended for the prediction of crude oil prices in Nigeria.