Theprediction Accuracy Research Articles

Prediksi Daya Output Sistem Pembangkit Listrik Tenaga Surya (PLTS) Menggunakan Regresi Linear Berganda

<table width="593" border="1" cellspacing="0" cellpadding="0"><tbody><tr><td valign="top" width="388"><p>The power generated by Solar Power Plants (Pembangkit Listrik Tenaga Surya/PLTS) from time to time is fluctuating due to the influence of weather and other external conditions. This study predicts the output power of PLTS Sumalata in North Gorontalo Regency with data analytics on datasets obtained from measurements at 2 plants in PLTS Sumalata. Data analytics to predict the output power of PLTS Sumalata is using a multiple linear regression approach, which is applied by implementing the Cross-industry standard for data mining (CRISP-DM) process model. The tools used are the Weka 3.0 application and Jupyter Notebook with the Python programming language. With data analytics using Weka 3.0 on datasets obtained from measurements at 2 plants in PLTS Sumalata, multiple linear regression equations were obtained as well as evaluation of prediction results using Correlation Coefficient (CC), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Relative Absolute Error (RAE), and Root Relative Squared Error (RRSE). The equation formed from the prediction of the output power in Plant 1 is Y = -22216632810.1123 - 771640073.1888 X<sub>1</sub> + 2349039057.8254 X<sub>2</sub> -25796134709.3552 X<sub>3</sub>. While the equation formed from the prediction of the output power in Plant 2 is Y = -2784.107 + 300.0146 X<sub>1</sub> – 173.7016 X<sub>2</sub> + 21773.3845 X<sub>3</sub>. Based on the test, the correlation coefficient on the Plant 1 dataset is 0.52 and the Plant 2 dataset is 0.92. Those can be concluded that the irradiation data, module temperature, and ambient temperature have a significant effect of 52% on the output power generated in the PLTS system at Plant 1 and 92% on Plant 2. Then the MAE, RMSE, RAE, and RRSE values in the Plant 1 dataset are higher than Plant 2, while the relationship between the independent variables and the dependent variables in the Plant 2 dataset is stronger than the Plant 1 dataset. In order to improve the accuracy of the prediction that can be used for evaluating the performance of the PLTS system, measurement data with a minimum measurement duration of one year is needed to be able to represent seasonal conditions throughout the year, such as the dry season, rainy season, and extreme weather conditions.</p></td></tr></tbody></table>

COVID-19 Diagnosis and Classification Using Radiological Imaging and Deep Learning Techniques: A Comparative Study.

In December 2019, the novel coronavirus disease 2019 (COVID-19) appeared. Being highly contagious and with no effective treatment available, the only solution was to detect and isolate infected patients to further break the chain of infection. The shortage of test kits and other drawbacks of lab tests motivated researchers to build an automated diagnosis system using chest X-rays and CT scanning. The reviewed works in this study use AI coupled with the radiological image processing of raw chest X-rays and CT images to train various CNN models. They use transfer learning and numerous types of binary and multi-class classifications. The models are trained and validated on several datasets, the attributes of which are also discussed. The obtained results of various algorithms are later compared using performance metrics such as accuracy, F1 score, and AUC. Major challenges faced in this research domain are the limited availability of COVID image data and the high accuracy of the prediction of the severity of patients using deep learning compared to well-known methods of COVID-19 detection such as PCR tests. These automated detection systems using CXR technology are reliable enough to help radiologists in the initial screening and in the immediate diagnosis of infected individuals. They are preferred because of their low cost, availability, and fast results.

Multi-step Time Series Forecasting Based on Informer-XGBoost-GA

Multi-step prediction of time series has great significance in practical application. To improve the accurate prediction of multistep time series, time series prediction models have good long-time dependence and are able to analyze the correlation between information in time series. In this paper, a multi-step prediction model on the basis of Informer-XGBoost-GA was proposed, first establishing a combination of Informer and XGBoost, and then optimizing the weight of the combined model through GA. By comparing the experimental results, it can be seen that the proposed that the proposed prediction model significantly outperforms the comparison model and raise the accuracy of the prediction with advantage.

Forecasting Directional Movement of Stock Prices using Deep Learning

Stock market’s volatile and complex nature makes it difficult to predict the market situation. Deep Learning is capable of simulating and analyzing complex patterns in unstructured data. Deep learning models have applications in image recognition, speech recognition, natural language processing (NLP), and many more. Its application in stock market prediction is gaining attention because of its capacity to handle large datasets and data mapping with accurate prediction. However, most methods ignore the impact of mass media on the company’s stock and investors’ behaviours. This work proposes a hybrid deep learning model combining Word2Vec and long short-term memory (LSTM) algorithms. The main objective is to design an intelligent tool to forecast the directional movement of stock market prices based on financial time series and news headlines as inputs. The binary predicted output obtained using the proposed model would aid investors in making better decisions. The effectiveness of the proposed model is assessed in terms of accuracy of the prediction of directional movement of stock prices of five companies from different sectors of operation.

Crankshaft HCF Research Based on the Simulation of Electromagnetic Induction Quenching Approach and a New Fatigue Damage Model

In recent decades, the electromagnetic induction quenching approach has been widely applied in the surface treatment process of steel engine parts such as crankshafts. In this paper, the strengthening effect of this approach was selected to be the object of study. First, the multi-physics coupling phenomenon was established by a 3D finite element simulation approach. Then, the fatigue property of the crankshaft was predicted based on the combination of the residual stress field obtained in the previous step and a chosen multi-axial fatigue damage model. Finally, a corresponding experiment verification was carried out to check the accuracy of the prediction. The results showed that the method proposed by this paper can provide high enough accuracy in predicting the fatigue property of two types of commonly used steel crankshafts.

Electric Vehicle Accident Alert System

Electric Vehicle Accident Alert System is a Machine Learning Integrated, IoT based real-time alerting system, operating on ESP8266 microcontroller. Our designed system acts as an alerting system in its truest sense. It uses KNN classification algorithm to analyse the GPS location of the vehicle and generate a prediction indicating whether the vehicle is in an accident-prone area or not. And, if the accident occurs, the Electric Vehicle Accident Alert System sends a SOS alert message with a link to google maps (for directions), to the emergency contact saved in the system. The whole system is based on a plug n play concept, i.e., the system would be ready to use once powered up. To provide this agility and flexibility, we have designed a registration web portal for the device which registers the user and their device with a unique UID on the Cloud. After successful registration, the system becomes ready and can be simply fitted inside a vehicle to be used. Moreover,the ML algorithm saves those GPS coordinates in its dataset to further improve the accuracy of the prediction. The whole IoT stack integrated with ML algorithm, Web Portal and a Cloud server (implemented on Google Firebase), makes our project a self-improving, agile and a user-friendly system.

Using Deep Learning to Fill Data Gaps in Environmental Footprint Accounting.

Environmental footprint accounting relies on economic input-output (IO) models. However, the compilation of IO models is costly and time-consuming, leading to the lack of timely detailed IO data. The RAS method is traditionally used to predict future IO tables but suffers from doubts for unreliable estimations. Here we develop a machine learning-augmented method to improve the accuracy of the prediction of IO tables using the US summary-level tables as a demonstration. The model is constructed by combining the RAS method with a deep neural network (DNN) model in which the RAS method provides a baseline prediction and the DNN model makes further improvements on the areas where RAS tended to have poor performance. Our results show that the DNN model can significantly improve the performance on those areas in IO tables for short-term prediction (one year) where RAS alone has poor performance, R2 improved from 0.6412 to 0.8726, and median APE decreased from 37.49% to 11.35%. For long-term prediction (5 years), the improvements are even more significant where the R2 is improved from 0.5271 to 0.7893 and median average percentage error is decreased from 51.12% to 18.26%. Our case study on evaluating the US carbon footprint accounts based on the estimated IO table also demonstrates the applicability of the model. Our method can help generate timely IO tables to provide fundamental data for a variety of environmental footprint analyses.

Endoplasmic Reticulum Stress-Related Signature Predicts Prognosis and Drug Response in Clear Cell Renal Cell Carcinoma.

Clear cell renal cell carcinoma (ccRCC) is the most common type of kidney cancer. The maximum number of deaths associated with kidney cancer can be attributed to ccRCC. Disruption of cellular proteostasis results in endoplasmic reticulum (ER) stress, which is associated with various aspects of cancer. It is noteworthy that the role of ER stress in the progression of ccRCC remains unclear. We classified 526 ccRCC samples identified from the TCGA database into the C1 and C2 subtypes by consensus clustering of the 295 ER stress-related genes. The ccRCC samples belonging to subtype C2 were in their advanced tumor stage and grade. These samples were characterized by poor prognosis and malignancy immune microenvironment. The upregulation of the inhibitory immune checkpoint gene expression and unique drug sensitivity were also observed. The differentially expressed genes between the two clusters were explored. An 11-gene ER stress-related prognostic risk model was constructed following the LASSO regression and Cox regression analyses. In addition, a nomogram was constructed by integrating the clinical parameters and risk scores. The calibration curves, ROC curves, and DCA curves helped validate the accuracy of the prediction when both the TCGA dataset and the external E-MTAB-1980 dataset were considered. Moreover, we analyzed the differentially expressed genes common to the E-MTAB-1980 and TCGA datasets to screen out new therapeutic compounds. In summary, our study can potentially help in the comprehensive understanding of ER stress in ccRCC and serve as a reference for future studies on novel prognostic biomarkers and treatments.

Development and validation of a meta-learning-based multi-modal deep learning algorithm for detection of peritoneal metastasis.

The existing medical imaging tools have a detection accuracy of 97% for peritoneal metastasis(PM) bigger than 0.5cm, but only 29% for that smaller than 0.5cm, the early detection of PM is still a difficult problem. This study is aiming at constructing a deep convolution neural network classifier based on meta-learning to predict PM. Peritoneal metastases are delineated on enhanced CT. The model is trained based on meta-learning, and features are extracted using multi-modal deep Convolutional Neural Network(CNN) with enhanced CT to classify PM. Besides, we evaluate the performance on the test dataset, and compare it with other PM prediction algorithm. The training datasets are consisted of 9574 images from 43 patients with PM and 67 patients without PM. The testing datasets are consisted of 1834 images from 21 testing patients. To increase the accuracy of the prediction, we combine the multi-modal inputs of plain scan phase, portal venous phase and arterial phase to build a meta-learning-based multi-modal PM predictor. The classifier shows an accuracy of 87.5% with Area Under Curve(AUC) of 0.877, sensitivity of 73.4%, specificity of 95.2% on the testing datasets. The performance is superior to routine PM classify based on logistic regression (AUC: 0.795), a deep learning method named ResNet3D (AUC: 0.827), and a domain generalization (DG) method named MADDG (AUC: 0.834). we proposed a novel training strategy based on meta-learning to improve the model's robustness to "unseen" samples. The experiments shows that our meta-learning-based multi-modal PM predicting classifier obtain more competitive results in synchronous PM prediction compared to existing algorithms and the model's improvements of generalization ability even with limited data.

Virtual Screening of Drug Proteins Based on the Prediction Classification Model of Imbalanced Data Mining

We propose a virtual screening method based on imbalanced data mining in this paper, which combines virtual screening techniques with imbalanced data classification methods to improve the traditional virtual screening process. First, in the actual virtual screening process, we apply k-means and smote heuristic oversampling method to deal with imbalanced data. Meanwhile, to enhance the accuracy of the virtual screening process, a particle swarm optimization algorithm is introduced to optimize the parameters of the support vector machine classifier, and the concept of ensemble learning is brought in. The classification technique based on particle swarm optimization, support vector machine and adaptive boosting is used to screen the molecular docking conformation to improve the accuracy of the prediction. Finally, in the experimental construction and analysis section, the proposed method was validated using relevant data from the protein data bank database and PubChem database. The experimental results indicated that the proposed method can effectively improve the accuracy of virus screening and has practical guidance for new drug development. This research regards virtual screening as a problem of imbalanced data classification, which has obvious guiding significance and also provides a certain reference for the problems faced by virtual screening technology.

A Delayed Detached Eddy Simulation Model for the Simulation of Complex Turbulent Flow

A new turbulent model based on Delayed Detached Eddy Simulation (DDES) with non-linear eddy viscosity model (NLEVM) was developed to predict the complex turbulent flow. The numerical simulation of the triangular cylinder and the centrifugal pump was carried out to investigate the ability and applicability of the DDES model based on NLEVM (DDES_NL). Compared to the turbulent model based on the eddy viscosity model, the computational results of the triangular cylinder showed the advantage of the non-linear eddy viscosity modification in the DDES_NL model which can improve the accuracy of the prediction in the flow phenomenon with a relatively simple turbulence structure. Regrettably, some small-scale turbulent structures among those still cannot be captured accurately. The numerical simulation of the centrifugal pump predicted by the DDES_NL model shows more abundant flow structures and gets close to the realistic statistical characteristics. It also proves the good applicability of the DDES_NL model in complex flow. This study aims to contribute to the growing area of turbulence modeling by exploring it.

A novel pyroptosis-related lncRNA prognostic signature associated with the immune microenvironment in lung squamous cell carcinoma

BackgroundA growing body of evidence suggests that pyroptosis-related lncRNAs (PRncRNAs) are associated with the prognoses of tumor patients and their tumor immune microenvironments. However, the function of PRlncRNAs in lung squamous cell carcinoma (LUSC) remains unclear.MethodsWe downloaded the transcriptome and clinical information of 551 LUSC samples from the The Cancer Genome Atlas (TCGA) database and randomly separated patients with complete information into two cohorts. Based on the training cohort, we developed a pyroptosis-related signature. We then examined the signature in the test cohort and all retained patients. We also clustered two risk groups in each cohort according to the signature and performed survival analysis, functional analysis, tumor immune microenvironment analysis and drug sensitivity analysis.ResultsA prognostic signature containing five PRlncRNAs (AP001189.1, PICART1, LINC02555, AC010422.4, and AL606469.1) was developed. A principal component analysis (PCA) indicated better differentiation between patients with different risk scores. Kaplan–Meier (K–M) analysis demonstrated poorer survival among patients with higher risk scores (P < 0.001). A receiver operating characteristic (ROC) curve analysis provided evidence confirming the accuracy of the signature, and univariate (p = 0.005) and multivariate (p = 0.008) Cox regression analyses confirmed the independent value of the risk score in prognoses. Clinical subgroup validation indicated that the signature was more suitable for patients with early-stage LUSC. We also created a nomogram to increase the accuracy of the prediction. Moreover, functional analysis revealed that pathways related to tumor development and pyroptosis were enriched in the high-risk group. Furthermore, the prognostic signature was proven to be a predictor of sensitivity to immunotherapy and chemotherapy.ConclusionsWe developed a novel pyroptosis-associated signature with independent value for the prognosis of LUSC patients. PRlncRNAs are closely associated with the tumor immune microenvironment in LUSC and might offer new directions for immunotherapy.

Macroeconomic attention and stock market return predictability

Our investigation evaluates the novel macroeconomic attention indices (MAI) of Fisher et al. (2022) in terms of their ability to predict stock market returns based on dimension reduction methods and shrinkage methods. Our results demonstrate that macroeconomic attention indices can predict stock market returns with a significant degree of accuracy. In addition, the components of MAI indices based on partial least squares (PLS) and the least absolute shrinkage and selection operator (LASSO) methods have a greater capacity to improve the accuracy of the prediction of stock market returns than the components of the traditional macroeconomic variables. Moreover, we find that shrinkage methods can generate performances superior to those of the other models for forecasting stock market returns. We further demonstrate that macroeconomic attention indices embody superior predictive ability during the COVID-19 pandemic and over longer periods of time. Our study sheds new light on stock market returns’ prediction from the perspective of macroeconomic fundamentals.

Performance Prediction of Cement Stabilized Soil Incorporating Solid Waste and Propylene Fiber.

Cement stabilized soil (CSS) yields wide application as a routine cementitious material due to cost-effectiveness. However, the mechanical strength of CSS impedes development. This research assesses the feasible combined enhancement of unconfined compressive strength (UCS) and flexural strength (FS) of construction and demolition (C&D) waste, polypropylene fiber, and sodium sulfate. Moreover, machine learning (ML) techniques including Back Propagation Neural Network (BPNN) and Random Forest (FR) were applied to estimate UCS and FS based on the comprehensive dataset. The laboratory tests were conducted at 7-, 14-, and 28-day curing age, indicating the positive effect of cement, C&D waste, and sodium sulfate. The improvement caused by polypropylene fiber on FS was also evaluated from the 81 experimental results. In addition, the beetle antennae search (BAS) approach and 10-fold cross-validation were employed to automatically tune the hyperparameters, avoiding tedious effort. The consequent correlation coefficients (R) ranged from 0.9295 to 0.9717 for BPNN, and 0.9262 to 0.9877 for RF, respectively, indicating the accuracy and reliability of the prediction. K-Nearest Neighbor (KNN), logistic regression (LR), and multiple linear regression (MLR) were conducted to validate the BPNN and RF algorithms. Furthermore, box and Taylor diagrams proved the BAS-BPNN and BAS-RF as the best-performed model for UCS and FS prediction, respectively. The optimal mixture design was proposed as 30% cement, 20% C&D waste, 4% fiber, and 0.8% sodium sulfate based on the importance score for each variable.

Analytical and Numerical Investigation of Embedded Through-section GFRP-Strengthened RC Beams with a Developed Bonding-based Model

This study analytically and numerically investigated the shear performance of rectangular concrete beams strengthened with embedded through-section (ETS) glass fiber-reinforced polymer (GFRP) bars. The key features of the bonding-based approach were first formulated and then validated by comparing their mechanism with the actual shear resistance mechanism of a strengthening system. Thereafter, a finite element method (FEM) incorporating the bonding-based approach was used to simulate the shear behaviors of ETS-GFRP-strengthened beams. Conversely, the shear crack angle model for ETS-GFRP-strengthened beams was empirically developed to enhance the accuracy of the prediction. The experimental data recently investigated by the authors regarding pullout tests and beam tests were utilized to verify the numerical and analytical results. The analyses indicate that the FE model coupled with the bonding-based approach is a universal and powerful method for simulating the behavior of ETS-GFRP-strengthened beams with and without anchorage. The results demonstrate the great potential of the analytical model, established on the basis of the bond mechanism, for assessment of the shear resistance of ETS-GFRP-strengthened beams.

A novel type curve for estimating oil recovery factor of gas flooding

A novel type curve is presented for oil recovery factor prediction suitable for gas flooding by innovatively introducing the equivalent water-gas cut to replace the water cut, comprehensively considering the impact of three-phase flow (oil, gas, water), and deriving the theoretical equations of gas flooding type curve based on Tong’s type curve. The equivalent water-gas cut is the ratio of the cumulative underground volume of gas and water production to the total underground volume of produced fluids. Field production data and the numerical simulation results are used to demonstrate the feasibility of the new type curve and verify the accuracy of the prediction results with field cases. The new type curve is suitable for oil recovery factor prediction of both water flooding and gas flooding. When a reservoir has no gas injected or produced, the gas phase can be ignored and only the oil and water phases need to be considered, in this case, this gas flooding type curve returns to the Tong’s type curve, which can evaluate the oil recovery factor of water flooding. For reservoirs with equivalent water-gas cuts of 60%—80%, the regression method of the new type curve works well in predicting the oil recovery factor. For reservoirs with equivalent water-gas cuts higher than 80%, both the regression and assignment methods of the new type curve can accurately predict the oil recovery factor of gas flooding.

Application of Machine Learning in Enterprise Risk Management

With the development of China’s economic globalization, more and more enterprises have realized the importance of risk management. As a new technology, machine learning has injected new vitality into enterprise risk management. This paper analyzes the factors of enterprise risk from the perspective of risk management, explores and analyzes the application of machine learning in enterprise risk management, and points out the direction for scientific enterprise risk management. This paper first establishes a set of credit evaluation factors and then builds a model based on this evaluation factor set, which ensures the accuracy of the evaluation and also adds pertinence. With the increase of the number of samples, the accuracy of the prediction results also increases. Also, basically, with the increase of training samples and realistic samples, the accuracy rate continues to increase. In the data of the experimental results, it is shown in the process of increasing the number of samples from 400 to 600. The accuracy of the traditional method is 0.81 and 0.8316, respectively. The accuracy of the improved method is 0.842 and 0.905, respectively. This paper studies the process of enterprise risk assessment from the perspective of risk management and attempts to build an enterprise risk management model based on industry risk coefficients, which have certain practical significance.

The Effect of Speed-Varying Micro-Cutting Tool Dynamics on Stability During High-Speed Micromilling of Ti6Al4V

Abstract Chatter-free machining is necessary in micromilling to avoid the catastrophic failure of micro-end mill. The accuracy of the prediction of chatter-free machining conditions in high-speed micromilling has been improved in this work by including speed-varying micro-end mill dynamics. An optimum design of exponential window has been devised to remove the unwanted spindle dynamics from the displacement signal to construct the speed-dependent frequency response function (FRF) of micro-end mill. The stiffness of the micro-end mill has been found to be increasing with increase in spindle speed and the natural frequency of the micro-end mill has been found to be changing with change in spindle speeds. The cutting velocity-chip load-dependent cutting coefficients have been included to predict the stability using Nyquist criterion. The predicted stability lobe with speed-varying micro-end mill dynamics has increased chatter-free depth of cut significantly compared to the chatter-free depth of cut predicted with static micro-end mill dynamics. The increase in depth of cut with speed-varying dynamics has been found to be 28% at 20,000 rpm, 150% at 52,000 rpm, and 250% at 70,000 rpm. A critical value of acceleration of the workpiece has been identified for chatter onset detection and it has been validated with machined surface image analysis. The magnitude of acceleration in both feed and normal to feed direction has been characterized to analyze the effect of spindle speed and depth of cut on the vibration of workpiece.

A MRI-based radiomics model predicting radiation-induced temporal lobe injury in nasopharyngeal carcinoma.

To develop and validate a radiomics-based model for predicting radiation-induced temporal lobe injury (RTLI) in nasopharyngeal carcinoma (NPC) by pretreatment MRI of the temporal lobe. A total of 216 patients with diagnosed NPC were retrospectively reviewed. Patients were randomly allocated to the training (n = 136) and the validation cohort (n = 80). Radiomics features were extracted from pretreatment contrast-enhanced T1- or fat-suppressed T2 weighted MRI. A radiomics signature was generated by the least absolute shrinkage and selection operator (LASSO) regression algorithm, Pearson correlation analysis, and univariable logistic analysis. Clinical features were selected with logistic regression analysis. Multivariable logistic regression analysis was conducted to develop three models for RTLI prediction in the training cohort: namely radiomics signature, clinical variables, and clinical-radiomics parameters. A radiomics nomogram was used and assessed with respect to calibration, discrimination, reclassification, and clinical application. The radiomics signature, composed of two radiomics features, was significantly associated with RTLI. The proposed radiomics model demonstrated favorable discrimination in both the training (AUC, 0.89) and the validation cohort (AUC, 0.92), outperforming the clinical prediction model (p < 0.05). Combining radiomics and clinical features, higher AUCs were achieved (AUC, 0.93 and 0.95), as well as a better calibration and improved accuracy of the prediction of RTLI. The clinical-radiomics model showed also excellent performance in predicting RTLI in different clinical-pathologic subgroups. A radiomics model derived from pretreatment MRI of the temporal lobe showed persuasive performance for predicting radiation-induced temporal lobe injury in nasopharyngeal carcinoma. • Radiomics features from pretreatment MRI are associated with radiation-induced temporal lobe injury in nasopharyngeal carcinoma. • The radiomics model shows better predictive performance than a clinical model and was similar to a clinical-radiomics model. • A clinical-radiomics model shows excellent performance in the prediction of radiation-induced temporal lobe injury in different clinical-pathologic subgroups.

Machine Learning Approach for Big-Mart Sales Prediction Framework

The amounts of data predicted to increase at an exponential rate in the future. The modifications are essential to meet transaction speeds as well as the anticipated growth in data and customer behaviors. The information derived from prior data is extensively relied upon by the majority of companies. One of the primary goals of the suggested system is to identify a reliable sales trend prediction mechanism that is executed using machine learning techniques in order to maximize income. Sales forecasting advises managers about how to manage a company's employees, working capital and assets. It's a requirement for strategic planning and decision-making in the corporate world. Reasonable forecasts enable the company to increase market growth while increasing revenue generating. Operations, marketing, sales, production, and finance all use sales predictions as inputs in their decision-making processes. The concept of sales data and sales forecast has been examined in the suggested system. Machine learning algorithms such as GLL (Generalized Linear Model), GBT (Gradient Boosted Trees), and Decision Trees were used to develop the model, and the optimum model for prediction was established based on the results analysis. A best-fit prediction model for anticipating sales trends is offered based on a performance review. The effectiveness and accuracy of the prediction and forecasting approaches used are discussed in the findings. The Gradient Boost Algorithm has been demonstrated to be the best fit model for forecasting and predicting future sales. The sales projection is done using Gradient Boosted Trees, which predicts which product will be sold in what quantity in the future.