American Journal of Business and Operations Research

Inflation, an omnipresent economic phenomenon, is marked by a continual upsurge in the overall price levels of commodities and services within an economy. Accurately predicting inflation within a data-abundant setting poses a formidable challenge and remains a dynamic area of research encompassing several unresolved methodological inquiries. Among these, a significant query pertains to the identification and extraction of data offering the highest predictive capability for a targeted variable, particularly in scenarios characterized by numerous closely interconnected predictors, as encountered in the context of inflation prediction. Recently, the application of machine learning (ML) models has gained traction in predicting inflation parameters. The predictive accuracy of such models hinges significantly on the selection of an appropriate framework. Ensemble models, designed to amalgamate multiple base models, have emerged as a compelling strategy to yield superior predictive outcomes. In this study, we introduce a novel weighted average ensemble model tailored for the prognostication of inflation prediction. The proposed approach leverages three foundational base models: Linear Regression (LR), Polynomial Regression (PR), and Moving Average (MA) regression. The critical aspect of this ensemble lies in optimizing the weights assigned to each base model, thereby accentuating their individual strengths. To achieve this, we employ the Waterwheel Plant Optimization Algorithm (WWPA), a proficient optimization algorithm, to discern the optimal weight distribution for the base models. Comparative evaluations are conducted, pitting the proposed model against three another base models. Empirical findings conclusively demonstrate the superiority of the proposed weighted average ensemble model, underscoring its capacity to predict inflation with exceptional efficiency.

To get around the drawbacks of conventional classification algorithms that required manual feature extraction and the high computational cost of neural networks, this paper introduces a deep convolutional neural network with multiple instance learning approaches, namely dynamic max pooling and sparse representation. For the categorization of tuberculosis lung illness, this model combines deep convolutional neural networks and multiple instance learning. The design was composed of four phases: pre-processing, instance production, feature extraction, and classification. To perform feature extraction, a model based on a customized version of the VGG16 architecture was trained from scratch. Multiple instance learning techniques such as Diverse Density (DD) and the Maximum pattern bag formulation of the Support Vector Machine were used to evaluate how well the proposed classification algorithm performed in comparison (SVM).The numerical findings demonstrated that the new method offered a higher level of accuracy than the methods that had been used in the past. When evaluating the efficacy of the current method, accuracy, specificity, sensitivity, and error rate were all taken into consideration. The accuracy of the max-pooling based framework and the sparse representation framework was found to be greater than that of the other multiple instance strategies, coming in at 91.51% and 89.84%, respectively, when compared to that of the other methods. The improved accuracy of the present system that makes use of deep neural networks is mostly attributable to the contributions made by features such as transfer learning and automatic feature extraction.

This paper presents a comparative analysis of traditional forecasting methods and machine learning (ML) techniques for sales prediction in e-commerce.  We first review the literature on both traditional and ML methods for sales prediction in e-commerce, highlighting their strengths and weaknesses. The study uses a dataset of daily sales from an e-commerce retailer to conduct a comprehensive empirical study thar compares the performance of literature methods from both categories. The analysis considers different forecasting horizons and evaluates the accuracy of the predictions using various performance metrics, such as mean absolute error and mean squared error. The study finds that ML techniques generally outperform traditional methods, especially for longer forecasting horizons. However, some traditional methods, such as the Holt-Winters method, can also perform well under certain conditions. Our study provides insights into the relative strengths and weaknesses of traditional and ML methods for sales prediction in e-commerce and can guide practitioners in selecting appropriate methods for their specific requirements.

Credit risk assessment is a critical task for financial institutions to determine the creditworthiness of their potential customers. Business intelligence (BI) and machine learning (ML) techniques have gained popularity in recent years as effective tools for credit risk assessment. In this paper, we propose a decision support system (DSS) for credit risk assessment that integrates BI and ML techniques. The proposed DSS employs BI tools to extract and transform data from various sources, and ML techniques to analyze the data and generate predictive models for credit risk assessment. We evaluate the proposed DSS using a real-world dataset of a financial institution. The results show that the proposed DSS achieves a high level of accuracy in credit risk assessment. The results showed that the system was able to accurately predict credit risk, with an accuracy of 88%. The system also outperformed traditional credit scoring models, which highlights the potential of our system for credit risk assessment. The system provides decision-makers with actionable insights to make informed decisions, thereby reducing the risk of default and increasing the profitability of the financial institution.

In the current time, global warming has compelled the automotive vehicle tech sector to undertake a paradigm shift from internal combustion engines that are fueled by fossil fuels to electrical motors that are used for traction instead. It has become an important problem to evaluate BEV options in a thorough manner from the perspective of the consumer because of the recent fast expansion that the BEV industry has seen. This evaluation may be carried out by looking at the fundamental characteristics of every BEV. In addition, effective tools for making the correct choice on the purchase of a BEV are those that use multiple criteria decision making (MCDM). The selection process of BEVs involves vague and uncertainty problem, so that, this work aims to introduce a new multi-criteria decision-making model based on the neutrosophic sets and TOPSIS method to overcome this problem.  The results concluded that the proposed model could handle unclear information and uncertainty which exist usually in the sekection process and present an effective model to rank and select best BEVs.

While it is still in its infancy in comparison to other forms of renewable technology, there is a growing amount of interest and backing for wave energy as a potentially useful renewable resource that could replace a portion of the existing energy supply. In the context of sustainable development, the choice of technology represents a multi-criterion decision-making (MCDM) challenge that may affect the competitive advantages enjoyed by an organization or a nation. The purpose of this study is to evaluate the many wave energy technologies that are now in use as possible choices for green and sustainable technologies that may be used in the seas and oceans. However, requirements like ecological, financial, and technological factors that are based on the fundamental idea of sustainability calls for unclear or unreliable expert assessments that can be solved using single-valued neutrosophic sets (SVNSs). Because of this, the selection of sustainable wave energy technology requires the creation of a one-of-a-kind framework that can analyze both clear and ambiguous data simultaneously without sacrificing any of the information in either category. This study developed a framework that uses measurement alternatives and ranking based on compromise solution (MARCOS) within the context of SVNSs to assist decision-makers in the process of resolving real-time energy problems. An application of the process of selecting the wave energy technology is taken into consideration here as a means of illustrating how applicable the suggested framework is.