Comprehensive Analysis of Stock Price Dynamics Using Ensemble Machine Learning Models for Enhanced Prediction Accuracy

Vinamra

doi:https://doi.org/10.54216/AJBOR.120103

Fusion: Practice and Applications | 2025

Feature Selection based on Improved Differential Evolution (DE) Algorithm for E-mail Classification

Spam e-mail has become a pervasive nuisance in today's digital world, posing significant challenges to efficient communication and information dissemination. Dealing with huge amounts of data with irrelevant and redundant features, which leads to high dimension. Nowadays, with the growth of using the internet, finding the secure E-mail classification system for cloud computing is a very important topic. Additionally, determining the best algorithm for choosing a subset of features has a big impact on how well automatic email classification works, making it one of the major issues. Among these is the Differential Evolution (DE) algorithm, which is computationally costly because of the slow convergence rate and evolutionary process. To address these issues, this study offers an intelligent scheme called Opposition Differential Evolution (ODE), which combines the Opposition Based Learning (OBL) and DE algorithms for effective automated feature subset selection. Its effectiveness is assessed using the support vector machine (SVM) to present a strong performance when evaluating the e-mail spam classification rate. Moreover, the OBL is used to accelerate and increase the convergence rate of traditional DE. To determine which features, contribute most to the reliability of the email spam classification, the subset features based on ODE that was used as feature subset selection are used.To assess the effectiveness of the proposed scheme, extensive experiments are conducted on spambase” and “spamassassin” benchmark email datasets, comprising a diverse collection of spam and non-spam emails. The results demonstrate that the Opposition Differential Evolution (ODE) algorithm yields superior performance compared to traditional machine learning and evolutionary techniques, displaying its robustness and efficiency in identifying spam emails accurately. The ODE algorithm effectively handles high-dimensional feature spaces, enhancing the model's discriminatory power while maintaining computational efficiency. Compared to the suggested ODE-SVM technique, which yields a result of 96.79 percent, the full-feature accuracy result was 93.55 percent. Additionally, empirical results demonstrate that our scheme may efficiently increase the number of features needed to improve the accuracy of the email spam classification.

groups

Nadir Omer mail

link https://doi.org/10.54216/FPA.170229

Volume & Issue

Vol. Volume 17 / Iss. Issue 2

Details open_in_new

Fusion: Practice and Applications | 2025

Robust Multimodal Fusion of Transfer Learning Framework for Leukemia Cancer Detection and Classification using Biomedical Images

Leukemia is a form of blood cancer that targets white blood cells (WBC) and stands as a major cause of mortality worldwide. During the center of human bones, leukaemia is presented and provides blood cell generation with leukocytes and WBC, and if some cell comes to be blasted, then it grows a fatal illness. For that reason, the analysis of leukaemia in its initial stages aids significantly in the treatment accompanied by saving the life. At present, leukemia analysis is done by visual assessment of biomedical images of blood cells, which is time-consuming, tedious, and wants to train specialists. Consequently, the lack of an early, automatic, and effectual leukemia recognition model is a major problem in hospitals. A few automated techniques like deep learning (DL) and Machine learning (ML) methodologies for leukemia cancer identification are presented that offer remarkable and effectual results. This study develops a Robust Multimodal Fusion of Transfer Learning Framework for Leukemia Cancer Detection and Classification (RMFTLF-LCDC) algorithm. The RMFTLF-LCDC system mostly suggests identifying and classifying the existence of leukemia cancer on biomedical imaging. At first, the RMFTLF-LCDC model applies image preprocessing using a kernel correlation filter (KCF) to eliminate the noise. For the feature extraction process, the multimodal fusion of CapsNet models, including RES-CapsNet, VGG-CapsNet, and GN-CapsNet are implemented to improve the representation of features by providing more accurate initial information to subsequent capsule layers. In addition, the recurrent spiking neural network with the spiking convolutional block attention module (RSNN-CBAM) technique is performed for the leukemia cancer detection process. At last, the improved Harris hawk optimization (IHHO) approach-based hyperparameter choice can be executed to improve the classification outcomes of the RSNN-CBAM system. The efficiency of the RMFTLF-LCDC method has been validated by comprehensive studies using the benchmark image dataset. The numerical result shows that the RMFTLF-LCDC method has better performance and scalability across other recent techniques.

groups

Arwa Darwish Alzughaibi mail

link https://doi.org/10.54216/FPA.170230

Volume & Issue

Vol. Volume 17 / Iss. Issue 2

Details open_in_new

International Journal of Neutrosophic Science | 2025

Soft Computing with Neutrosophic and fractional order frameworks: A state-of-the-Art review

This study reviews a comprehensive mathematical framework known as neutrosophic soft sets, which combines neutrosophic theory with the soft set theory. Also, we review neutrosophic fractional order functions. For decision making, this framework effectively conveys ambiguity and uncertainty. The developments in soft set theory and neutrosophic set theory are thoroughly examined in this article. We review the advancements of both theories in general. We examine the qualities, applications, and theoretical underpinnings of both theories. We study the combination of neutrosophic soft set theory and logic. The study talks about important new developments and techniques that make neutrosophic soft suites better at solving difficult real-world problems that aren’t always clear. To promote the advancement of the discipline, we also provide a comprehensive overview of the theories derived from literature methodologies, and propose potential avenues for future research. This review serves as an important resource for researchers and practitioners wishing to utilize neutrophil suites in their work. It provides a deeper understanding of the potential effects and applications. This review also addresses a discussion on fractional order neutrosophic sets (FONS). The fractional order component offers an additional degree of freedom, enhancing the adaptability of neutrosophic sets for many applications.

groups

Kottakkaran Sooppy Nisar mail -

Muhammad Farman mail -

Harish Garg mail -

Mahmoud Abdel-Aty mail

link https://doi.org/10.54216/IJNS.250345

Volume & Issue

Vol. Volume 25 / Iss. Issue 3

Details open_in_new

American Journal of Business and Operations Research | 2024

Development of Automated Statistical and Optimized Models with Soft Computing Techniques for Business finance Operations

As part of the scope of the Artificial Neural Network – Particle Swarm Optimization (ANN-PSO) notion, the computational capability of ANNs is integrated with the optimization potential of PSO. This method proves to be very effective in solving complex non-linear forecasting problems where traditional approaches would not be effective. The data interactions that exist are the ones that are modelled and captured by the ANN component. However, the PSO method is charged with the duty of minimizing the biases and weights used in the ANN to ensure that the model attains the global minimum without being trapped in tiny local minimum. The application of this framework can be extended to cash forecast used in business like the one above in which a days of cash requirement forecast is created based on experience and factors like holidays, pay check effects and working days. However, the given contribution of the PSO element in learning process is linked with continuous transformation of variables under the basic guidelines of swarming intelligence, it makes the learning session of ANN more efficient. Therefore, the degree of accuracy of forecasts that are given by such configurations is improved, especially in the conditions that are in a state of steady evolution. The ANN-PSO model mirrors similar attributes, including its ability to process data in parallel and furthermore, its high compatibility with large-scale data as well as it robustness when working with both non-linear and linear data set. Incorporating the PSO into a model enhances the range of possible solutions and given the peculiarity of the gradient-based approach, it reduces mistakes more effectively than the conventional techniques. They suggested that by applying ANN with PSO the framework act as an efficient tool for prediction and for solving various issues in several fields. In this case, the ANN-PSO strategy suggested here works out to an impressive overall accuracy of over 98% compared to the previous systems.

groups

Doug Young Song mail

link https://doi.org/10.54216/AJBOR.110202

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

American Journal of Business and Operations Research | 2024

Stochastic Diffusion Process Models for Driving Innovation in Market-Driven Product Development

In order to analyses the diffusion of new technological products in rapidly changing market environments, this paper presents two new stochastic diffusion models: SDM1 and SDM2. The two models also utilize stochastic market size function in capturing rather random growth of potential users, inherent in most real-world markets. SDM1 apply the exponential distribution to model the market growth rate to consider the cases characterized by the high increase, while SDM2 adapt the Erlang distribution to reflect the S-shape to consider the long-term adoptions. The presented models rely on stochastic differential equations with recourse to calculus, and they adopt stochastic geometric Brownian motion and logistic growth function for adoption rates. This makes it possible to capture effects of learning as well as the non-regularity of adoption over time. The empirical results of benchmark models by using Apple iPhones and Samsung Galaxy smartphones sales data show the better performance of SDM1 and SDM2. The performance of the methodologies is measured using parameters, the goodness-of-fit tests and the forecast accuracy that all show that the proposed methods are very efficient. These models have a rich theoretical background, which comprises the foundation for explaining adoption patterns, which in turn will facilitate the behaviour of managers and policymakers towards understanding consumers, controlling inventory, and designing significant marketing strategies for technology products in a stochastic world. Both SDM1 and SDM2, the suggested algorithms, outperform the state-of-the-art techniques in terms of accuracy. SDM1 outperforms the other models with an accuracy of 95.32 percent. SDM2's greater accuracy in forecasting is shown by its outperformance of all techniques, which stands at 97.3%.

groups

Saurabh Singh mail

link https://doi.org/10.54216/AJBOR.110203

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

American Journal of Business and Operations Research | 2024

PSO-Driven Parametric Estimation and Simulation for Banking Applications Using Soft Computing Approaches

For the purposes of maintaining a healthy liquid balancing and maximizing cash flow, accurate cash forecasting is very necessary for banking operations. In order to overcome the shortcomings of conventional forecasting techniques, such as linear regression, which do not take into account dynamic elements like pay impacts and vacations, this research, offers a Cash Management Model (PSO-CMM) that is based on Particle Swarm Optimization. Taking into account a number of characteristics, such as working days, holiday impacts, and pay patterns, PSO-CMM improves its coefficients for cash prediction. This allows for both short-term and long-term predictions. By swarm intelligence, the model is able to improve the accuracy of its predictions, hence providing greater resilience to continuously modifying surroundings. In addition to the development of linear and hybrid models that combine PSO with artificial neural networks (ANNs), the incorporation of adaptive computing approaches to improve weights is one of the most important advances. Furthermore, in order to prevent local optimums and to promote universal convergence, erratic patterns were incorporated in the most sophisticated systems. The results of this evaluation revealed a significant rise in the accuracy of cash projections. This study presents a comprehensive methodology for predicting cash requirements, which makes it possible for micro financial organizations to get useful insights and improves their operating effectiveness in situations that are always changing. When compared with Normal Data, the suggested PSO-CMM method's overall accuracy is around 91%.

groups

Yogesh Khandokar mail

link https://doi.org/10.54216/AJBOR.110204

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

American Journal of Business and Operations Research | 2024

Optimized Cash Forecasting Models for Banking Applications Using Soft Computing Techniques

The Artificial Neural Network-based Cash Forecasting Model (ANN-CFM) is introduced in this part as one way of mitigating the vices that are characterised with linear approach to financial management. This paradigm is quite helpful when the analysis is focused on non-linear and, generally, troublesome data. ANN-CFM, therefore, simultaneously takes both the linear and non-linear information for improving on the cash forecasting. Due to this fact, it is able to realise and leverage over advantage from the computational competence that neural systems provide. The hidden, output and input layers use randomised initial biased and weights. These include biases together with weighting that is altered regarding a standard basis with the use of a learning strategy to try to find the greatest cash needs. This design is actually composed of three various layering. This is exactly what the ANN-CFM is capable of dealing with and it accepts inputs, both for LT and ST forecasting. Among these inputs, you have factors of the working days, wages’ impact, and the impacts of holidays. The ANN-CFM is a system that revolutionises the way a human would perform his/her decisions and is a highly parallelized and efficient analytical tool for large data. As a result, this results in enhancement of precision to that which is predicted. The kind of architecture used in the system is feed forward neural network, which uses back propagation to help in reducing the numbers of errors that prevail at the time of prediction. In this part, extensive application of ANN, including its ability to learn in environments that may be constantly evolving is also highlighted. Thus, this innovative approach allows for sure receipt of accurate solutions for the management of these funds by companies operating in the financial industry. Comparing it with Normal Data, it is clear that the ANN-CFM technique proposed here provides an overall accuracy of approximately 95%.

groups

Manjeet rege mail

link https://doi.org/10.54216/AJBOR.110205

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

American Journal of Business and Operations Research | 2025

Optimizing Business Process through Fault-Tolerant Scheduling in Cloud Environments: A Comparative Study

The fault tolerance study carried out in this research explores Bidirectional Long Short-Term Memory (LSTM) and Generative Adversarial Networks (GAN) to improve cloud computing dependability and functionality. Being an integral part of the rage for business operations, cloud-computing fundamentals of resource provisioning and fault tolerance have a bearing on the overall cost-dynamics, ROI and OpEx. Reliability covers such issues as hardware failures, configuration problems and other network issues that may have financial implications and even lead to revenue loss, and failure to meet service level agreement (SLA). The work develops a novel GAN-BiLSTM model for the accurate prediction of faults and the enhancement of recovery management, resulting in resource efficiency and cost of capital reduction (CapEx). Evaluation criteria involve deadline guarantee ratio, average task delay, and system scalability, confirming that the proposed model has better financial performance than DPSO and ANFIS. Cutting wastage of resources and increasing energy capacity in a system, the model displays attractive cost reduction and operating effectiveness for cloud service providers. In the simulation, important results of the model are demonstrated in the business continuity, financial risk reduction as well as maintaining accurate and resourceful service in high demand situations. All these developments have placed the fault tolerant systems powered by machine learning as indispensable instruments that can also enhance profitability, resources utilisation and sustainable competitiveness in the cloud computing business.

groups

Anil Audumbar Pise mail

link https://doi.org/10.54216/AJBOR.120101

Volume & Issue

Vol. Volume 12 / Iss. Issue 1

Details open_in_new

American Journal of Business and Operations Research | 2025

Gorilla Troop Optimizer-Driven Fault-Tolerant Scheduling for Cloud-Based Business Workflows

The study proposes a GTO-FTASS (Gorilla Troop Optimizer-Based Fault Tolerant Aware Scheduling Scheme) for improving the reliability and performance in the cloud computing context. Cloud systems are more likely to fail due to the architecture of these layers and dependence on both the hardware and software, therefore require more sophisticated fault-tolerant solutions. The preliminary to this work is the design of an adaptive GTO-FTASS with a fitness function based on two constraints: Expected Time of Completion (ETC) and Failure probability that were derived from the gorilla value system. The approach provides resource utilization and task planning with the provision of fault recovery hence reducing exposure to time loss and operational vulnerability. MGS outperforms several state-of-the-art models, such as MTCT, MAXMIN, ACO, NSGA-II, and DCLCA in terms of makes pan, failure ratio and failure slowdown. Finally, the applicability of experimental validation with various situations and fluctuating intensities demonstrates the scalability of the model and its stability under pressure, decreased failure rates and increased effectiveness of performed tasks. Through the approaches to latency, resource, and error correction, GTO-FTASS is an investment that stewards have to make to cut costs and achieve high performance on clouds. The framework also provides competitive benefit and robustness for cloud enterprising in fluctuating and crucial strategic applications.

groups

Takura Wekwete mail

link https://doi.org/10.54216/AJBOR.120102

Volume & Issue

Vol. Volume 12 / Iss. Issue 1

Details open_in_new

American Journal of Business and Operations Research | 2025

Comprehensive Analysis of Stock Price Dynamics Using Ensemble Machine Learning Models for Enhanced Prediction Accuracy

Stock price prediction is an important component of the financial analysis because the results influence the increase in economic growth and investment. This work aims to develop an ensemble SL technique that consists of mainly PCA, PSO, and SVM to achieve better prediction. Hence, through PCA, large numbers of stocked data dimensions are compressed without compromising on the crucial feature of data set. The problem of parameter selection for non-linear datasets is handled by using a bio-inspired optimization technique known as PSO in order to optimize the SVM hyperparameters. As the core accurate predictor model, the SVM employs the Radial Basis Function to provide the substantial regression capacity for sophisticated financial data sets. The ensemble framework was used with actual stock price data and the information set into training and testing sets. The acknowledgement of probable manifold values indicated that the proposed approach is more accurate than conventional approaches, with an accuracy rate of 95.5 %, when benchmarked using RMSE or MAE. In particular, the forecasts of stock prices by integrating PCA for feature reduction and PSO for parameter tuning with SVM regression is a notable improvement. The proposed methodology can be easily applied to scale for financial analytics since it manages to solve for the issues of noisy and non-linear high dimensional data.

groups

Vinamra Nayak mail

link https://doi.org/10.54216/AJBOR.120103

Volume & Issue

Vol. Volume 12 / Iss. Issue 1

Details open_in_new