Metaheuristic algorithm

Cloud Computing has drastically simplified the management of IT resources by introducing the concept of resource pooling. It has led to a tremendous improvement in infrastructure planning. The major goals of cloud computing include maximization of computing resources with minimization of cost. But the truth is that everything has a price and cloud computing is no different. With Cloud computing there comes a number of security concerns which need to be addressed. Cloud forensics plays a vital role to address the security issues related to cloud computing by identifying, collecting and studying digital evidence in cloud environment. The aim of the research paper is to explore the concept of cloud forensic by applying optimization for feature selection before classification of data on cloud side. The data is classified as malicious and non-malicious using convolutional neural network. The proposed system makes a comparison of models with and without feature selection algorithms before applying the data to CNN. A comparison of different metaheuristics algorithms- Particle Swarm Optimization, Shuffled Frog Leap Optimization and Fire fly algorithm for feature optimization is done based on convergence rate and efficiency.

Financial statements are critical to users, as the increasing fraud cases have left behind irreversible impacts. Hence, this study aims to identify the appropriate financial ratios for fraud risk prediction in the financial statements of companies listed on the Tehran Stock Exchange within the 2014–2021 period. The study is based on data from 180 companies listed on the Tehran Stock Exchange, encompassing a total of 1440 financial statements. To select the most appropriate ratios for fraud risk prediction, all financial ratios were tested by three metaheuristic algorithms, i.e., genetic algorithm, grey wolf optimization, and particle swarm optimization. Metaheuristic and data mining methods were employed for data analysis, and these analyses were conducted using MATLAB R2020a (MATLAB 9.8). According to the research results, the fitness function yielded 0.2708 in particle swarm optimization (PSO). With an accuracy of 72.92% after 19 iterations, PSO was more accurate and converged faster than the other algorithms. It also extracted 11 financial ratios: total debts to total assets, working capital to total assets, stock to current asset, accounts receivables to sales, accounts receivables to total assets, gross income to total assets, net income to gross income, current assets to current debt, cash balance to current debt, retained earnings and loss to equity, and long-term debt to equity. The support vector machine (SVM) classifier was then employed for fraud risk detection at companies through the ratios extracted by the proposed algorithms. The accuracy and precision of financial ratios extracted by PSO and SVM were reported at 80,60% and 71,20%, respectively, which indicates the superiority of the proposed model to other models. Considering that the results obtained from the performance evaluation of financial ratios provided by PSO-SVM demonstrate the capability of this method in predicting the likelihood of fraud in financial statements, it can assist financial statement users. By incorporating these ratios about the performance of the target companies and comparing them with those of other companies, users can make more informed decisions in economic decision-making, investments, credit assessments, and more, ultimately minimizing potential losses and risks.