Search Results for machine-learning

Abstract

This study aimed to evaluate the financial sustainability of Iraqi industrial companies using machine learning models and to compare their performance with traditional statistical models. Five key financial variables were analyzed: Working Capital Ratio (WCR), Current Ratio (CR), Return on Equity (ROE), Return on Assets (ROA), and Asset Turnover (AT). A Financial Sustainability Index (FSI) was constructed to classify companies. The results showed that WCR and CR were the most influential variables in explaining financial sustainability, while AT had the least impact. All variables demonstrated statistically significant relationships with the FSI. Moreover, machine learning models—particularly XGBoost and Random Forest—outperformed the traditional logistic regression model in predictive accuracy, with variations in variable importance rankings across models. The findings highlight the importance of focusing on working capital management and liquidity improvement to enhance financial sustainability, as well as the use of machine learning tools in financial forecasting to support more accurate strategic decision-making. This study contributes by providing a modern analytical framework to support the financial sustainability of industrial companies, with the potential for future application in other sectors. Specifically, machine learning models improved predictive accuracy by approximately 9.4% over the traditional logistic regression model, while SHAP-based analysis confirmed that WCR and CR carry the greatest weight in predicting financial sustainability. Accordingly, the study recommends that industrial companies prioritize working capital and liquidity management, that financial analysts and investors adopt machine learning models such as XGBoost and Random Forest in financial sustainability forecasting, that more comprehensive sustainability indices combining financial and operational performance be developed, and that decision-makers invest in training on machine learning tools and SHAP interpretation to support more reliable, evidence-based financial decisions.

Abstract

This study presents a machine learning-based framework for predicting and analyzing student academic performance using behavioral, psychological, demographic, and digital learning features. The dataset consists of 14,003 student records collected from higher education institutions in 2024 and obtained from Kaggle. It includes 16 variables covering study behaviors, engagement patterns, learning resources, psychological factors, and demographic characteristics. The target variable (response) was Final Grade, which was categorized into four performance levels: Low / Fail / Poor performance, below average performance, Good / Average performance, and Excellent / High performance, representing overall academic performance. Three machine learning algorithms XGBoost (Extreme Gradient Boosting), CatBoost (Categorical Boosting), and Support Vector Machine (SVM) were applied to classify student performance levels. The models were evaluated using confusion matrices, accuracy, F1-score, MCC, R², RMSE, MAE, and feature importance analysis. Results show that XGBoost achieved the highest performance (99.96% accuracy), followed by CatBoost (99.82%) and SVM (98.96%). Analysis of the importance analysis revealed that Exam Score is the most influential predictor across all models. The study highlights the effectiveness of advanced machine learning techniques in predicting academic success and supporting data-driven educational decision-making.

Abstract

This study aimed to evaluate the financial sustainability of Iraqi industrial companies using machine learning models and to compare their performance with traditional statistical models. Five key financial variables were analyzed: Working Capital Ratio (WCR), Current Ratio (CR), Return on Equity (ROE), Return on Assets (ROA), and Asset Turnover (AT). A Financial Sustainability Index (FSI) was constructed to classify companies. The results showed that WCR and CR were the most influential variables in explaining financial sustainability, while AT had the least impact. All variables demonstrated statistically significant relationships with the FSI. Moreover, machine learning models—particularly XGBoost and Random Forest—outperformed the traditional logistic regression model in predictive accuracy, with variations in variable importance rankings across models. The findings highlight the importance of focusing on working capital management and liquidity improvement to enhance financial sustainability, as well as the use of machine learning tools in financial forecasting to support more accurate strategic decision-making. This study contributes by providing a modern analytical framework to support the financial sustainability of industrial companies, with the potential for future application in other sectors. Specifically, machine learning models improved predictive accuracy by approximately 9.4% over the traditional logistic regression model, while SHAP-based analysis confirmed that WCR and CR carry the greatest weight in predicting financial sustainability. Accordingly, the study recommends that industrial companies prioritize working capital and liquidity management, that financial analysts and investors adopt machine learning models such as XGBoost and Random Forest in financial sustainability forecasting, that more comprehensive sustainability indices combining financial and operational performance be developed, and that decision-makers invest in training on machine learning tools and SHAP interpretation to support more reliable, evidence-based financial decisions.

Abstract

Breast cancer is one of the most prevalent and life-threatening diseases worldwide, making early and accurate diagnosis essential for improving patient outcomes. This study aims to evaluate the performance of two machine learning models, Extreme Gradient Boosting (XGBoost) and Gaussian Naive Bayes, for breast cancer classification using the Breast Cancer Wisconsin Diagnostic dataset. A comparative analysis was conducted based on several performance metrics, including Accuracy, Precision, Recall, F1-score, Specificity, Receiver Operating Characteristic (ROC) curve, Area Under the Curve (AUC), Balanced Accuracy, and Confusion Matrix. In addition, feature importance analysis was performed to identify the most influential variables contributing to the classification process. The experimental results demonstrated that both models achieved high classification performance. However, the Gaussian Naive Bayes model showed slightly superior performance, achieving an Accuracy of 96.5% and an AUC of 0.997, compared with 96.0% and 0.993, respectively, for the XGBoost model. Furthermore, the analysis revealed that features related to tumor morphology, particularly concave points mean, perimeter worst, and radius worst, were among the most influential predictors of breast cancer classification. The findings indicate that both machine learning models provide reliable and effective tools for supporting early breast cancer diagnosis, with Gaussian Naive Bayes demonstrating a slight advantage for the dataset used in this study. These results highlight the potential of machine learning techniques to enhance diagnostic accuracy and support clinical decision-making.

Abstract

This study aims to compare and improve the methods of building investment portfolios for a sample of Iraqi banks listed on the Iraq Stock Exchange, by comparing traditional methods such as the Markowitz model with modern techniques based on machine learning. The Markowitz model is key to balancing return and risk across the medium-variance optimization framework, a traditional model that many financial institutions rely on. The study focused on exploring the extent to which machine learning techniques such as key component analysis (PCA), supporting vector machine (SVM), logistic regression, and random forest can improve the performance of the investment portfolios of these banks in a volatile environment such as the Iraq Stock Exchange. These techniques rely on processing and analyzing huge financial data to discover hidden patterns and relationships that help increase returns and reduce risk more effectively compared to traditional methods. The historical financial data related to the shares and assets of the banks of the research sample in the Iraq Stock Exchange was used to evaluate the performance of portfolios according to indicators such as expected return, variance, and Sharpe ratio. The study aims to provide innovative solutions that help banks make smarter and more effective investment decisions, commensurate with the local market conditions and the economic and political challenges they face.     

Abstract

This research aims to provide a theoretical and applied framework for employing machine learning algorithms in management accounting and costing systems.

The research focuses on the importance of improving resource consumption monitoring, accurately tracking cost behavior, identifying unutilized energy, and supporting decision-making through historical data analysis to enhance the accuracy of production reports.

To achieve the research objective, a descriptive approach was adopted, drawing on available studies. A field study was also used, using a questionnaire to collect data from the research sample (the Electrical Cables and Wires Factory - Ur General Company).

The research also reached a number of conclusions, most notably that employing machine learning algorithms contributes to improving the prediction of quantitative resource consumption, which helps detect deviations and identify their potential causes, and enhances the accuracy and comprehensiveness of production reports.

The research concluded with a set of recommendations, most notably the need to establish an integrated data management system that includes operational data processing to provide real-time solutions and alternatives that contribute to supporting decision-making related to rationalizing resource consumption.

Abstract

Neuromarketing is an emerging field that combines neuroscience and marketing to better understand the factors influencing consumer decisions. Consumer purchasing behavior has long been a focus for marketers seeking to create compelling advertisements and increase engagement. It relies on objective tools to identify consumer responses through physiological and neurological measurements. This research aims to apply brain science to marketing research, specifically using electroencephalography (EEG) and eye-tracking to determine whether a consumer will purchase a product. “The association between physiological reactions and self-reported purchase intention to marketing cues was investigated by analysing physiological data from 210 participants. The link between the independent factors (product category and advertisement type) and the dependent variable (purchase intention) was investigated using descriptive, correlational, and regression methods. The results confirm that knowledge of neuromarketing significantly boosts practicality (β = 0.726, p < 0.001), which enhances marketing efforts (β = 0.611, p < 0.001) and social media communication effectiveness (β = 0.712, p < 0.001).” The study found that both frontal lobe alpha and beta wave activity (neural reactivity) and fixation duration and number of fixations (visual attention) were strong predictors of purchase intention, with fixation duration being the strongest predictor. Results suggest that emotional advertising elicited greater purchase intention than informational advertising, with the effects of emotion moderated by product category. This research presents evidence for a model that uses EEG and eye-tracking in conjunction to develop insights into consumer behavior. The paper provides an immediately applicable framework for marketers trying to revamp their advertising strategy.

Abstract

Rapid evolution of the Internet of Things (IoT) has transformed the urban infrastructure because it is possible to interconnect systems that can improve transportation, healthcare, security, energy management, and public services. Implementation of IoT network in Iraqi cities faces numerous constraints including lack of infrastructural development, network saturation, unstable connections, poor cybersecurity and lack of resource management. These factors largely affect dependability and functionality of smart city applications. This paper examined how Artificial Intelligence (AI) methods applied to improve IoT network performance in Iraqi cities through effective data routing, predictive maintenance, anomaly detection, bandwidth optimization, and intelligent resource management. Machine learning, Deep Learning and reinforcement learning methods were commonly used methods to address the conditions of dynamic networks and maximize operational efficiency. The study aims at the application of AI techniques to strengthen current IoT systems to enhance scalability, minimize latency, enhance security, and energy efficiency. The research, based on analytical modeling and comparative performance analysis, shows that AI-based IoT systems significantly improve technological resilience of cities and assist Iraq in the development of smart cities. The results offer viable advice to policy makers, network engineers and urban planners who wish to develop modern cities in Iraq using intelligent technology.

Abstract

Given the swift digital changes occurring in the Banking industry, the purpose of this paper is to examine how well artificial intelligence systems can forecast and protect against future disasters.  By utilizing its skills in big data analytics, forecasting financial behavior, and more accurately and effectively managing risks, artificial intelligence (AI) is increasingly regarded as a crucial component in the development of banking systems and improving their operational efficiency.

 By enhancing client satisfaction, tailoring banking services to meet the demands of each individual, and cutting down on operational errors and administrative expenses, banks hope to gain a competitive edge by utilizing these technologies.  AI also helps to speed up credit decisions, make it possible to identify financial crime early, and create clever marketing plans based on forecasts of future market trends.

In order to ensure financial sustainability and achieve integration between digital transformation and the demands of banking innovation, studies show that the future of AI encompasses strategic, cultural, human, technological, and organizational dimensions in addition to technical ones.

 The paper also examined a number of anticipated long-term effects of AI applications, such as increased forecasting precision, lower operating expenses, better customer satisfaction, increased worker productivity, and assistance with investment choices.  The findings show that implementing AI applications in the banking sector is a strategic requirement to guarantee long-term growth and competitiveness in the digital era, not a technical luxury.

In order to enhance lending decisions and lower default risks, the paper also assesses how well a number of categorization algorithms work in assessing loan applicants' creditworthiness.  Using a dataset that represented the traits and financial activities of clients, seven machine learning techniques were used: Gradient Boosting, Random Forest, Extra Trees, Gaussian Naive Bayes, Logistic Regression, SVC-RBF, and KNN.

The paper used a database of 21 variables for loan applicants. Numerical variables included (age, income, credit score, debt-to-income ratio, and loan amount). Descriptive variables included (loan purpose, region, marital status, employer, educational level, and application channel). Binary variables included (whether or not the applicant had a history of default). These variables were used to predict the approval or rejection decision, with the dependent variable being represented by two values: 0 for rejection and 1 for approval.

The models were evaluated using the following six key performance indicators: Accuracy, Precision, Recall, F1 Score, Receiver Operating Characteristic Area Under the Curve (ROC AUC), and Brier Score.   The findings demonstrated that the Gradient Boosting algorithm performed best overall in both probability prediction quality and customer differentiation across different risk levels.  The Random Forest algorithm, which showed stability and balanced metrics, came next.  On the other hand, despite its moderate performance, Logistic Regression provided great interpretability, while the Gaussian Naive Bayes algorithm demonstrated high sensitivity in identifying high-risk customers.  In terms of overall accuracy and probability quality, some models—like SVC-RBF and KNN—performed worse.

Abstract

Smart universities constitute a complex environment in which thousands of Internet of Things (IoT) devices are interconnected, rendering them an increasingly attractive target for sophisticated cyberattacks. This study aims to design an integrated framework combining Artificial Intelligence (AI) and IoT technologies for the early detection of cyberattacks within smart university environments, employing machine learning and deep learning algorithms — namely Random Forest and a hybrid CNN-LSTM model — to classify network traffic and detect anomalous patterns in real time. The study adopted a descriptive-analytical approach supported by a field questionnaire administered to a sample of 150 participants from IT and network security personnel across universities. Results revealed excellent questionnaire reliability (Cronbach's Alpha = 0.927) and a strong positive correlation (r = 0.674) between cybersecurity awareness and AI detection effectiveness, alongside the superiority of the proposed model over traditional systems by an accuracy margin of approximately 25 percentage points (96.3% vs. 71.2%). The study recommends adopting a five-layer architectural framework integrating IoT devices, data processing, AI inference, attack detection, and real-time response layers as a viable model for deployment in Arab smart universities.

Abstract

The study aims to identify the conceptual framework of sustainable accounting and the role of Artificial Intelligence (AI) technologies in enhancing corporate reporting. Furthermore, it demonstrates the impact of adopting sustainability accounting on the financial reporting of a sample of five companies listed on the Iraq Stock Exchange from the year (2019) to the year (2022). The study relied on the descriptive analytical approach using the SPSS program. The main findings reveal the absence of a statistically significant relationship regarding the influence of disclosure in financial reports to apply sustainability standards on investors' decisions, as well as the absence of a significant relationship between the impact of sustainability accounting and the quality of financial reports. Integrating AI tools—such as machine learning and automated ESG analytics—can bridge these gaps by improving data accuracy, reducing reporting bias, and enhancing transparency. The study recommended a set of recommendations, the most important of which is obligating auditors specialized in the profession of auditing companies not to express an opinion on the validity of financial reports if they are devoid of sustainability accounting reports and supported AI-driven disclosure metrics.

Abstract

This research aims to measure the impact of using Internet of Things (IoT) and machine learning techniques on achieving the characteristics of accounting information and the extent to which this is reflected in the conceptual statement/8.This is based on the ability of artificial intelligence techniques to solve business problems and overcome obstacles facing companies by using highly advanced devices and applications, as it is a system capable on simulating human intelligence. According to the importance and problem of the research, and for the purpose of testing its hypotheses, a questionnaire was designed using Google Forms and distributed electronically to the research sample members, who were selected using a deliberate (intentional) sample consisting of employees from the departments of (financial affairs, auditing and internal control). (50) responses were obtained. Using the Statistical Package for the Social Sciences (SPSS), the results of the questionnaire were analyzed. According to the applied frame wall of the research, a set of conclusions were reached, the most important of which was that artificial intelligence techniques achieve the qualitative characteristics of accounting information, as stated in Conceptual statement/8. The researchers presented a set of The most prominent recommendations include the adoption by the Accounting Department of the Basrah Municipality Directorate of artificial intelligence techniques when performing various accounting tasks.

Abstract

This paper reviews the growing role of AI in accelerating the discovery and design of inorganic nanomaterials for sustainable development through a six-stage pipeline: Data → AI → Prediction → Design → Fabrication → Application. Using a critical-analytical approach supported by quantitative metrics (99.99% correlation for TiO₂ band gap, screening of >10 million oxides, >2 compounds/day at A-Lab), it systematically compares ML, DL, and GNN techniques, alongside DFT-based high-throughput screening, inverse design, active learning, and self-driving laboratories. Based on 30 peer-reviewed references (2013–2026) selected using PRISMA guidelines from Scopus, Web of Science, and Google Scholar.

Abstract

This research aims to analyze the role of integrating artificial intelligence (AI) technologies in promoting digital transformation in the banking sector. This is of great importance due to its significant impact on monetary policy. Through the application of digital technologies, banks and financial institutions can greatly improve their procedures, leading to changes in banking liquidity. Digital transformation can contribute to increased cash flows and improved liquidity management, thus affecting the central bank's ability to implement monetary policy effectively. Thanks to big data analytics and AI, banks can better identify economic needs, influencing interest rate decisions. Therefore, it can be said that digital transformation plays a significant role in shaping monetary policy and opens the door to significant improvements in the management of fiscal and monetary policies and creative thinking skills. Traditional reliance on auditing methods does not keep pace with rapid technological developments, and auditors' weak adoption of creative thinking methods in addressing audit problems negatively affects the quality of audits and the accuracy of their results. The importance of this research lies in its aim to provide a comprehensive framework that combines modern technological tools and creative approaches to digital work, thereby contributing to the development of critical practices and enhancing the quality of its results according to standards of efficiency and effectiveness. The advanced statistical software Smart Pals and appropriate statistical methods were used to test the hypotheses. The results showed a statistically significant effect of using artificial intelligence technologies (expert systems, machine learning, continuous auditing, natural language processing, and robotic process automation) across all dimensions. This indicates the importance of AI applications in improving the quality of external auditing. Furthermore, the use of AI technologies has a positive impact on investor decisions through the quality of external auditing. Based on these findings, the researcher recommends that all accountants and auditors keep pace with technological developments and respond to changes in the surrounding environment by developing their AI skills. Audit firms should also be encouraged to adopt AI technologies in their work by replacing manual systems with computerized systems. The study aimed to analyze the impact of digital transformation indicators (ATMs, electronic payment cards, electronic point-of-sale systems, and instant settlement transfers). Adopting digital technologies has contributed to reducing reliance on cash and increasing the efficiency of monetary policy. However, in Iraq, limited effects were observed due to structural challenges such as weak infrastructure and economic instability.

Abstract

Artificial intelligence (AI) applications are essential for enhancing payment security and improving customer experience in financial institutions. This research aims to examine the challenges faced by Mastercard in Australia and Visa in Singapore in implementing these technologies. The study focuses on three main areas: challenges related to fraud and privacy, the strategies employed by both companies to address these challenges, and the impact of AI applications on customer satisfaction and trust. The research adopts an analytical methodology that combines both qualitative and quantitative data, relying on diverse sources including academic studies and annual reports. The findings reveal that the effectiveness of AI applications varies across the two markets, reflecting different responses to local challenges. For instance, Mastercard demonstrates a greater reliance on machine learning technologies in Australia to combat fraud, while Visa in Singapore focuses more on enhancing privacy and data protection. The study offers strategic recommendations aimed at improving payment security and customer experience, such as increasing transparency in data usage and strengthening communication with customers. This research contributes to a deeper understanding of the role of AI in financial services, providing valuable insights for companies and practitioners in the sector. By addressing the unique challenges of each market, customer satisfaction can be enhanced and greater trust in digital payment services can be fostered.