Volume 4, Issue 1 , June 2025
Mahmoud Lari Dashtbayaz; Zahra Rezazadeh; Toktam Safdel
Abstract
Audit fees are considered important issues for the client and the auditors and a critical component of corporate governance in business environment. Abnormal audit profits or costs can signal underlying issues in audit quality, auditor independence, or financial health of the audited entity. Understanding ...
Read More
Audit fees are considered important issues for the client and the auditors and a critical component of corporate governance in business environment. Abnormal audit profits or costs can signal underlying issues in audit quality, auditor independence, or financial health of the audited entity. Understanding the factors that affect audit fees is essential for stakeholders to ensure transparency and assurance in financial reporting. Effective communication between auditors and clients is crucial in addressing these issues and ensuring accurate and reliable audit outcomes. The present study examined the relationship between abnormal audit profits or costs and factors influencing audit fees. The research sample comprised 139 companies (695 firm-years) listed on the Tehran Stock Exchange (TSE) during 2016–2020. This period was selected due to significant regulatory changes and economic events that impacted audit practices and fee structures. A multiple regression model was employed to test the research hypotheses. The results showed that abnormal audit profits or costs significantly affect audit quality, Type I audit error, and audit report lag. On the other hand, abnormal audit profits or costs hurt Type II audit errors. The findings suggested that higher audit fees in Iran are more likely to represent the actual costs of conducting audits rather than excess profits. Consequently, higher audit fees lead to increased audit production costs (APC) and improved audit quality.
Djalila Boughareb; Hazem Bensalah; Zineddine Kouahla
Abstract
The increasing demand for personalized advertising based on user preferences is driving a surge in popularity. Social networks utilize millions of user’ data to suggest ads based on specific criteria. However, many of these ads can be uninteresting. This paper presents a collaborative advertisement ...
Read More
The increasing demand for personalized advertising based on user preferences is driving a surge in popularity. Social networks utilize millions of user’ data to suggest ads based on specific criteria. However, many of these ads can be uninteresting. This paper presents a collaborative advertisement recommendation system that leverages users’ preferences along with geographic and demographic data to deliver engaging ads. The system employs the K-dtree algorithm to efficiently organize users into interest-based communities and model complex patterns within those communities to enhance ad relevance. The dataset, collected via Hazmit provides a rich source of information. The system’s performance was evaluated based on precision, recall, F-score, and accuracy metrics, as well as running time measurements. The results highlighted the superior effectiveness of the K-dtree-based approach in accurately targeting the right customers for advertisements. Overall, the K-dtree method improves ad targeting accuracy, especially for food and demographics, but struggles with news due to subjectivity and regional biases.
Raziyeh Moghaddas; Farinaz Tanhaei; Maryam Al Moqbali; Solmaz Safari
Abstract
The incorporation of machine learning (ML) approaches into business intelligence (BI) results in a great impact on fields that require predictive analysis, such as the real estate market. An accurate prediction of housing prices can provide benefits to stakeholders such as developers, investors, and ...
Read More
The incorporation of machine learning (ML) approaches into business intelligence (BI) results in a great impact on fields that require predictive analysis, such as the real estate market. An accurate prediction of housing prices can provide benefits to stakeholders such as developers, investors, and policy planners. This study aims to explore the application of ML techniques to property valuation by creating a dataset based on real-world house price data collected from various areas in Muscat, Oman. Several ML models, including Linear Regression, Ridge Regression, Gradient Boosting, Random Forest, and Support Vector Regression, were applied and examined on the created dataset to estimate the house prices. Besides, hyperparameter tuning is used for each model in order to improve their predictive accuracy. Finally, we assessed the performance of each model using standard evaluation metrics, i.e., Mean Absolute Error (MAE), Mean Squared Error (MSE), and the R-squared (R²) score. The findings of this research work provide a comparative analysis of model efficiency that highlights both the capabilities and limitations of each model. This study demonstrates the practical power of ML techniques in real-state analytics and its wider applicability in improving BI systems subsequently.
Bekan Kitaw Mekonen
Abstract
Effective communication through digital platforms often faces issues like misspellings and inefficient typing. A next-word prediction system that suggests probable words can significantly enhance sentence construction, especially for Afaan Oromo - a Cushitic language spoken by over 41.7 million people ...
Read More
Effective communication through digital platforms often faces issues like misspellings and inefficient typing. A next-word prediction system that suggests probable words can significantly enhance sentence construction, especially for Afaan Oromo - a Cushitic language spoken by over 41.7 million people in Ethiopia. Despite its importance as the official language of Oromia and its complex linguistic features, Afaan Oromo lacks advanced digital tools. This study evaluates various deep learning models, including Long Short Term Memory (LSTM), Attention-based LSTM, Bidirectional LSTM (Bi-LSTM), Attention-based Bi-LSTM, and Recurrent Neural Network (RNN), to determine the most accurate model for Afaan Oromo next word generation. Our methodology involves developing and benchmarking these models using a comprehensive dataset of 201,538 words sourced from various media, academic literature, and religious texts. The Attention-driven Bi-LSTM model emerged as the most effective, achieving an accuracy of 95.0% and a low loss value of 0.27. These findings highlight the potential of the Attention-driven Bi-LSTM model to improve the next word generation for Afaan Oromo texts. This advancement addresses specific linguistic challenges and enhances the overall digital interaction experience for Afaan Oromo speakers.
Wael Eldesouki Bedda; Heba Al-Ashry; Hamed A. Ead
Abstract
This study systematically reviews global practices in AI-enabled digital marketing for startups and examines the implementation challenges faced by Egyptian entrepreneurs. By systematically reviewing 21 studies (2017–2023) from Scopus, Web of Science, and Science Direct, the research highlights ...
Read More
This study systematically reviews global practices in AI-enabled digital marketing for startups and examines the implementation challenges faced by Egyptian entrepreneurs. By systematically reviewing 21 studies (2017–2023) from Scopus, Web of Science, and Science Direct, the research highlights how AI enhances engagement with customers, improves campaign results, and boosts prediction abilities. Key findings reveal that AI tools, such as chatbots and predictive analytics, improve personalization, with some studies reporting an increase in conversion rates in emerging markets. However, Egyptian startups face significant barriers, including limited AI adoption, infrastructural gaps, and skill shortages. The study also examines Egypt’s entrepreneurial ecosystem, noting government-backed incubators like INTILAC and a youth-driven, tech-savvy population as potential enablers for AI integration. Challenges such as ethical concerns, algorithmic bias, and cultural readiness still persist. The paper concludes with recommendations for policymakers and entrepreneurs to bridge these gaps, emphasizing the need for AI literacy, targeted incentives, and ethical frameworks to foster sustainable growth in Egypt’s digital economy.
