AI and Machine Learning Applications in Islamic Financial Services: Literature Review

Project: Research

Project Details


The use of artificial intelligence (AI) application considers the new area of modern business practices. AI method, including machine learning and deep learning, is transformative technologies which disrupts many sectors but offer numerous potential benefits. It enables business in different areas such as process automation, forecasting and planning among others. In addition, the AI provides new innovative business solutions. For instance, the machine learning modelling allows banks to automate several businesses such as account opening, lending, and robo advising as personal assistant, security, and recommendations. AI is expected to save 22% in the traditional financial services business by 2030, according to some estimates [1]. To overcome regulatory concerns and instill trust in AI-based systems, AI can be utilized to replace human resources or complement existing infrastructure, which is the current practice. Islamic financial services as distributive innovation of conventional financial system provides huge opportunities for artificial intelligence and machine learning techniques applications. Especially, applications of AI in ISLAMIC finance become critical to sustain development of bigger Islamic financial markets such as in the Kingdom of Saudi Arabia. In addition, AI and machine learning allows such market to maintain its competitive advantage compared with other markets. Furthermore, the business model of Islamic financial services considers a rich area for further innovation. For instance, several issues of Islamic finance could be used as business problem in artificial intelligence applications. However, the question is still valid regarding development and potential of current AI and machine learning applications for Islamic financial services. Prior studies cover different machine learning applications in Islamic financial services. This included fintech, Chatbot, customer loyalty assessment, bank performance, fatwa, Sukuk rating, robot, index prediction [2, 3]. Such research, the [2] proposed an interactive Chatbot based on artificial Intelligence called 'Chatbot as Islamic Finance Expert' (CaIFE). The CaIFE receives automatic robot support related to Islamic finance and banking by having users communicate with a robot having knowledge accumulated by machine learning. It responds in real time to any inquiries about Islamic finance and banking. It then provides an analysis of the CaIFE example and discusses its features and limitations. However, CaIFE is still in its infancy and continuously improving and growing and is initiative to solve the facing challenging tasks in chatbot assistant, which are understanding the complex sentence structures and provide appropriate response. Authors in [3] propose an Intelligent Information Retrieval (IIR) Framework for Shariah sources using Support Vector Machine (SVM) for Shariah decision making in Islamic Financial Industry (IFI). In addition, this study also discusses the requirements for an automated indexing platform for Shariah sources. SVM is a machine learning, mainly in the classifying, regressing, and finding of information. Nonetheless, the literature of AI and machine learning applications in Islamic finance are still fragmented without clear directions. This limitation does not allow for further model performance evaluation and comparison. Furthermore, the current literature still misses the state of art model (base) in each area of the interest. it might be necessary to understand the current literature of machine learning and AI in Islamic financial services. Therefore, the current study intends to fill this gap by systematically review and scoping the current literature of machine learning applications in Islamic financial services. Besides, this review will help new researchers to understand the present role of ML and AI in the Islamic financial services opening new opportunities for researchers to continue with future work.
Effective start/end date1/10/2131/12/22


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.