<![CDATA[This research was conducted in order to determine the method of fraud detection. The method used in this research is Systematic Literature Review (SLR) with PRISMA Protocol. The steps taken include identification, screening, feasibility testing, and finalization of relevant studies from various leading databases such as Google Scholar, Scopus, Web of Science, Emerald, and IEEE Xplore. The results of the study identified eight main methods of fraud detection, namely big data analysis, machine learning algorithms, Natural Language Processing (NLP); Accounting Information Systems, Extremely Randomized Trees (ERT), rule-based systems, blockchain technology, Asexual Reproduction Optimization (ARO), Bayesian A/B testing, and authentication systems. These methods show varying degrees of effectiveness in detecting suspicious transactions and reducing the risk of digital fraud. The research findings not only discuss fraud detection methods, but also analyze the weaknesses and advantages of each method. The findings can be used as a practical reference source in preventing fraud by conducting fraud detection based on the right method.]]>
Copyrights © 2026
