PRISMA Approach for Assessing Fingerprint Classification Models Based on Artificial Super-Intelligence Techniques

The aim of this research paper focused on using PRISMA to reveal most artificial intelligence techniques that were used for fingerprint classification. Biometric technology such as fingerprints plays a key role in authenticating and identifying people’s identities. Therefore, with the increasing number of population and the usage of biometrics for authentication, fingerprint classification systems are becoming important and indispensable for recognizing and authenticating individuals. Therefore, Artificial Super-Intelligence (ASI) techniques such as bioinspired algorithm, deep learning and machine learning were used to improve fingerprint classification accuracy. The proposed method aimed to assess fingerprint classification models based on ASI. The researchers employed PRISMA approach, which is based on systematic analysis and is used to select, evaluate and analyze journals. Although IEEEXplore and Web of Science were utilized to extract journal articles from 2019 to 2023. As a result, 1350 articles were found in both databases. Furthermore, a total of 35 publications were assessed to determine their eligibility and 19 articles were eliminated with reasons and 16 matched the requirements for a meta-analysis. Our findings demonstrate and highlight the need for developing a new approach to improve fingerprint classification accuracy.