A scientific study conducted by Dr. Hussein Abdul Kareem Younis, in collaboration with researchers from Japan, India, and the Southern Technical University, investigated the development of multilevel closed high-utility pattern mining techniques using deep learning. The study aimed to improve data-processing efficiency and enhance the discovery of valuable patterns.

The study proposed a system called DeepHUPM, which utilizes a feedforward neural network to dynamically predict utility thresholds based on data density and length. The system also integrates utility-list techniques and closure checking to reduce redundant patterns and improve the quality of the discovered patterns.

The results demonstrated that the proposed system achieved better performance compared with state-of-the-art methods in terms of execution time, memory consumption, and accuracy in discovering closed patterns. These findings highlight the potential of deep learning techniques in analyzing multilevel data, with possible applications in retail data analysis and cybersecurity.

The study was published in a scientific journal indexed in the international databases Scopus and Web of Science, reflecting the significance of the research and its contribution to the fields of data mining and deep learning applications.