Scientific Study at the University of Basrah Explores the Development of Pattern Mining Techniques
A scientific study at the University of Basrah explored the development of multi-level, closed, high-utility pattern mining techniques.
The study, presented by Dr. Hussein Abdul-Kareem Younis, in collaboration with researchers from Japan, India, and the Southern Technical University, aims to develop data mining techniques using deep learning to improve the efficiency of high-utility pattern detection and reduce processing time and memory consumption.
The study included the development of a DeepHUPM system based on a neural network for predicting utility thresholds and examining closed patterns. It achieved better results than modern methods in terms of execution speed, memory consumption, and pattern detection accuracy, with potential applications in commerce data analysis and cybersecurity.
Department of Media and Government Communication
