Master’s Thesis at University of Basrah Explores the Use of Persistent Homology in Data Space Classification for Artificial Neural Networks
A master’s thesis at the College of Science, University of Basrah examined the use of persistent homology for classifying data spaces in artificial neural networks.
The thesis, presented by Sajad Kazem Hassan, aimed to demonstrate the benefits of an innovative methodology in artificial neural networks and data classification, focusing on the application of Topological Data Analysis (TDA) techniques, particularly Persistent Homology.
The study evaluated the potential advantages of applying these advanced techniques in the processing and analysis of geophysical well logs, specifically investigating their role in improving the accuracy of predicting compressional sonic wave transit time (DT).
This research represents a significant contribution in bridging applied mathematics with geophysics by employing modern topological tools that enhance the understanding of subsurface geophysical properties.
Department of Media and Governmental Communication
