Interactive machine learning by visualization: a small data solution

• Main Author: Beh, Jeff Zhiwen
• Other Authors: Zheng Jianmin
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering::Information systems::Information interfaces and presentation
• Online Access: https://hdl.handle.net/10356/171952
• Institution: Nanyang Technological University

Machine learning algorithms and conventional data mining processes typically necessitate a substantial amount of data to train models tailored to the algorithms. Often, there is limited to no user feedback throughout the model construction phase. This approach, which hinges on leveraging "big data" for automated learning, can be impractical in scenarios where gathering or processing data is arduous or costly, such as in the context of clinical trials. Furthermore, domains like biomedical sciences can greatly benefit from the incorporation of domain expertise during the model creation process. This report proposes a novel approach to interactive machine learning and visual data mining. It involves a visual analytics framework that facilitates user engagement. This report encompasses sections with code excerpts, screen captures, and elucidations of the implementation procedure. The primary objective is to furnish comprehensive documentation of the development trajectory