This talk from the Applied Algebraic Topology Network explores the comparative effectiveness of topological data analysis versus machine learning methods for classification tasks. Dive into a naive analysis that puts persistent homology methods head-to-head with traditional machine learning and deep learning techniques for label-efficient classification. Follow along as the speaker evaluates simple topological approaches—including persistence thresholding and Bottleneck distance classification—against conventional learning algorithms and hybrid methods. The analysis focuses on two practical binary classification tasks: surface crack detection and malaria cell identification. Perfect for those interested in understanding which approach might be more suitable for specific data analysis challenges.