This course provides attendees with a basic working knowledge of signal processing and organization of digital documents using diffusion geometry. We will provide an overview of computational tools for data driven empirical analysis. In particular, constructions of basis functions, generalizating Wavelets, and Fourier modes play a central part. We will discuss the basic methodology for data driven computational learning. Examples of the technique include: organization of sensor data such as sound signatures, images, spectra, text documents, and psychological and/or security questionnaires. Other applications include heterogeneous digital data fusion, subjective search in the digital domain, filling in missing data, denoising, classification diagnostics, and anomaly detection. Relevant emerging applications for this technique include improved filtering of irrelevant and inaccurate data from high dimensional relational security & defense databases such as no-fly and watch lists, allowing resources to be concentrated on valid targets while improving privacy for the overall populace.