Data analysis and data interpretation are fundamental to many data dependent disciplines, such as bioinformatics, speech synthesis and machine vision. Data analysis is an established field of study in the engineering sciences, but has only marginally been applied to art practices (of all flavors) until recently. This course will allow students to dive into, and critically examine, this challenging field. The course will be held in both theory (one credit) and production (3 credits) mode, so students lacking comfort in computer programming can participate.
Topics include: fundamentals of applied statistics, basics of language/text processing, genetic algorithms and neural networks and select AI techniques. From these rich but also demanding engineering science fields we will concentrate on aspects that facilitate the interpretation of data.
All exercises and examples (for the 3 credit version) will be in the open-source and platform independent programming language python. Readings and discussions will complement the technical materials.

>>> Prerequisites: programming experience, college algebra and calculus, MediaRoboticsII or equivalent + dedication.


W1
Overview

W2
Applied Statistics: basics

W3
Applied Statistics: descriptive statistics

W4
Applied Statistics: practical issues

W5
Natural Language Processing: overview

W6
Natural Language Processing: parsing

W7
Natural Language Processing: text processing

W8
Genetic Algorithms: basics

W9
Genetic Algorithms: fitness functions

W10
Genetic Algorithms: practical issues

W11
Neural Networks: overview

W12
Neural Networks: perceptrons

W13
Neural Networks: hopfield nets

W14
Neural Networks: kohonen maps

W15 - 17
Project development

>> Scripts on all four seminar topics - Applied Statistics, Text Processing, Genetic Algorithms
and Neural Nets - are available here for registered students: SCRIPTS