Creative Data Mining Timetable

16.02.2015 Introduction to Classification
 What is it good for and how is it used. The pizza example
 23.02.2015  Introduction to the Environment
 Including R Studio in your digital toolchain. The AI example.
 02.03.2015  The Math underneath
 The Simplicity of Machine Learning. The Flower example.
 09.03.2015  Automatic Collage 
 The Sketch-Scan-Render Workflow. The buildingblock example.
 16.03.2015  Seminar week (No lecture)
 23.03.2015  Kick-Off Project Work
 Brainstorming, Discussion and Roadmap.
30.03.2015 Applied Classification
Algorithms, Data and Usecases. Student project.
20.04.2015 Workshop I – Analysis
Finishing the first half of the student project.
27.04.2015 Urban Sensing and Simulation
Automatic data acquisition and Urban Hacking. Project work Q&A.
04.05.2015 Workshop II – Interpretation
Finalising the individual project work.
11.05.2015 Final iA critique
Combined critique with the other iA courses


FS2015 | Creative Data Mining

Intuitively Analysing Design Ideas

It is increasingly important to know how to work with big amounts of data – especially in such highly complex fields like architecture and urban planning. This knowledge is necessary for comparing different design variations with each other or to compare design sketches to existing architecture. Thus participants will learn how to to create whole landscapes of designs in a semi-automatic manner.
At first the lectures and exercises will adress the topic of how to automatically create a good overview when multiple design alternatives are given. This seminar will start with introducing the technique of classification as powerful way of sorting all kinds of things. The first few lessons will shed a light on the math underneath and the easiest ways of applying it. Therefore we will present a special workflow of scanning selfmade design sketches and handmade heatmaps to automatically create collages out of it. This allows to find similarities in different designs or built environments and further to tag those automatically.
In a second block, the participants will use the presented method to work on a  small project on their own. They will deepen their knowledge in how to automatically compare multiple design alternatives and how to interprete the results for enriching their concepts. This block is interwoven with lessons regarding additional usecases, data acquisition and theoretical discussions. Finally the participants should be able to decide which of the big amount of data available is worth considering in their designs and how to do it.

Room & Timeslot:
HIT H12 on Mondays from 10:00 to 12:00

Matthias Standfest
Artem Chirkin

Requirement: Former knowledge of any digital tool or coding language is most welcome but NOT required at all. You only need to provide  a reasonable amount of motivation and of course a notebook.