1 week, 2019

Linnea Groot, Senne Friederichs

What if....

we designers use AI in our creative process to get unexpected but relevant input.

My thoughts

Often people say that creative jobs are not likely to be replace by AI because they lack creativity. But what if we use AI as an tool to enhance our creativity. This project explores if we can use the power of machine learning to crush big databases and find patterns in that to stimulate our creativity.

Concept

When you are stuck in a brainstorm our algorithm can help you by giving unexpected but relevant input in your design process. This all based on relevance between concepts in design which has been thought to the algorithm by learning it 500 design research paper abstracts.

How does it work

To realize this project my team mate created a database of 500 design research paper abstracts.

I made a python program to first process the database to filter out stopwords, numbers and punctation.

After that I use Googles Word2Vec algorithm which is a shallow 2 layer neural network to train an algorithm with the processed data as test set.

By selecting relevant words a suggestion can be giving plus the input is use to keep training the algorithm

back to overview