Machine learning algorithms in the construction sector
In the CityCNN project, Arkkitehdit Sankari Oy experiment with the use of machine learning algorithms in the construction industry.
What kind of problem is the project trying to solve?
The project looks for new opportunities of using the enormous amount of open data material published by municipalities that, at the moment, is hard to use to support design.
Objective: Experiment with machine learning algorithms for using open data in the construction sector
The objective of the project is to experiment with applying the machine learning algorithms that progress at a rapid pace in information systems science in the context of construction sector. The experiment utilises the enormous amounts of data that municipalities have published that can fairly easily and automatically be prepared for use as the basis for teaching machine learning models.
What is done in this project?
In the project, various machine learning algorithms and their variants are compared and their usability is tested in supporting tasks in urban and building design. The project uses the enormous amount of data that municipalities have published as a basis for teaching machine learning models.
The project also involves preparing a dataset that allows experimenting with applying an artificial intelligence model to problems that are usually regarded as requiring the expertise and experience of the designer, such as finding locations for complementary construction.
What are the expected results of the experiment?
The objective of the experiment is to provide examples, as well as suggestions for tools and methods concerning a total work flow that allows machine learning algorithms to be used in supporting a building or urban designer in their everyday tasks. The experiment results in demonstrations and presentations that can be used as a basis for further development.
Who benefits?
The understanding of the conditions of using machine learning models benefits construction companies and municipalities that publish the data. In addition, it may create new business models to IT companies.