At UDL, we have created a unique database of more than 300 attributes describing each and every built-up place in Switzerland at a very high resolution going all the way down to individual streets, buildings, and even individual users. This is what is powering our predictive models and what brings the big value to our customers.
But as you can see from the image, each of the attributes tells a different story. It is then our role, to find those that are relevant for you - be it a prediction of household income, retail location or an assessment of the best place to build a new student housing.
At the beginning of a collaboration, our customers generally tend to select specific attributes, such as income, proximity to public transport or density of young households. They are all experts in their field and want to define their expert model. This can instantly and interactively be done in our Web application. However, when they are asked to define what is “close to a public transport and what is far” and whether proximity is the same in rural places or urban areas, they realize the millions of possible combinations.
We have developed data driven methods to review the best combinations, and this is why each model we develop is somehow unique. Some use cases may only need a few, carefully selected attributes. Other, potentially hundreds to find a unique combination of many capturing the phenomena we are trying to predict.
We can assure you about one thing - as long as you interact with the physical world, there's something in for you as well ;-) Get in touch if you want to see how we can together figure out what this is!