Big data analytics for the City of Copenhagen

The City of Copenhagen has a lot of data about the citizens available that is currently not utilized. The overall aim of this project is to develop big data analytics for the City of Copenhagen in order to improve the welfare of its citizens. We focus on problems related to prediction of the necessary frequency and type of homecare and support in the decision making carried out by employees in the city.

We developed an algorithm for predicting the future needs for elderly citizens in terms of homecare, thus being able to assist the citizen quicker and to foresee the resources needed for homecare in the near future. We are currently working on a system to support employees in their decision making regarding complaints from the citizens.

The solutions are based on machine learning and natural language processing. To predict the amount of healthcare needed, we aim at learning a rich temporal representation of each citizen in order to utilize their similarities when predicting each elderly citizen’s needs. In order for us to assist the city’s employees we employ natural language processing models to capture the meaning of citizen’s complaints in order to effectively match them to optimal replies.

The project generates knowledge about how to predict homecare in a large city by utilizing other kinds of data about the citizens. This knowledge is valuable not only to the City of Copenhagen but also to other cities around the world. Furthermore, the project yields important knowledge about automated handling of documents in order to support employees in faster decision making.

Predicting the homecare required by the elderly citizens makes it possible for the city to provide faster and more adequate assistance to ailing citizens, thus increasing the well-being of the citizens. Supporting the employees in handling citizens’ complaints will speed up the process, free up resources currently required to handle complaints, improve citizen satisfaction due to reduced response time, etc.

This project is a joint project between Department of Computer Science, University of Copenhagen and the City of Copenhagen.