HOME / Exhibition Program / Estimating risk of infection in a city
Exhibition Program
Science of Machine Learning

Estimating risk of infection in a city

People flow reconstruction based on anonymous sensor data

Estimating risk of infection in a city

We can measure the risk of infection in a city by knowing the movement routes of infected people, but collecting information of people's movements infringes privacy. We thus propose a new method for estimating multiple routes on the basis of anonymized passage information to estimate the risk of infection while preserving privacy. For a better estimation, we need to choose more appropriate path patterns of people that correctly explain the anonymized passage information. Therefore, we consider a movement model, estimate the transit probability between passage information, and find the most likely set of routes efficiently on the basis of the model. It can improve the accuracy of the estimation. Infectious disease control will be one of functions of smart cities to be realized in the future.By using our work, the risk of infection can be estimated without collecting personal movement information.

Estimating risk of infection in a city

[1] K. Matsuda, H. Ikeuchi, Y. Takahashi, T. Toyono, “People Flow Reconstruction Based on Anonymous Sensor Data toward Smart City Infrastructure for Estimating Infection Route,” IEICE Technical Committee on Network Systems, 2020.


Kotaro Matsuda / NTT Network Technology laboratories
Email: cs-openhouse-ml@hco.ntt.co.jp

Click here for other research exhibits