Point-of-interest categorisation: towards putting the human back in the loop - Naver Labs Europe
loader image

Abstract

Point of Interest (POI) categories can facilitate several services, such as location-based search and place recommendation. However, such information can be incomplete and/or incorrect, especially in crowdsourcing environments. In the literature, machine learning (ML) based category imputation has been suggested to tackle this problem. However, none of the existing works has studied how to optimize the involvement of a human in creating such models. In this paper, we discuss specific challenges, opportunities and first steps towards designing an appropriate what-if system for ML model exploration and understanding.

Related Content