Dynamic and multi-source semantic annotation of raw mobility data using geographic and social media data

Nowadays, positioning technologies have become widely available providing then large datasets of individuals' mobility data. Actually, annotating raw traces with contextual information brings semantics to them and then provides a better understanding of people behavior. To do so, literature work explored novel techniques to enrich raw mobility data with contextual information using either geographic context represented by landmarks/points of interest or widely used social media feeds. Accordingly, in this work, a novel approach integrating three data sources: raw mobility data, geographic information and social media feeds for a two-fold trajectory semantic annotation process is presented. In a first step, structured trajectories are constructed using geographic information. Later, the former are annotated by event-related words grasped from social media. Indeed, combining both data sources could result in a more complete annotation of trajectories. The proposed approach is experimented and evaluated on datasets of tourists in Kyoto. Results showed that the proposed approach quantitatively performed well compared to previous work in terms of precision of annotation words that maintained similar or equal to 0.9 when recall reached 50%, while improving its quality by consolidating both sources of semantics. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

This study explores enriching individuals' mobility data with contextual information from geographic data and social media, presenting a novel approach that integrates these three data sources for trajectory semantic annotation. Experimental results show that the proposed approach improves the precision of annotation words while maintaining similar recall rates, enhancing the quality of trajectory semantics by combining both data sources.