Early Warning of Human Crowds Based on Query Data from Baidu Maps: Analysis Based on Shanghai Stampede

Abstract

Without sufficient preparation and on-site management, the mass scale unexpected huge human crowd is a serious threat to public safety. A recent impressive tragedy is the 2014 Shanghai Stampede, where 36 people were killed and 49 were injured in celebration of the New Year's Eve on December 31th 2014 in the Shanghai Bund. Due to the innately stochastic and complicated individual movement, it is not easy to predict collective gatherings, which potentially leads to crowd events. In this paper, with leveraging the big data generated on Baidu map, we propose a novel approach to early warning such potential crowd disasters, which has profound public benefits. An insightful observation is that, with the prevalence and convenience of mobile map service, users usually search on the Baidu map to plan a routine. Therefore, aggregating users' query data on Baidu map can obtain priori and indication information for estimating future human population in a specific area ahead of time. Our careful analysis and deep investigation on the Baidu map data on various events also demonstrates a strong correlation pattern between the number of map query and the number of positioning users in an area. Based on such observation, we propose a decision method utilizing query data on Baidu map to invoke warnings for potential crowd events about 1 ∼ 3 hours in advance. Then we also construct a machine learning model with heterogeneous data (such as query data and mobile positioning data) to quantitatively measure the risk of the potential crowd disasters. We evaluate the effectiveness of our methods on the data of Baidu map.

Measuring economic activity in China with mobile big data

“…retails, cinemas, auto sales, restaurants) by corresponding locations search volumes (map queries). As shown in our previous research [ 40,41 ], searching a location on maps leads to high probability of offline arrival in the near future, and on the aggregate level, we find that the the volume of map queries of one location is highly correlated with its offline foot traffic. We therefore use the volume of map queries to estimate the volume of visited consumers, and define the trends of location queries of different industries as consumer spending indicators.…”

On the statistical properties of viral misinformation in online social media

“…The relevance of our results is not necessarily limited to the field of computational social science coping with misinformation [47,48,49]. Indeed, despite the prediction of extremely viral posts and rare events remains an hard task [50,51], we believe that both our findings and the methodology used herein may be of interest to the broader field of computational social science dealing with forecast-ing and tracking of viral contents and events [52,53,54,55,56,57,58,59,60,61].…”

The impact of position errors on crowd simulation

“…Franke et al [15] implemented a smartphone based crowd management system, which also uses a heat map representation of the crowd state and its evolution. Zhou et al [68] proposed a solution based on Baidu map and developed a prediction model to perceive the crowd anomaly and to assess the risk of the crowd event.…”

“…Besides GPS-based solutions, vision-based systems have also been developed to detect congestion spots and pedestrian behaviors. The accuracy of vision-based systems can be affected by three factors: camera coverage, camera resolutions, and level of illumination [68,25]. The existing computer vision work focuses on real-time detection of the stampedes [24,35,26,36,69,42,43] rather than proactive prediction, making it complementary to our goal.…”