Next Place Prediction Based on Spatiotemporal Pattern Mining of Mobile Device Logs

AbstractDrought is a common natural disaster that greatly affects the crop yield and water supply in China. However, the spatiotemporal characteristics of drought in China are not well understood. This paper explores the spatial and temporal distributions of droughts in China over the past 40 years using multiscale standardized precipitation evapotranspiration index (SPEI) values calculated by monthly precipitation and temperature data from 612 meteorological stations in China from 1980 to 2019 and combines the space-time cube (STC), Mann-Kendall (M-K) test, emerging spatiotemporal hotspot analysis, spatiotemporal clustering and local outliers for the analysis. The results were as follows: 1) the drought frequency and STC show that there is a significant difference in the spatiotemporal distribution of drought in China, with the most severe drought in Northwest China, followed by the western part of Southwest China and the northern part of North China. 2) The emerging spatiotemporal hotspot analysis of SPEI6 over the past 40 years reveals two cold spots in subregion 4, indicating that future droughts in the region will be more severe. 3) A local outlier analysis of the multiscale SPEI yields a low-low outlier in western North China, indicating relatively more severe year-round drought in this area than in other areas. The low-high outlier in central China indicates that this region was not dry in the past and that drought will become more severe in this region in the future.

Download Full-text

Adding intelligence to your mobile device via on-device sequential pattern mining

Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing Adjunct Publication - UbiComp '14 Adjunct ◽

10.1145/2638728.2641285 ◽

2014 ◽

Cited By ~ 7

Author(s):

Abhishek Mukherji ◽

Vijay Srinivasan ◽

Evan Welbourne

Keyword(s):

Mobile Device ◽

Pattern Mining ◽

Sequential Pattern Mining ◽

Sequential Pattern

Download Full-text

Watch Out for the Tailings Pond, a Sharp Edge Hanging over Our Heads: Lessons Learned and Perceptions from the Brumadinho Tailings Dam Failure Disaster

Remote Sensing ◽

10.3390/rs13091775 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1775

Author(s):

Deqiang Cheng ◽

Yifei Cui ◽

Zhenhong Li ◽

Javed Iqbal

Keyword(s):

Pattern Mining ◽

Hot Spot ◽

Economic Loss ◽

Lessons Learned ◽

Disaster Mitigation ◽

Dam Failure ◽

Extreme Weather Events ◽

Spatiotemporal Pattern ◽

Tailings Dam ◽

Depth Analysis

A catastrophic tailings dam failure disaster occurred in Brumadinho, Brazil on 25 January 2019, which resulted in over 270 casualties, 24,000 residents evacuated, and a huge economic loss. Environmental concerns were raised for the potential pollution of water due to tailings waste entering the Paraopeba River. In this paper, a detailed analysis has been carried out to investigate the disaster conditions of the Brumadinho dam failure using satellite images with different spatial resolutions. Our in-depth analysis reveals that the hazard chain caused by this failure contained three stages, namely dam failure, mudflow, and the hyperconcentrated flow in the Paraopeba River. The variation characteristics of turbidity of the Rio Paraopeba River after the disaster have also been investigated using high-resolution remote sensing images, followed by a qualitative analysis of the impacts on the downstream reservoir of the Retiro Baixo Plant that was over 300 km away from the dam failure origin. It is believed that, on the one hand, the lack of dam stability management at the maintenance stage was the main cause of this disaster. On the other hand, the abundant antecedent precipitation caused by extreme weather events should be a critical triggering factor. Furthermore, the spatiotemporal pattern mining of global tailings dam failures revealed that the Brumadinho dam disaster belonged to a Consecutive Hot Spot area, suggesting that the regular drainage inspection, risk assessment, monitoring, and early warning of tailings dam in Consecutive Hot Spot areas still need to be strengthened for disaster mitigation.

Download Full-text