Data Mining Meets the Needs of Disaster Information Management

Zheng, Li; Shen, Chao; Tang, Liang; Zeng, Chunqiu; Li, Tao; Luis, Steven; Chen, Shu‐Ching

doi:10.1109/thms.2013.2281762

Cited by 62 publications

(23 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Kelly et al, 2011) • Data-mining for deriving information about the status of disaster relieve and how the community is recovering. (Zheng et al, 2013) • Twitter data for detecting earthquakes. (Sakaki et al, 2010) For more applications, we refer the reader to Hristidis et al (2010) and Schmitt et al (2007).…”

Section: Related Workmentioning

confidence: 99%

Real-Time Optimization for Disaster Response: A Mathematical Programming Approach

Chan¹,

Wong²,

Kuo³

et al. 2015

STBD

View full text Add to dashboard Cite

Disasters are sudden and calamitous events that can cause severe and pervasive negative impacts on society and huge human losses. Governments and humanitarian organizations have been putting tremendous efforts to avoid and reduce the negative consequences due to disasters. In recent years, information technology and big data have played an important role in disaster management. While there has been much work on disaster information extraction and dissemination, real-time optimization for decision support for disaster response is rarely addressed in big data research. With big data as an enabler, optimization of disaster response decisions from a systems perspective would facilitate the coordination among governments and humanitarian organizations to transport emergency supplies to affected communities in a more effective and efficient way when a disaster strikes. In this paper, we propose a mathematical programming approach, with real-time disaster-related information, to optimize the post-disaster decisions for emergency supplies delivery. Since timeliness is key in a disaster relief setting, we propose a rounding-down heuristic to obtain near-optimal solutions for the provision of rapid and effective response. We also conduct two computational studies. The first one is a case study of Iran that aims to examine the characteristics of the solutions provided by our solution methodology. The second one is to evaluate the computational performance, in terms of effectiveness and efficiency, of the proposed rounding-down heuristic. Computational results show that our proposed approach can obtain near-optimal solutions in a short period of time for large and practical problem sizes. This is an extended work of Kuo et al., 2015, which

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Optimization for Disaster Response: A Mathematical Programming Approach

Chan¹,

Wong²,

Kuo³

et al. 2015

STBD

View full text Add to dashboard Cite

show abstract

“…Recently Ushahidi provides a platform to crowd source news stories and crisis information using multiple channels and prepares visualization and interactive maps [24] and Geo-VISTA monitors tweets to form situation alerts on a map-based user interface according to the geo-locations associated with the tweets [25]. These situation-specific tools provide query interfaces, GIS and visualization capabilities to support user interaction and query [26]. However, they do not generate textual storylines to improve the situation awareness.…”

Section: Related Workmentioning

confidence: 99%

Generating textual storyline to improve situation awareness in disaster management

Zhou

Shen

et al. 2014

Proceedings of the 2014 IEEE 15th International Conference on Information Reuse and Integration (IEEE IRI 2014)

Self Cite

View full text Add to dashboard Cite

Abstract-Hurricane Sandy affected the east coast of U.S. in 2012 and posed immense threats to businesses, human lives and properties. In order to minimize the consequent loss of a catastrophe like this, a critical task in disaster management is to understand situation updates about the disaster from a large number of disaster-related documents, and obtain a big picture of the disaster's trends and how it affects different areas. In this paper, we present a two-layer storyline generation framework which generates an overall or a global storyline of the disaster events in the first layer, and provides condensed information about specific regions affected by the disaster (i.e., a locationspecific storyline) in the second layer. To generate the overall storyline of a disaster, we consider both temporal and spatial factors, which are encoded using integer linear programming. While for location-specific storylines, we employ a Steiner tree based method. Compared with the previous work of storyline generation, which generates flat storylines without considering spatial information, our framework is more suitable for largescale disaster events. We further demonstrate the efficacy of our proposed framework through the evaluation on the datasets of three major hurricane disasters.

show abstract

“…Commercial systems such as Web EOC and E-Team are usually used by Emergency Management departments located in urban areas [9], [10]. Recently many disaster situation-specific tools provide query interfaces, GIS and visualization capabilities to support user interactions and queries to improve situation awareness [11] in a specific disaster event.For example, Ushahidi [12] provides a platform with visualization and interactive maps to crowd source news stories and crisis information using multiple channels and GeoVISTA [13] monitors tweets to form situation alerts according to the geo-locations associated with the tweets. However, these tools do not answer the comparative queries about all disaster related data of different cities.…”

Section: Related Workmentioning

confidence: 99%

A bipartite-graph based approach for disaster susceptibility comparisons among cities

Zhou

Shen

et al. 2014

Proceedings of the 2014 IEEE 15th International Conference on Information Reuse and Integration (IEEE IRI 2014)

Self Cite

View full text Add to dashboard Cite

Abstract-People are attracted to large cities because of more employment opportunities, convenient facilities, and rich cultural activities. However, large cities are also more vulnerable to natural disasters, which have caused widespread physical destructions, great loss of life and property, and immense havoc. "Which city is less susceptible to natural disasters?" is thus one of the most critical questions one faces when making decisions on travelling or job and business relocation. In this work, we propose a bipartite-graph based framework to compare the impacts of disasters on two cities by answering different queries using textual documents collected online. Besides intuitive simple comparison using statistics, our system also generates textual comparative summaries to better describe the differences between the two cities in terms of safety. Although a number of online services provide disaster events statistic information for cities, our framework compares the impacts of disasters on cities in a more straightforward and comprehensive way.

show abstract

Data Mining Meets the Needs of Disaster Information Management

Cited by 62 publications

References 17 publications

Real-Time Optimization for Disaster Response: A Mathematical Programming Approach

Real-Time Optimization for Disaster Response: A Mathematical Programming Approach

Generating textual storyline to improve situation awareness in disaster management

A bipartite-graph based approach for disaster susceptibility comparisons among cities

Contact Info

Product

Resources

About