Extremely simple logistics.

ACT is an event response coordination system that facilitates multi-organization responses to an event, including natural disasters, disease epidemics, and unrest. ACT is a fast, eeffective, and open way to streamline communication and coordination.

Extremely powerful analysis.

ACT leverages crowdsourced information and simplifies complex data analysis for a groupsourced response. ACT helps visualize available and deployed resources on interactive maps according to their distribution. ACT's data analysis increases efficiency by avoiding duplicate responses to the same incidents.

Step 1. Collect

ACT collects two types of requests: requests from crowds (crowdsourcing) and requests from groups (groupsourcing). Crowdsourcing refers to requests submitted by people (e.g. victims, volunteers) who are not from certified organizations. The groupsourcing requests originate from responding organizations such as United Nation, Red Cross, etc.

Step 2. Visualize

The data analysis takes advantage of both data mining technology and expert knowledge to iteratively capture the essential content of raw requests. After data analysis, ACT visualizes the requests on a map with information of requests category and quantity.

Step 3. Respond

ACT is able to coordinate the response from various relief organizations to avoid conflicts. Each request is assigned to a state ranging from available to delivered. ACT can also organize requests based on which organization responded to the request, when the request was addressed, and whether any further action must be taken.

How does it work?

ACT (ASU Coordination Tracker) is an open disaster relief coordination system. It contains five functional modules: request collection, request analysis, response, coordination, and situation awareness. Raw requests from users are collected via crowdsourcing and groupsourcing. Based on the demand and disaster background, raw requests are analyzed and classified into various categories, and stored in a requests pool. The system visualizes the requests pool on its crisis map. Organizations respond to requests and coordinate with each other through the crisis map directly. A statistics module runs in the background to help track relief progress.

Publications

  • "Harnessing Crowdsourcing Power of Social Media for Disaster Relief", Under Review by IEEE Intelligent Systems (CPSS).
  • "Promoting Coordination for Disaster Relief - From Crowdsourcing to Coordination"”", Social Computing, Behavioral-Cultural Modeling, and Prediction (SBP 2011).
  • "Making Social Media Work for Humanitarian Assistance and Disaster Relief" (Poster Paper). HSCB Focus 2011. February 8-10, 2011. Chantilly, Virginia.

ACT is built at the Data Mining and Machine Learning Lab at Arizona State University.

Photo
Huiji Gao is a Ph.D. student working at the Data Mining and Machine Learning Lab.
Photo
Mohammad-Ali Abbasi is a Ph.D. student working at the Data Mining and Machine Learning Lab.
Photo
Dr. Huan Liu is the director of the Data Mining and Machine Learning Lab.

Contact Us: (480) 727-7349 | 699 S. Mill Ave. Suite 501, Tempe, AZ 85281

DMML Logo

This project is funded by the Office of Naval Research.