BIG DATA INNOVATION FOR SOCIAL IMPACT Hemant Purohit Assistant Professor, Information Sciences and Technology 1 @hemant_pt ICICT-2015 Keynote, Oct 9 International Congress on Information & Communication Technology Computer Society of India, ACM Udaipur
21
Embed
ICICT-15 keynote: Big Data Innovation for Social Impact, Hemant Purohit
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
BIG DATA INNOVATION FOR
SOCIAL IMPACT
Hemant PurohitAssistant Professor, Information Sciences and Technology
1@hemant_pt
ICICT-2015 Keynote, Oct 9International Congress on Information & Communication Technology
Computer Society of India, ACM Udaipur
20th Century Connectivity
2@hemant_pt
21st Century Connectivity
3@hemant_pt
Consequence: Data Deluge
4@hemant_pt
Data Deluge => Innovation Opportunity
5@hemant_pt
WHY, WHAT &HOW to leverage?
ORGANIZATIONS CITIZEN COMMUNITIES
6@hemant_pt
Overload of Citizen-‐generated data
ORGANIZATIONS CITIZEN COMMUNITIES
7
Structured Roles & Defined Tasks
Highly Unstructured
@hemant_pt
Lets tap citizen behaviors that meet organizational information needs ..
Social Data: Goldmine of Human Behavior!
Uni-directional communication(TO people)
Bi-directional(BY people, TO people)
Web 2.0media
8@hemant_pt
BUT .. Mining Behavior is challenging
9
Human Behavior
Emotion
Intent
AttitudeBelief
Sentiment
@hemant_pt
Social data: Multimodal, ambiguous, unconstrained natural language
USE-CASE-1. Can Societal Belief Modeling inform Policy Design?
10@hemant_pt
USE-CASE-1. Can Societal Belief Modeling inform Policy Design?
• Surveys:– Static by design– Time & Resource costs
• Social Data:– Dynamic & Cheap commodity
11@hemant_pt
@User15 no place is THAT safe and I don't really want to take the risk especially
when I'm alone :((( rape and robbery are very …
Purohit, et al. (2015). Gender-based violence in 140 characters or fewer: a #BigData case study of Twitter. arXiv preprint arXiv:1503.02086.
USE-CASE-2. Can Disaster Response Orgs leverage Social Data at Scale?
12@hemant_pt
20M+ Tweets during #Sandy
HURRICANE SANDY RESPONSE: $60+ Bln in damage
Does anyone know how to donate clothes to hurricane #Sandy victims?
Consequence: Creation of 2nd Disaster
• Resource demand-supply mismatch
Hurricane Sandy, “Thanks, but no thanks”, NPR, Jan 12 2013http://www.npr.org/2013/01/09/168946170/thanks-but-no-thanks-when-post-disaster-donations-overwhelm
@hemant_pt 13
DISASTER Event
@hemant_pt
Solution: Mine Big Social Data to Improve Disaster Coordination
CITIZENSensors
14
RESPONSE Organizations
DISASTER Event
2.a.) Filtering Big Disaster Data: Intentions
15
CITIZENSensors
RESPONSE Organizations
Me and @CeceVancePR are coordinating a clothing/food drive for families affected by Hurricane Sandy. If you would like to donate, DM us
Does anyone know how to donate clothes to hurricane #Sandy victims?
[SEEKING
[OFFERING
Intent-Classifiers as a Service
Purohit et al. (2013) First Monday, & Purohit et al. (2014) JCSCW@hemant_pt
DISASTER Event
2.b.) “We TRUST people!” => Filtering Users to Engage
16
CITIZENSensors
RESPONSE Organizations
Tool to mine Important users
@hemant_pt Purohit et al. (2014) ISCRAM
Want to help animals in #Oklahoma?
@ASPCA tells how you can help:
http://t.co/mt8l9PwzmOx
RESPONSE TEAMS (including humanitarian
org. and ‘pseudo’ responders)
VICTIM SITE
Where do I go to help out for volunteer work around Moore? Anyone know?
Anyone know where to donate to help the animals from Oklahoma
disaster#oklahoma #dogs
Matchable
Matchable
If you would like to volunteer
today, help is desperately needed in
Shawnee. Call 273-5331 for more info
Future: Online Match-making Crisis Coordination
17
CITIZEN SENSORS
Need Offer
@hemant_pt 17Purohit et al. (2013) First Monday
Exemplary Cyber-Human System: Transforming Public Goodwill into Action
18
@InCrisisRelief : presented by @parrySingh
@hemant_pt Purohit et al. (2015) Empowering Crisis Response-led Citizen Communities. IGI Global
Time to Start modeling Citizen Behavior for Social Impact!
• Data challenges:
19
Ambiguity
Sparsity
Diversity
Scalability
@hemant_pt
Need for the Solution at the intersection
Conclusion
• Start from organizational information needs to create Big Data systems for social impact
• Interdisciplinary Research a necessity!
• Less is More! -- We need to focus on actionable data, not Big Data deluge