TY - GEN
T1 - Little data, big stories
T2 - 2nd IEEE International Conference on Collaboration and Internet Computing, IEEE CIC 2016
AU - Dudas, Patrick M.
AU - Weirman, Samantha
AU - Griffin, Christopher
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2017/1/6
Y1 - 2017/1/6
N2 - With the proliferation of Big Data, Social Science projects being developed, this work takes a step back to design research avenues that specifically look at smaller, real-time Social Science projects. Building on an already developed platform, called Dynamic Twitter Network Analysis (DTNA), we build out exploration into multiple world event types, which were captured in real-time and used smaller datasets to allow the user the ability to seek location and topic-specific data collections in parallel to events occurring. With these datasets, we first establish what could be learned during the event that mimics larger projects in the same domain. Secondly, we compare the events to help bring awareness to strategies that can evolve as specific events occur. The datasets examined are from a 24-hour period from specific locations of relevance with a focus on polarizing events. This includes: 1)Boston Marathon Bombing, 2) Sandy Hook Elementary Shooting, 3) Gezi Park Riots, 4) Hurricane Sandy, 5) Batkid, Make-a-Wish Foundation, 6) Brazil World Cup Protests, and 7) 2014 NBA Championship (Game 5). These networks will be analyzed both from social network analysis (SNA) and natural language processing (NLP) approaches (including sentiment analysis and part of speech tagging comparing personal pronoun use).
AB - With the proliferation of Big Data, Social Science projects being developed, this work takes a step back to design research avenues that specifically look at smaller, real-time Social Science projects. Building on an already developed platform, called Dynamic Twitter Network Analysis (DTNA), we build out exploration into multiple world event types, which were captured in real-time and used smaller datasets to allow the user the ability to seek location and topic-specific data collections in parallel to events occurring. With these datasets, we first establish what could be learned during the event that mimics larger projects in the same domain. Secondly, we compare the events to help bring awareness to strategies that can evolve as specific events occur. The datasets examined are from a 24-hour period from specific locations of relevance with a focus on polarizing events. This includes: 1)Boston Marathon Bombing, 2) Sandy Hook Elementary Shooting, 3) Gezi Park Riots, 4) Hurricane Sandy, 5) Batkid, Make-a-Wish Foundation, 6) Brazil World Cup Protests, and 7) 2014 NBA Championship (Game 5). These networks will be analyzed both from social network analysis (SNA) and natural language processing (NLP) approaches (including sentiment analysis and part of speech tagging comparing personal pronoun use).
UR - http://www.scopus.com/inward/record.url?scp=85013187196&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85013187196&partnerID=8YFLogxK
U2 - 10.1109/CIC.2016.071
DO - 10.1109/CIC.2016.071
M3 - Conference contribution
AN - SCOPUS:85013187196
T3 - Proceedings - 2016 IEEE 2nd International Conference on Collaboration and Internet Computing, IEEE CIC 2016
SP - 474
EP - 482
BT - Proceedings - 2016 IEEE 2nd International Conference on Collaboration and Internet Computing, IEEE CIC 2016
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 1 November 2016 through 3 November 2016
ER -