Jiawei Zhang,
Purdue University, zhan1486@purdue.edu PRIMARY
Shehzad Afzal, Purdue University, safzal@purdue.edu
Dallas Breunig, Purdue University, dbreunig@purdue.edu
Jing Xia, State Key Lab of CAD&CG, Zhejiang University, xiajing@zjucadcg.cn
Jieqiong Zhao, Purdue University, jieqiongzhao@purdue.edu
Isaac Sheeley, Purdue University, isheeley@purdue.edu
Joseph Christopher, Purdue University, christ33@purdue.edu
David Ebert, Purdue University [Faculty Advisor], ebertd@ecn.purdue.edu
Chen Guo, Purdue University, guo171@purdue.edu
Shang Xu, Purdue University, xu537@purdue.edu
Jun Yu, Purdue University, yujun_2005@hotmail.com
Qiaoying Wang, Purdue University, wang1925@purdue.edu
Chen Wang, Harbin Institute of Technology, chwang@hit.edu.cn
Zhenyu Qian, Purdue University [Faculty Advisor], qianz@purdue.edu
Yingjie Chen, Purdue University [Faculty Advisor], victorchen@purdue.edu
Student Team: YES
Our
work utilized and extended SMART system, done by Purdue University VACCINE
Center for social media visual analytics.
Approximately how many hours were spent
working on this submission in total?
180
May we post your submission in the
Visual Analytics Benchmark Repository after VAST Challenge 2014 is complete? YES
Video:
http://pixel.ecn.purdue.edu:8080/~zhan1486/Purdue-Zhang-MC3/Purdue-Zhang-MC3.wmv
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Questions
Please note - this challenge contains
a question that is time-dependent.
Within 3 hours of starting the final data stream, send an email to VASTChal2014MC3@vacommunity.org containing your answer to question
MC3.1. Please include a copy of your
answer to MC3.1 in your final answer form also. Your answers to MC3.2 and
MC3.3, along with your video, are due July 8.
The
responses to these questions should be incrementally built, as you (the
contestant) acquire information from each streaming data segment you
receive. Your submission will answer these questions in consideration of
all of the streaming data segments.
MC3.1 -
Within 3 hours after start the final data stream, send an email to VASTChal2014MC3@vacommunity.org containing:
a.
An image showing the streaming data in
your visual analytics tool. In this image, identify an event of interest that
you intend to investigate further.
b. The content of the final message in the data stream
The image showing the streaming data in the visual analytics
tool:
Fig 1: screenshot of
SMART system showing real time analysis of streaming data. In this image, the
user is able to identify fire and standoff, the two major events during the
time period from 20:00 to 20:30.
The content of the final
message in the data stream:
Time:
2014-01-23 21:32:00
User
name: gardener4958
Message:
RT @KronosStar There has been an explosion from inside the apartment building.
Several people are down. #KronosStar #DancingDolphinFire #AFDHeroes
MC
3.2 - Describe the timeline of up to five major events that you discover in the
streaming data. This timeline should include information from all three
segments of the data stream if needed. Use specific microblog
records and call center data to support your description, but do not simply
mimic back the data stream. Provide a concise description of important
participants, locations and durations. Focus your response on the events
themselves, rather than on the individuals reporting the events. Please limit your
answer to no more than ten images and 1500 words.
We
discovered 4 major events in the 3 data segments: POK rally, fire &
explosion, hit & run, and shooting & standoff.
Fig 2: POK rally held in
Abila City Park. Part of Parla St was closed due to concerns of public safety.
Several key roles appeared in the rally including Sylvia Marek and Lucio Jakab.
Some news media like abilapost actively posted updates of the rally. The system
consists of multiple linked views including a) time series view; b) topic view;
c) retweet/reply networks view; d) call center data table; e) microblog data
table and f) map view. In time series view, blue bar charts represent microblog
data, and red line charts represent call center data. In map view, red dots
denote call center data and blue dots denote microblog data.
Fig 3: Fire event
occurred at Dancing Dolphin Apartment at around 18:41.
Fig 4: A firefighter was
injured in the fire event. Multiple linked views reveal the fact accordingly.
Fig 5: An unexpected
explosion occurred at Dancing dolphin Apartment, at around 21:30.
Fig 6: A black van hit a
car and a bicyclist, and escaped. Based on geo-located call center data and
microblog data, the system helps investigate the possible trajectory of the
suspect.
Fig 7: The black van
chased by police was blocked at Gelato Galore and then opened fire.
Fig 8: The evolution of
the shooting & standoff events.
MC 3.3 –
Select one of your five major events from question MC 3.2 that you consider to
be most likely to provide additional clues to the investigation of the GASTech
disappearances. Describe the roles of the participants.
Describe how other events you identified in MC3.2 may have influenced your
selected event. Provide a hypothesis and evidence as to whom you suspect as
being directly involved in the GAStech
disappearances, either as perpetrators or victims. Please limit your
response to no more than five images and 500 words.
According
to the timeline described in MC3.2, the shooting & standoff events provide
clues to the GASTech disappearances. Based on monitoring the microblog and call
center data traffic we think the disappearances of GASTech employees are due to
the following two reasons:
However,
all the above assumptions should be doubted due to the subjectivity of
microblog users and lack of validation from multiple data sources.
In
addition, all major events happening during the data stream are most likely
related. Here is some evidence: (Fig.9)
Based
on the evidence above, it is possible the POK members kidnapped two employees
of GASTech and stayed somewhere near the Dancing Dolphin Apartments. Due to the
fact that the hit & run was close to the Dancing Dolphin Apartments, it can
be assumed the hit & run happened very soon after the kidnappers left their
hideout. However, because the hit & run happened a significant period of
time after the beginning of the fire, the kidnappers most likely were hiding in
a location near the Dancing Dolphin Apartments and were “flushed” out of their
hiding spot either by the crowd and heightened activity or by police officers
ordering them to evacuate. The kidnappers most likely hit the car and the
bicyclist by accident while trying to flee. It can be assumed the perpetrators
of the hit & run and the shooting & standoff crisis are the same. It is
possible the suspects travelled the route near the site of the POK rally
because there were large crowds, which they thought may help them escape the
pursuit of police officers.
In
addition, the explosion following the fire event at Dancing Dolphin Apartments
seems suspicious. The explosion occurred less than 10 minutes after the fire
was under control. Some people were injured by the explosion. It can be
possible that the explosion was caused by a gas leak, or by a deliberate
attack. For a deliberate attack, it is possible that the fire event was also a
man-made terror. Furthermore, the terrorists may try to murder victims by a
fire event, which was later controlled by the firefighters. Finally they
carried out the bombing attack.
Fig 9: Analysis of
correlations among uncovered major events based on the timeline and spatial
distribution of call center records.
Fig 10: Retweet/reply
networks formed by users who mentioned GASTech related topics. Influential
microblog users (having high degree centrality in the networks) like
officia1abilapost and friendsofkronos provide clues to help explain the
disappearance of GASTech employees. However, the statements should be doubted
due to the subjectivity of microblog users and lack of validation from multiple
data sources.