Entry Name: UCAS-Zhu-MC2

VAST Challenge 2015
Mini-Challenge 2

Team Members:

Yuefan Zhu, Chinese Academy of Sciences, zhuyuefan_ncepu@163.com PRIMARY

Jingjing Fu, Chinese Academy of Sciences, fujingjing@iie.ac.cn

Chao Wang, Chinese Academy of Sciences, 251995640@qq.com

Yi Du, Chinese Academy of Sciences, duyi@cnic.cn SUPERVISOR

Danhuai Guo, Chinese Academy of Sciences, guodanhuai@cnic.cn SUPERVISOR

Student Team: YES

Did you use data from both mini-challenges? NO

Analytic Tools Used:

Excel

MATLAB

Python

Data-Driven Documents

Approximately how many hours were spent working on this submission in total?

128 hours

May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2015 is complete? YES

Video:

UCAS-Zhu-MC2.wmv

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Questions

MC2.1 – Identify those IDs that stand out for their large volumes of communication. For each of these IDs

a. Characterize the communication patterns you see.

b. Based on these patterns, what do you hypothesize about these IDs?

Limit your response to no more than 4 images and 300 words.

Look at picture 1.1:

These are the top ten IDs of communication for each day in the whole park. The larger the area is, the greater the amount of communication is.

Picture 1.1

ID: 1278894 and 839736

According to picture 1.1, the two IDs had quite a lot of communication (ranking first in all IDs).

For the two IDs, we refined their behaviors:

Look at picture 1.2: The first two pie charts show the proportion of the three days of communication for each ID. The third and fourth one show the proportion of the regional communication volume.

Picture 1.2

In picture 1.2, we found that they appeared every day and communicated only in Entry Corridor. Their communication volume changed with the number of visitors.

We hypothesize that the two IDs were staffs working at the gate whose job was to install APP for visitors.

ID: 1508923

In picture 1.1, we can find ID 1508923 ranks third on Friday. Similar to picture 1.2, we can get the proportion of the regional communication volume of ID 1508923. It appeared every day. And it had different amount of communication in different area each day.

We hypothesize that ID 1351786 was a normal visitor.

ID: 312564

Look at picture 1.3:

Picture 1.3

It ranks first among the communication volume in Coaster Alley on Friday. Similar to picture 1.2, we can get the proportion of the regional communication volume of ID 312564. We find that it had little communication on Sunday and had large amount of communication in Wet Land in three days.

We hypothesize that ID 312564 was a person who participated in the crime.

ID: 19249

It communicated a lot in Coaster Alley. And it communicated quite a lot on Sunday morning in Wet Land.

We hypothesize that ID 19249 was a big fan of Scott Jones.

ID: 1351786

In picture 1.1, we can find ID 1351786 ranks fourth on Saturday. Similar to picture 1.2, we can get the proportion of the regional communication volume of ID 1351786. It appeared only one day. And it appeared at every part with approximately equal amount of communication in each part.

We hypothesize that ID 1351786 was a normal visitor.

MC2.2 – Describe up to 10 communications patterns in the data. Characterize who is communicating, with whom, when and where. If you have more than 10 patterns to report, please prioritize those patterns that are most likely to relate to the crime.

Limit your response to no more than 10 images and 1000 words.

Pattern 1:

Communication pattern: people who vandalized the pavilion.

They sent text messages to anyone within their designated group. They wandered inside and outside Creighton Pavilion to make a criminal plan. Look at picture 2.1(the proportion of the regional communication volume in the criminal group), they made massive communication in Wet Land and Coaster Alley (which surrounded Creighton Pavilion) on Friday to make the plan. They communicated a lot in Wet Land on Saturday to look for the opportunity to commit the crime. They contacted with each other frequently in Wet Land on Sunday to find the latest development.

Picture 2.1

Pattern 2:

Communication pattern: staff members

They installed APP for visitors. They communicated with visitors who hadn’t installed the APP when visitors checked in the amusement park in Entry Corridor.

Pattern 3:

Communication pattern: big fans of Scott Jones

They sent text messages to anyone within their designated fan group. They added other fans they met to their fan group when they watched the stage show led by Scott Jones. They communicated a lot when they visited Creighton Pavilion which exhibited the awards, trophies, and the Olympic Gold medal accumulated by Scott Jones. They mostly communicated in Coaster Alley.

Pattern 4:

Communication pattern: family members

Look at picture 2.2: in this picture, there are two family groups who communicated a lot in Kiddle Land. Red points means that this ID is active in Kiddle Land. Green points means that this ID is not active in Kiddle Land. Pie chart shows the ratio of these two types of IDs.

Picture 2.2

Family members sent text messages to anyone within their designated family group. They communicated frequently when they waited for kids on various rides in Kiddle Land. They communicated in Kiddle Land most of the time because their children preferred to stay there. They also communicated a lot before lunch time to find a place for a meal.

Before they left the amusement park, there was a communication peak to get together.

Pattern 5:

Communication pattern: friends

Friends sent text messages to anyone within their designated friend group. They communicated with each other when they found something interesting especially in a pavilion or a show. They communicated before lunch time to have meal together. They also contacted with each other when there’s some big news (For example, the sudden closure of Creighton Pavilion). Before they left the amusement park, there was a communication peak to get together.

MC2.3 – From this data, can you hypothesize when the crime was discovered? Describe your rationale.

Limit your response to no more than 3 images and 300 words.

We hypothesize the crime was discovered between 11:00 a.m. and 11:28 a.m. on Sunday. That is eleven to eleven twenty-eight in the morning of June 8th 2014.

Picture 3.1

Look at picture 3.1. These are communication volume curves in Coaster Alley Area on Friday, Saturday and Sunday (Creighton Pavilion and Grinosaurus Stage are located in Coaster Alley). There was a peak of communication at both 11 a.m. and 4 p.m. on Friday and Saturday. But on Sunday, we can only find one communication peak at 11 a.m. Based on this, we hypothesize Scott Jones was scheduled to appear at his stage show at 11 a.m. and 4 p.m. in these three days. When shows began, many visitors checked in the Grinosaurus Stage. The “check-in” messages and related communication texts brought about the communication peak. However, a crime was committed on Sunday, which forced the closure of Creighton Pavilion and the cancel of stage show in Grinosaurus Stage. Therefore, we lost a “check-in” peak at 4 o’clock on Sunday afternoon in Coaster Alley.

Look at picture 3.2. These are communication volume curves in Wet Land Area on Friday, Saturday and Sunday. Once you come out from the only exit of Creighton Pavilion, you’ll enter the Wet Land Area. And the communication curve of Sunday in Wet Land has an significant exception (compared to curves of Friday and Saturday). The communication volume has a notable increase at 11:28 a.m. Based on this, we assume that the crime was discovered between 11:00 a.m. and 11:28 a.m. and then visitors were forced to leave without check-out. These visitors quickly spread out this news when they were driven into Wet Land, which caused an obvious communication increase in Wet Land.

Picture 3.2