Entry Name: "TJU-Li-MC2"

VAST Challenge 2015
Mini-Challenge 2

Team Members:

Kang Li, Tianjin University, China, lanelim@foxmail.com PRIMARY

Fan-Rui Sun, Tianjin University, China, frsun@foxmail.com

Jie Li, Tianjin University, China, vassilee@tju.edu.cn SUPERVISOR

Kang Zhang, Tianjin University, China, kzhang@tju.edu.cn SUPERVISOR

Student Team: YES

Did you use data from both mini-challenges? NO

Analytic Tools Used:

MySql

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

About 60 hours (60 days and 1 hours per day)

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

Video:

MC2.wmv

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

Approach Description

We attempt to use a multi-attribute ranking visualization to resolve the MC 2, as shown in Fig. 0.

(e) 5-attribute

(d) 2-attribute

(c) 3-attribute

(b) 2-attribute

(a) 1-attribute

Fig. 0. Multi-attribute ranking. (a-c) illustrates a quadtree metaphor. Each level represents an attribute and is divided into 4 groups having the same number of IDs. The grayscale of a grid represents the number of IDs belonging to the grid. (d) Five important attributes. (e) An example containing all the five attributes, that uses grayscale to represent the number of IDs. By observing the grayscale distribution, various communication patterns can be identified.

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?

a. Characterize the communication patterns you see.

We identify the top five IDs who have the largest sum of all the sent and received messages each day.

Communication pattern 1:

Examples: 839736

The five attributes of this type of IDs are all at level 1, and tended to send group messages, and sent and received roughly equal amount of messages. They always replied to the

messages they received and also received replies whenever they sent.

We believe that these IDs are the site staff who maintain various facilities on site.

Communication pattern 2:

Examples: (Fri: 1508923)(Sat:1351786 and 1292409)(Sun: 620184, 19249 and 1952914)

This type of IDs tended to send large volumes of messages, only during the park’s opening hours.

We believe that these IDs are security staff, site director, tour guides, or tour organizers.

Communication pattern 3:

For instance, (Fri: 1278894)(Sat: 1278894)(Sun: 1278894)

This type of IDs showed unusual behaviors. They sent large volumes of messages to a large number of people regularly every 2 hours.

We believe that these IDs machines that kept tracking the number of visitors entering the park and the number of visitors not using any facilities.

Communication pattern 4:

For instance, (Fri: 825466 and 809736)(Sat: 2082743)

This type of IDs tended to send group messages only when the park is open, and sent to far more people than those they received from.

We believe that these IDs are visitors who liked to initiate communication.

(a) Friday

(b) Saturday

Fig.1-1 Top 5 IDs and their attributes including the volumes of sending and receiving messages each day and the sum of the first two attributes.

(a) The top 5 IDs’ detailed information on Friday.

(b) The top 5 IDs’ detailed information on Saturday.

Fig.1-2 The top 5 IDs’ detailed information on three days.

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

839736, 1278894 => machines

1508923, 1351786, 1292409 => staff

825466, 809736, 2082743, 620184, 19249, 1952914 => visitors

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.

1. This type of IDs is one of the major types of visitors, mostly engaged in one-to-one communication, with a few group messaging.

Fig.2-1 Communication pattern 1

2. This type of IDs is one of the major types of visitors, frequently engaged in one-to-one communication.

Fig.2-2 Communication pattern 2

3. This type of IDs sent mass group messages frequently.

We believe that they are tour organizers, tour guides and staff of the park.

Fig.2-3 Communication pattern 3

4. This type of IDs is one of the major types of visitors, engaged mostly in one-to-one communication, with a few group messaging. They sent far more messages than they received.

We believe that this type of visitors made most use of the park facilities and liked to initiate communication.

Fig.2-4 Communication pattern 4

5. This type of IDs always engaged in one-to-one communication, low volumes of messages. They sent messages upon receiving from those who sent messages every two hours.

We believe that they are mostly single or small-group visitors.

Fig.2-5 Communication pattern 5

6. This type of IDs primarily engaged in group messaging, and received far more messages than they sent, particularly high during the opening hours.

We believe that these IDs are lottery machines, staff handling complaints, tour organizers, and cheer leaders.

Fig.2-6 Communication pattern 6

7. This type of IDs never receives any messages, and sent few messages, always on one-to-one communication.

We believe that these IDs are single visitors who may communicate with the outside, or the people with special purposes.

Fig.2-7 Communication pattern 7

8. This type of IDs sends to and receives from exactly the same set of IDs, sends and receives exactly one message to each of their communication targets.

We believe that these IDs either had no desire to communicate, only communicating with staff, or with special purposes since they minimized their communication to hide themselves.

Fig.2-8 Communication pattern 8

9. This type of IDs liked to send messages to others in high volume, yet ignored part of the received messages.

We believe that these IDs were people with special purposes or visitors.

Fig.2-9 Communication pattern 9

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

Minute

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

Minute

Hour

s

(a) Calendar (b) Clock metaphor

Fig.3-1. Two types of calendar Heatmap.

The periods on Sunday highlighted by two yellow rectangles in Figure 3-1 (a) illustrate that, an abnormal communication pattern appeared, clearly distinct from patterns at the same time on Friday and Saturday, with over twice communication volumes more than the other two days. We therefore suspect that the incident was discovered at around 11:36 on Sunday (pointed to by the short black arrow on the clock metaphor in Figure 3-1 (b)). At 12 o’ clock on Sunday (pointed to by the long red arrow), a large volume of communication data occurred continuously for about 30 minutes, spreading the news, via broadcast channels by staff ID 839736. The communication messages were to inform all the park staff of the event and explain to all the visitors.