Entry Name: ”HKUST-SiWei-MC1”

VAST Challenge 2015
Mini-Challenge 1

Team Members:

Si Wei Fu, Hong Kong University of Science and Technology, fusiwei339@gmail.com

Shao Yu Chen, Hong Kong University of Science and Technology, schenan@connect.ust.hk

PURI Abishek, Hong Kong University of Science and Technology, apuri@connect.ust.hk

Tian Yu Wang, Hong Kong University of Science and Technology, twangad@connect.ust.hk

Yeuk Yin Chan, Hong Kong University of Science and Technology, yychanae@connect.ust.hk

Dong yu Liu, Hong Kong University of Science and Technology, ustdongyu@gmail.com

Hua min Qu, Hong Kong University of Science and Technology, huamin@cse.ust.hk

Student Team:

Yes

Did you use data from both mini-challenges?

Yes

Analytic Tools Used:

Visio

ParkVis, developed by student team led by HKUST VisGroup for the challenge

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

100 Hours

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

Yes

Video Download

Video:

See the supplementary files.

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

Questions

MC1.1 – Characterize the attendance at DinoFun World on this weekend. Describe up to twelve different types of groups at the park on this weekend.

a. How big is this type of group?

b. Where does this type of group like to go in the park?

c. How common is this type of group?

d. What are your other observations about this type of group?

e. What can you infer about this type of group?

f. If you were to make one improvement to the park to better meet this group’s needs, what would it be?

Limit your response to no more than 12 images and 1000 words.

Group 1:

a) This kind of group has 8 people, and this kind has only one group like that.

b) They seem to like to stay around the Grinosaurus Stage, as they always go straight to the stage then leave then go to stage then leave again (Fig 1.1).

c) We observed only one type of group like this.

d) They walk all around the park, only stopping at the stage. They also enter the park twice on the first two days, but only once on Sunday.

e) They seem to be patrolling the park, going all around. Because they stop at the stage at the same time as the performances by Scott Jones, they might be his security detail, entering and leaving with him.

f) If they are a security detail, it would be easier for them to have a designated path to follow, and have that path made exclusive for them. However this is impractical as it would be hard to just cordon off one area of an amusement park.

Fig 1.1 Maps and Group View showing path of the group members

Group 2:

a) This kind of group commonly has 1 to 5 people.

b) They seem to like the restaurants and shops in the park, as they do not check in to any rides, just enter and go straight to the shopping and eating area (Fig 1.2).

c) Over the course of three days, such groups are observed less than 10 times.

d) They do not stay in the park for a long time, simply going for a meal then leaving. Their paths are slightly varied, each group taking different routes to get to the shopping area, some eating then shopping others vice versa.

e) They might be coming to the park because the shops and food here are better than in the nearby area, or maybe they work near that area so only stay there and go no where else.

f) For visitors who are only interested in eating and shopping, they would like to buy bulk food coupons at a discounted rate, saving money if they come regularly.

Fig 1.2 Maps and Group Views showing different paths of different IDs

Group 3:

a) This kind of group commonly has 1 – 10 people.

b) They seem to really like going on thrill rides, as they go on more than 8 of the thrill rides in a day. Their paths also indicate that the go on some rides 2-3 times, indicating a strong interest in thrill rides (Fig 1.3).

c) We observed many groups falling into this category (Fig 1.6), making them quite a common occurrence.

d) The members of this group have quite high numbers of thrill rides relative to the rest of the parks visitors. In Fig 1.3 you can see that one group has an average distribution of almost 4 tries per ride, which is quite high for one day. They also tend to not go on any other rides, showing a strong inclination towards thrill rides.

e) Most likely the people composing this group are thrill junkies, enjoying the adrenaline rush that comes from going on the thrill rides.

f) As these groups tend to avoid other types of rides, they would benefit from a discounted thrill ride pass, which only allows them to go on thrill rides. This way maybe more of these groups would come, gaining revenue for the park and benefiting the people in these groups.

Fig 1.3 Group Views showing details of different IDs

Group 4:

a) This kind of group commonly has 1 – 2 people.

b) They do not have any particularly favorite location. They just walk around the park path all day.

c) They are fairly common, being seen tens of times on each of the three days.

d) Mostly they are simply walking around the path, but sometimes they will stay near a ride but will not check in to the ride, indicating that they are not playing the ride merely waiting around it.

e) As they don’t seem interested in doing anything particular, simply walk around the path the whole day, they might be the elderly citizens simply enjoying the parks atmosphere.

f) As they seem particularly interested in the parks atmosphere, the park should add more greenery by planting more flowers and trees all over the park and opening an area to see the ecology such as a greenhouse or a bird-viewing platform. This idea would really increase the enjoyment of the group members.

Fig 1.4 Group View and Map showing paths of different IDs

Group 5:

a) This kind of group commonly has 20 - 42 people.

b) They like to go to the thrill rides most, and different groups have various preferences.

c) They are detected everyday, about 10 such kind of groups are found everyday.

d) They play together and most communication happened within the group. Most of them come to the park for only one day, and most of them only have inner-group connections.

e) Maybe they are classmates and make fun together.

f) The park may provide group tickets for them.

Fig 1.5 Group View and MDS View showing a big group of people

Group 6:

a) This kind of group commonly has 3 - 10 people.

b) They like to go to the thrill rides most, and different groups have various preferences. We have found one group of this kind that has only play kiddie rides.

c) They are detected very common. About 100 group every day.

d) Only two or three people in the group have inner-group communication or communication with two IDs 839736 and 1278894 who are suspected to be park officers in the park. Most people in the group do not have any communication in the park. They do not stay in the park for a long time, and most of them go back home early.

e) Maybe they are families and kids do not use the APP to send out messages.

f) If they are family with kids, the park could provide specific devices to them so that parents can track their children’s path.

Fig 1.5 The MDS View, Group View and Map View for a “Family-like” group

Group 7:

MC1.2 – Are there notable differences in the patterns of activity on in the park across the three days? Please describe the notable difference you see.

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

One of the main differences we found was that the total number of visitors to the pavilion on Sunday was significantly lower than on the other two days, even though traffic on Sunday was much higher (Fig 2.2). This is due to the fact that after 12 pm on Sunday, there is no one checked into the pavilion until closing (Fig 2.1). This is likely related to the closing of the pavilion after the crime happens.

Another notable difference is that on Friday and Saturday, the stage has two peak periods starting at 9:30 am and 2:30 pm respectively (Fig 2.1). However, on Sunday, the stage only has one peak period, starting at 9:30 am (Fig 2.1). This is likely related to the news we are given that Scott Jones cancelled the second show after the crime happened, so no one went to the stage at 2:30 pm unlike the previous two days.

The last difference we found was that on Friday, there is a significantly lower peak level of traffic for thrill rides than on the other two days, even when accounting for overall park traffic differences. On Friday, percentage of people at peak time on thrill rides is 21% (Using data from Fig 2.2, 2.3). On Saturday and Sunday, however, it is 32% and 31% respectively (Using data from Fig 2.2, 2.3).

Fig 2.1 Graphs Showing traffic of the park at the stage (left) and pavilion (right) on different days

Fig 2.2 Line Graphs showing traffic of total park on different days

Fig 2.3 Graph showing traffic at the thrill rides on different days

MC1.3 – What anomalies or unusual patterns do you see? Describe no more than 10 anomalies, and prioritize those unusual patterns that you think are most likely to be relevant to the crime.

Limit your response to no more than 10 images and 500 words.

1. ID 1983765 exhibits unusual behavior on Sunday, after exhibiting normal behavior on Friday and Saturday. This is seen through a deviation in its path on Sunday as well as through a general lack of connections to other park visitors and external. On Sunday he visited the pavilion quite early and then stayed in one of the rides for a long time. Considering that the pavilion was the scene of the crime on that day, this is suspicious behavior.(Fig 3.1).

2. One extremely unusual pattern is that some people somehow get into the park without checking in at any entrance gate. This might indicate that they have somehow managed to hack into the app, increasing the chance that such individuals are the criminals. In Figure 3.2, you can see that there are no red check-in boxes anywhere on the map, indicating that the ID did not check in at any entrance gate.

3. The pavilion is empty from 10 to 11:30 am and 3 to 4:30 pm on Friday and Saturday, but on Sunday there are 2 people in the pavilion between 10 to 11:30 am. It is unusual for fans to not be at the stage to see the performance. (Seen in Fig. 2.1).

4. There are two people who go on rides on Friday and Sunday, but on Saturday they do not go on any rides. In Fig 3.3 you can see this unusual pattern.

5. There is a group of people who only play children rides, no other type of ride. This is anomalous behavior; most people in the park play the thrill and the rides for everyone even if they only come for one day. You can see in Fig 3.4 that they do not go to any rides except for the kiddie ones.

6. One unusual pattern was that not one person went on all the kid’s rides, even though many went on all the thrill rides. This is quite unusual as all the kids rides are right next to each other, however not one person went on all of them.

Fig 3.1 Workflow showing details about ID 1983765

Fig 3.2 Maps showing the movement of ID 657863 on Friday and Saturday

Fig 3.3 Group Views showing the details of two IDs on different days

Fig 3.4 Group View showing the details of four IDs on different days