Entry Name: “PKU360-Ye-GC”

VAST 2015 Challenge
Grand Challenge

Team Members:

Tangzhi Ye, Peking University, yetangzhi66@gmail.com PRIMARY

Youfeng Hao, Peking University, ajihyf@gmail.com

Zhenhuang Wang, Peking University, wangzhenhuang.zeek@gmail.com

Chufan Lai,Peking University, chufan.lai.1990@gmail.com

Siming Chen, Peking University, simingchen3@gmail.com

Xiaoru Yuan, Peking University, xiaoru.yuan@gmail.com

Jie Liang, Peking University, christy.jie@gmail.com

Zongru Li, Peking University, 1300013035@pku.edu.cn

Zhuo Zhang, Qihoo 360 Technology Co. Ltd, zhangzhuo@360.cn

Xin Huang, Qihoo 360 Technology Co. Ltd, huangxin-xy@360.cn

Zhanyi Wang, Qihoo 360 Technology Co. Ltd, wangzhanyi@360.cn

Chuanming Huang, Qihoo 360 Technology Co. Ltd, huangchuanming@360.cn

Fengchao Xu, Qihoo 360 Technology Co. Ltd, xufengchao@360.cn

Student Team: No

Analytic Tools Used:

PKU Space-Time explorer developed by Visualization research group at Peking University

360 visual analysis system developed by Qihoo 360 Technology Co. Ltd

Python pandas. http://pandas.pydata.org/

Python scikit-learn http://scikit-learn.org/stable/

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

200 hours

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

Video Download

Video:

http://vis.pku.edu.cn/vast2015/VC2015GC.mp4

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

Questions

For each of the following questions, consider both the movement and communications data.

GC.1 – Scott is not a paying customer and does not have an ID. Describe Scott Jones’ activities in the park during the three-day weekend. Who does he spend most of his time with? When does he arrive? When does he leave? What route does he follow?

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

Figure 1.1 #people stay in Grinosaurus Stage

Scott gave speech at No.63 Grinosaurus Stage regularly during the weekend. In figure 1.1, we found that two speeches were given from 10:00 a.m. to 11:00 a.m. and 15:00 p.m. to 16:00 p.m. on Friday and Saturday. On Sunday, however, the speech in the afternoon was not given as plan.

Then we found out 8 people in one group stayed with Scott in the park. They attended all the speeches over 3 days without any check-in records and didn’t go to any other POI. What’s more, their paths between east gate and Grinosaurus Stage were exactly the same over 3 days. These people’s IDs are 1787551, 1080969, 1935406, 1600469, 521750, 1781070, 1629516, 644885.

From their movement records, we could know the Scott’s route in the park.

Morning Route:

They came to the park through East Gate in the morning at 8:45 a.m.

Then they went to Grinosaurus Stage to attend the speech at 9:30 a.m. , which is 30 minutes before the speech started.

They left Grinosaurus Stage at 11:30 a.m., which is 30 minutes later than speech finished.

They left the Park and returned to the hotel, also through East Gate, at 12:15 p.m. to take a rest in the hotel.

Afternoon Route:

They came back to the park from hotel through East Gate at 13:45 p.m.

And they went to Grinosaurus Stage again to attend the second speech at 14:30 p.m., which is 30minutes ahead the speech started as well.

They left Grinosaurus Stage at 16:30 a.m., which is also 30 minutes later than speech finished.

Finally they left the Park through East Gate at 17.15 p.m.

Some essential time spots such as are marked in Figure 1.2, we also visualized Scott’s route on the map in Figure 1.4.

Figure 1.2 #Scott’s Group Routine on Saturday

We can see from the Figure 1.3, they followed the same route on Friday and Saturday as we described above. On Sunday, however, they didn't come back to the park in the afternoon.

Figure 1.3 #Scott’s Group Routine for 3 days

We draw the Scott path in the Figure 1.4. When he came to Grinosaurus Stage from East Gate followed path indicated by green arrow. He returned to East Gate from Grinosaurus Stage through same path with reverse direction indicated by blue arrow.

Figure 1.4 Scott’s path in the park

GC.2 – Identify up to 8 issues with park operations during the three-day weekend. Provide a rationale for your answers.

Limit your response to no more than 8 images and 800 words.

1. Food Capacity Issue: Park may increase the capacity of food shop, in order to meet the catering needs of visitors.

Figure 2.1 Food capacity issue, (a) people who didn’t eat in the park, went out seeking food, except one time had some sweet in the park; (b) Shilobite o’Pizza might run out of food or seat; (c) Mary Anning Beer Garden had more and more people in the last two days.

a) People from Camp and Hotel went outside for food at lunch and dinner time. (Figure 2.1 (a))

b) Some groups moved from one food shop to another, maybe because they could not get the seats or the shop run out of food. (Figure 2.1 (b))

c) Large sized group (30-45ppl) went to Beer Garden for food. It seems large groups could not fit in the size of food shops. (Figure 2.1 (c))

2. Policy for Trill Rides:

Figure 2.2.1 Comparison of people amount of Rides for Everyone (blue) and Trill Rides (purple)

Figure 2.2.2 Relationship between staying time and people amount for the TerrorSaur

There are significantly more people of Trill Rides than the Rides for Everyone, e.g. we can compare the people number of Ichthyoroberts Rapids4 and Fireball, we can see the difference from Figure 2.2.1. And for each Trill, when more and more people got together, the waiting time (inferred from staying time) increased (Figure 2.2.2).

In general, there were some rides of which staying(waiting) time was significantly longer than other rides in general, especially a number of popular rides: WenDisaurus Chase, Firefall and TerrorSaur.

3. Security staff issues:

a) There were fewer security staff at POI 32, before and during scott’s showcase at POI 63. It didn’t prevent the crimes from happening.

Figure 2.3 One people didn’t leave the park

b) A person with ID 657863 didn’t leave the park on Friday, his last record on the Friday was near No.20 rides. In the morning of Saturday, this person appeared in the near position, and left the park through North gate quickly. If there was no data error, we thought the park should increase the security power to carefully check people who didn’t leave the park (Figure 2.3).

4. Restroom issues:

Figure 2.4 People had to wait for the restrooms for a long time in the peak hour.

POI 66 - Tyrannosaurus Rest, in peak time especially on Sunday, e.g. 11:00, 14:30, it needed around 20-30 minutes to have all people finished using the restrooms. There must be a long queue waiting for using it. So we think the park can add some temporal toilets for the gathering activities next time.

5. Admissions issues: Park may provide other types of tickets, e.g. entry only(without rides), theater only, Scott show only.

Figure 2.5 Different type of tourists. (a) No ride checking, but only movement; (b) Scott’s fans; (c) People have interests others rather than rides.

a) There were a number of groups, who only visited the park, without any ride checking. (Figure 2.5(a))

b) Some Scott’s fans came only for his showcase and exhibition at Creighton pavilion. (Figure 2.5(b))

c) A number of groups were only interested in others rather fun rides, e.g.threater, shopping, eating. (Figure 2.5(c))

6. Park landscape issue: Park official may consider to re-structure the zones of map, by themes of interets, which include thrill, kiddle, everyone, show, and threater, and by function, which include food, rest area, shopping, and information.

Figure 2.6 landscape distribution in the park

a) Thrill Ride named No. 3 Auvilontops Express, No. 6 Keimosaurus Big Spin: Though waiting time was long in other thrill rides, there were not many people who came here. We found POI3 was located further at the end of road, in the northwest of Tundra Lan, where only had fewer number of visitors.

b) Many Thrill Ride lovers palyed a lot of thrill rides successively, but those rides were not placed at same region(Figure 2.6)

c) A type of group people went to many food stores successively, those food stores were not placed closely(Figure 2.6)

.

7. Show and theater schedule issue: Theaters and Scott’s show may need to be carefully planned

image:

Figure 2.7 #people in Sabretooth Theatre on Friday

a) We found some groups missed the theater shows, before or after Scott’s show.

b) Theatre finished an activity every 30 minutes, but many visitors check-in into the theatre during the activity. After an activity, many of them choose to stay in the theatre to attend the next activity. The park may need to schedule the theatre better to help visitors attend its activity on time.

8. Waiting on the Gate issues

Figure 2.8 Waiting in the gate and check-in in the gate

In Saturday and Sunday, the number of arrival people at the opening gate exceeds the amount of people that allows normal entrance. So we can infer the waiting situation in the gate.

GC.3 – For the crime, describe the following, and provide your rationale:

a. When did the crime occur?

b. Where did the crime take place?

c. Who are the most likely suspects in the crime?

Limit your response to no more than 5 images and 500 words.

The crime occurred during 10:00 - 11:30 on Sunday, in the Creighton Pavilion. The people who are most suspected with IDs 1502920, 416790 and 461004. They have another 4 partners, including 1123214, 1187909, 1350546 and 1000279. We believe that these are not security people due to their tourist-like behaviors (Figure 3.1). We provide our hypotheses about the crime as follows.

Figure 3.1 Seven people who are most likely suspects in the crime

These seven people came together and split into three teams to conduct the crime. In each step they entered a POI, they sent messages to each other to confirm something. At 9:00 on Sunday, 5 of them entered the Creighton Pavilion with check-in information. At about 9:30, 1123214, 1350546 and 1502920 came out, but 1502920 immediately returned to the pavilion. At around 10:00 when the pavilion was closed, all the 7 people had a brief communication. As an abnormal situation, 1502920, 416790 and 461004 stayed in the pavilion when it was locked up. The crime was most likely to take place between 10:00 and 11:30 when there were only the three people in the pavilion. Starting from 10:00, there were 37 people who didn't have check-in records, stayed around the pavilion and had a large inner group communications. They might be the watchers and helpers of these people (Figure 3.2).

Figure 3.2 Groups of 37 people with large inner-group communications, sharing stays with the three suspects in the pavilion in the morning

After 11:00, the three main crime people (1123214, 1350546 and 1502920) had more frequent communications with the other two groups of collaborators (Figure 3.3). We inferred they were checking the environements and sending requests and feedbacks for the crime. The three people left at 11:30 when the pavilion was opened again, and headed for the Keizmosaurus Big Spin. Another two people (1123214, 1187909) also set out for the same location at the same time. There was a notable subtle event that the two people chose different ways intentionally, though they had the same origin and headed to the same destination. It might be suspected as they were cautious. The five of them had a short meeting (for about 5 minutes), before the three suspects went back to the Pavilion at 11:46.

Figure 3.3 Three groups of people in the spatial view with the communication information

The crime was probably discovered at around 11:45, when external communications busted out and lasted until 12:00. There were also large communications among people in the pavilion during this period (Fig 3.4). Afterwards, there were large messages coming in and out from the two largest communication IDs. It might be the park administrator who told the tourists to be aware of the crime and had them evacuated from the pavilion(Fig 3.5).

Figure 3.4 Communications when the crime was discovered

The three groups stayed still and silent until 12:50, seemingly waiting for things to cool down. Then two of them (1187909 and 1000279) went to the pavilion to check the situation and quickly left (Figure 3.5). All groups were active again after 13:10, but did not have suspicious behaviors since then.

Figure 3.5 Two people checking things out after the event cooled down