Entry Name: "OvGU-Held-MC1"

VAST Challenge 2015
Mini-Challenge 1

Team Members:

Pascal Held, Otto von Guericke University Magdeburg, Germany, pascal.held@ovgu.de PRIMARY

Christian Braune, Otto von Guericke University Magdeburg, Germany, christian.braune@ovgu.de

Rudolf Kruse, Otto von Guericke University Magdeburg, Germany, rudolf.kruse@ovgu.de

Student Team: NO

Did you use data from both mini-challenges? YES

Analytic Tools Used:

Self-developed scripts to make analysis and visualization.

Python

Matplotlib

NumPy / SciPy

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

60h

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

Video:

OvGU-Held-MC1.wmv

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

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.

For the first eight groups we used frequent sequence mining to determine if there are some patterns in check-in behaviour that occur frequently. We used a minimum support of a sequence of 42 (sic!) and a minimum length of 5 check-ins. The first check-in here is always entering the park in the morning. We thought these parameters should be relatively restrictive already, yet they returned eight distinct patterns that occur each between 42 and 44 times. Each pattern belongs to an individual user and - to our surprise - spans their complete park visit. All of these groups arrived at the same time (+/- 1 second) and left at the same time (+/- 1 second).

Group 1 (determined from Freq_Path_Analysis):

[1003392 1060545 1072988 1126372 1165979 1214689 1240231 130479 1332781

1366161 1417237 1530629 1541776 1543533 15490 1563610 1574391 161071

1626469 1630045 1664117 1679714 1872654 1873144 187692 1917187 1950762

195664 2070394 2074636 275440 280057 368047 399712 429192 56627

588544 60390 669199 685081 773078 874468 874532 991984], sequence = "-1 25 18 -4 8 81 -6 -6 8 13 -6 8 21 5 2 1 19 81 6"

This group enjoys Thrill rides and spends quite some time on open areas, relaxing. They only visit the park on Sunday. They arrive and leave the park at almost exactly the same time. They spend a lot of time walking between the open area and Kiddie Land.

Group 2 (determined from Freq_Path_Analysis):

[1078782 1084844 1155375 1223245 1240651 1265098 1275178 1306931 1326004

1337155 1532450 1541863 1564557 1620988 163905 1702844 171580 1843796

1863151 1864772 1976732 1978182 2023167 2037599 2079054 254060 257199

311504 321688 382408 487801 518888 547746 549739 734884 798609

857114 86514 89827 91552 966510 99371 996825], sequence = "-1 10 26 14 14 15 2 14 6 13 26 11 31 9 25 3 5 7 8 29 31 21 7 7"

This group spends most of their time in Tundra Land, riding Rides for Everyone and some Thrill and Kiddie Rides. They do not visit any show and they do not spend time on open areas. They spend a lot of time walking between different rides from different areas. They visit the park on Sunday.

Group 3 (determined from Freq_Path_Analysis):

[1033080 1039546 1043632 1099456 119769 1200913 120395 123736 1357705

1376135 142484 1507270 1535709 1536913 1569944 1743710 1746062 1767476

1880273 22302 275309 361623 389062 390713 436 454659 498990

510486 519433 595416 610033 664290 664579 685723 691481 786543

816120 820353 904119 914685 952853 971885], sequence = "-2 6 5 2 3 5 2 16 64 6 14 15"

This group almost exclusively uses thrill rides. They wait quite some time in lines waiting, or walking between attractions. Except for the Auvilotus Express (3) they use every thrill ride twice. They visit the park on Sunday.

Group 4 (determined from Freq_Path_Analysis):

[1003429 1007782 1065976 1130591 1165756 120394 129113 1332925 1336047

1386224 1440279 1479205 1527025 1529254 1569172 1575235 1630646 171566

1749456 176233 1788092 1792869 1873203 1878171 1906645 1949874 1967677

1982027 2009792 2076332 2081811 337015 446534 468864 555404 578697

587738 646245 681305 781978 84567 860499 916289 968455], sequence = "-1 9 64 8 7 3 11 9 2 64 2 1 7 8 1 29 19 3 4 81 3 81 3 64"

This group really likes the Sabre Tooth Theatre (64) which they visit three times during their stay. The remaining time they spend is Thrill Rides from Coaster Alley, Wet Land and Tundra Land. They visit the park on Sunday.

Group 5 (determined from Freq_Path_Analysis):

[1090434 1102435 1128373 1159022 119892 1309498 1402103 1444951 1480467

1503578 15746 1606131 1615933 1618432 1666657 1686274 1705401 173216

1767050 1835254 1918476 1932021 1985387 2031085 2071376 316298 406680

419001 422675 528061 541576 549399 590342 607389 608616 636567

643897 71526 759250 847239 994072 99649], sequence = "-1 6 28 3 7 22 2 32 16 63 7 64 19 32 19 4 64"

This group spends most of its time in Kiddie Land and Wet Land, whre they enjoy Shows, Entertainment and Thrill Rides. They tend to switch between Thrill or Kiddie Rides and Some show during their stay. They visit the park on Saturday.

Group 6 (determined from Freq_Path_Analysis):

[1089925 1136565 1198950 1280903 144336 1446001 146927 1531238 1595626

1659314 1665479 169442 1703113 1769583 1794268 1818480 1829821 1908542

191700 1950642 1958051 2081357 240939 268572 343755 352310 426192

454864 463272 473009 484561 485069 495930 616186 640611 711089

755541 759761 786325 869640 896994 913806], sequence = "-1 22 21 81 11 31 63 8 6 21 24 32 81 2 64 6 4 29 25 2 21"

This group spends most of their time in Wet Land and Kiddie Land. They use a lot of Kiddie Rides and Thrill Rides. They visit the park on Saturday.

Group 7 (determined from Freq_Path_Analysis):

[1042280 1093721 1127694 1172390 125303 1262713 1284436 1304658 140461

1435805 1494697 1497676 1578523 1684033 173102 1739364 1745223 1773283

1836075 1837146 1860181 1901680 190992 1919714 1932220 203445 209832

218279 300315 32672 393354 401428 534277 538957 682629 750143

764255 804851 868659 912895 977669 98371 986021], sequence = "-2 10 32 4 17 81 63 30 10 1 32 2 17 7 3 7 15"

This group Spends a lot of time in Wet Land and Kiddie Land where they visit Shows, use Kiddie Rides and Thrill Rides. They visit the park on Friday.

Group 8 (determined from Freq_Path_Analysis):

[1071523 1104936 1223924 1307715 1322226 1336280 140935 1421817 14916

1555391 1556239 1574091 160988 1618002 1623196 1718705 1753321 1753347

1796573 1872559 194305 1983198 2005317 2026631 2059292 250958 277328

294250 407415 508780 551933 570247 58518 589448 675616 698636

705173 833313 835980 857616 974337 988355], sequence = "-2 6 14 2 1 8 31 6 5 -5 3 64 14 1 20 23 17 26"

This group visits the park on Sunday and uses mainly Thrill Rides and Rides For Everyone. Most of their checkins are on Coaster Alley and Tundra Land. They spend some time on the open area in the centre of the park.

Since the behaviour of checking into the park and leaving the park was quite strange we looked at a global enter-vs.-leave plot of all customers for all days. A pattern that can be seen throughout all days is that most people arrive early in the morning (between 8:00 a.m. and 10:00 a.m.) and leave the park either at 8 p.m., 9p.m. or 10p.m.

For leaving there is a bit of variation but almost no one leaves between 8:50 and 8:59 p.m. for example. It seems as if there were some buses leaving every full hour transferring people back to where they came from.

On Friday it also seems as if the park was closing on 8 p.m. Since no visitor is in the park after that anymore.

A very distinct group of people (magenta) is the group that arrives between 9 and 10 a.m. as no one of them leaves the park between 8 and 11 p.m. A small group stays until after 11 pm but most leave in the first quarter hour after 3, 4, or 5 p.m.

These pattern repeat equally on Sunday (with more visitors in the park) and similarly on Friday (less visitors, park closes at 8 p.m.).

Looking at the number of Checkins into the different lands of Dinofun World we can see four distinct groups of people. Most notably is a very small group of 18 people (here on Saturday), which does not check in into any Ride at all. Their only recorded check-ins are the entries into the park itself (for some of them twice). These people might be custodians that lead a group of kids through the park or they may be park employees that just enter and start to work.

From the distribution of checckins across all days we may also see, that there is one group of low-frequency check-inners and one group with a rather high check-in frequency. The low-frequency-check-in-group checks into rides less than once every two hours (['356903' '1149884' '521750' '953838' '1269018' '1283386' '655378' '1787551' '383013' '1476464' '1563594' '439584' '1748887' '753553' '1080969' '500084' '417205' '217719' '1540524' '1217381' '1340222'

'1658667' '847619' '1725365' '1307724' '47441' '1458915' '1081515' '1835861' '921888' '1711922' '1680161' '1600469' '122838' '1629516' '2095051' '415491' '373112' '334793' '626433' '159893' '2049974' '644885'

'903264' '1935406' '1473321' '1703818' '1781128' '430595' '970913' '1763672' '1601276' '1737703' '2090763' '2010501' '1781070']) While the high-frequency group (['1117855' '1143814' '1257239' '490718' '1763638' '1442611' '2048286' '248437' '1965716' '1954188' '1034569' '2019254' '624372' '1413244' '1280868' '396030' '1731552' '1513766' '1607208' '591150' '661923' '178304' '1467932' '1087978' '1489451' '1074829' '266590' '1548729' '353418' '1586378']) checks in more than 3.25 times per hour.

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.

From the times of entering and leaving the park, the most obvious anomaly is that the park is closing early on Friday night. Instead of being open until midnight all visitors have left the park at 8 pm.

The Auvilotus Express (3) ride usually has around 20 (Friday) to 40 (Saturday) checked-in visitors at the same time. This however increases significantly on Sunday (up to 600 checked-in visitors which in turn increases the average waiting time for the ride from 4 minutes to more than 50 minutes. This can be seen in the following figure, where the number of checked-in users (left, blue; moving average in red), the check-ins per hours (center column, 15 minute-resolution) and the average waiting time (right column) can be seen for each day.

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.

The Craighton pavillon is emptied regularly for preparing the next exhibition of Scott's memorabilia. For that every visitor has to leave the building. This happens regularly between 10 am and 11:30 am as well as 3 pm an 4:30 pm. Except on Sunday. Here some people stay in the building.

On the area in front of the Grinosaurus stage usually people only check-in if there is a show scheduled. When the show is over all people head back into the park. On Friday however it seems that one person is taking a nap on the field and staying right there for the second show of the day.

If we look at the total number of check-ins for each day and ride we might see some decrease for certain attractions on Sunday. This would definitely be an anomaly, since on Sunday the most people visit the park. Indeed attractions 32 and 63 show such a (significant) decrease. These two are the main attractions (show stage and exhibition hall). It seems all shows and exhibitions are cancelled after the crime has occured.

From the same figure we see, the shop 62 has almost no check-ins during any day of the weekend, but if, then the number of visitors is constant.

The total number of checkins also shows, that there is a very small decrease in the number of visitors for ride 12 on Sunday. The visitor statistic shows us, that for some time there are no check-ins into this ride and no guests on Sunday noon. Probably the ride was broken during that time.

For a similar reason ride 7 becomes interesting. The increase in the number of visitors is less high, than what we would expect to see. And indeed, although there is no real breakdown of the attraction, the number of checked-in guests drops by almost one hundred on Sunday, 2 pm.

Although the fluctuations in visitor intake do not deviate much from +/- 20 customers over time, the Firefall has some signiifacantly high change in visitors on Saturday afternoon.

Deviatian from the standard +/- 20 deviation is also the SabreTooth Theatre, which fluctuates twice as much, probably due to shows ending at fixed times and people entering right after that for the next show.