Entry Name: TJU-Chen-MC2
VAST Challenge 2014Mini-Challenge 2
Team Members:
Hui Chen, Tianjin University, leading2006@aliyun.com PRIMARY
Wentao Wang, Tianjin University, 1311572393@qq.com
Jiawan zhang, Tianjin University, jwzhang@tju.edu.cn
Ruimin Gao, Tianjin University, grm005@163.com
Han Tao, Tianjin University, th8678@163.com
Student Team: YES
Analytic Tools Used:
Arcgis Desktop,link: http://www.esrichina-bj.cn/
Excel
D3.JS ,link:http:d3js.org
Openstreetmap, link:http://www.openstreetmap.org/
Approximately how many hours were spent working on this submission in total?
About 300 hours.
May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2014 is complete?
YES
Video:
video link for download (NOTICE:Please open our video using windows media player)
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Questions
MC2.1 Describe common daily routines for GAStech employees. What does a day in the life of a typical GAStech employee look like?Please limit your response to no more than five images and 300 words.
Fig1.1 relationship of common consumption records
Firstly, we use this analysis tool to choose a typical person, we can see that Axel Calzas shared lots of common consumption records with the same department of engineering, but he did not have common consumption records with employees of other departments.
Fig1.2 Daily rountines during two weeks from January 6th t0 January 19th of Axel Calzas’s
From this picture, we can find that Axel Calzas’ daily routines are carnoro street, barwyn street, parla street, spetson street and Taxiarchon Arenue, his daily active areas are near the places named ALBERS FINE CLOTHING,COFFEE SHACK And BEAN THERE DONE THAT. We can judge that most of the Engineers live in this scope according to the relationship of common consumption records.(Fig1.1)
Fig1.3 January 6th Axel Calzas’ routine
Fig1.4. Daily consumption record about engineering Axel Calzas
From Fig 1.4, we can see the daily consumption record about engineering staff Axel Calzas from 2014/1/6 to 2014/1/19, in the weekdays he just went to the cafe shops and restaurants in most cases while he went for shopping at weekends, and in detail we can also see that he tended to have coffee in Bean There Done That at 12.00 pm for every weekday, which is his typical daily behavior.
MC2.2 Identify up to twelve unusual events or patterns that you see in the data. If you identify more than twelve patterns during your analysis, focus your answer on the patterns you consider to be most important for further investigation to help find the missing staff members. For each pattern or event you identify, describe
a. What is the pattern or event you observe?
b. Who is involved?
c. What locations are involved?
d. When does the pattern or event take place?
e. Why is this pattern or event significant?
f. What is your level of confidence about this pattern or event? Why?
Please limit your answer to no more than twelve images and 1500 words.
Pattern 1:
In the afternoon 3 pm of January 19, CEO、CIO、COO、CFO and ESA arranged to play golf together, the level of this event is A, because five bosses only met together for once in the two weeks before the incident and after this meeting they were missing, this kind of collective behavior needs to be noticed.
Fig2.1.Five Bosses played golf together in the afternoon of January 19
Pattern 2:
Henk Mies is a truck driver in facilities department, he just consumed in Airport on every Monday, Tuesday, Wednesday and Thursday in the two weeks, but his consumption record exists two exceptions, first one is in the 12pm of January 6 he paid 1873.79 by credit card without loyalty card(he indeed has one loyalty card), the second one is that in January 16 his loyalty card was added 3032 but his loyalty card was not lent to other staffs in the GAStech through program check. According to the above analysis, I make it as the level of B.
Fig2.2.Truck driver Henk Mies’ consumption record, exceptional points are in 1/6 and 1/16
Pattern 3:
Fig2.3.IT helpdesk Lucas Alcazar’s consumption record, exceptional point is in 1/13
Lucas Alcazar works in IT department as helpdesk, he paid for 10000 at Frydos Autosupply in the 7pm of January 13 by credit card without using loyalty card, from the web content and icon of Frydos, it may be shop selling machine parts, Lucas cost so much money without discount as a IT staff and the time was relatively late, so I think the level of this event is A.
Pattern 4:
Fig2.4.CEO assistant Mat Bramar and COO assistant common consumption record
Mat Bramar and Linda Lagos are the assistants of CEO and COO respectively, during the two weeks before incident they frequently had meal together even at weekends, before incident they went to Frydos for machine parts many times, in January 18 they had meal together and in January 19 they had no common consumption. In like manner, Rachel Pantanal (CIO assistant) and Anda Ribera (CFO assistant) also frequently had meal together, and Ribera met Linda Lagos, Mat Bramar successively. Because of the frequent contact, continuous consumption at Frydos before the incident and their special identity, I make the importance of event as level B.
Pattern 5:
Fig2.5.Truck driver Claudio Hawelon’s consumption record, exceptional point is in 1/10
Claudio Hawelon is a truck driver in facilities department, his total consumption was all in January 10 and he consumed 4039 at airport in that day without loyalty card, as a truck driver he just worked for one day and cost much in the airport without discount, all of these decide the event level as B in my viewpoint.
Pattern 6:
Fig2.6.CIO, COO and Lucas Alcazar’s consumption record in Kronos Mart, exceptional points are in 1/12 and 1/19
Kronos Mart is a supermarket in Kronos, many staffs came here for shopping, but there existed some abnormal patterns, COO and CIO both went shopping in mart at 4 am respectively in January 12 and January 19, Lucas Alcazar also went shopping in January 19 before dawn like CIO, but his loyalty card added record the day before he consumed, through analysis maybe he lent his loyalty card to others outside GAStech or it was just conflict data. All the above phenomena compose a puzzling pattern, the abnormal consumption events should attract our attentions, and I make this event as level A.
Pattern 7:
Fig2.7.CIO, ESA, Vira Frent and Lars Azada’s consumption record at Hippokampos before missing
Answer: January 19th 19:43，Two engineering staff named Vira Frent and Lars Azada were involved in the last dinner with the boss of the executive agency at Hippokampos before missing. the bosses are Ada Campo-Corrente (CIO),Willem Vasco-Pais (ESA). We find that usually, they do not share the common consumption record. However, they met each other this time. We believe that something must happend. It can't be a coincidence, and I make this event as level A.
Pattern 8:
Fig2.8.Abila Scrapyard only Dylan Scozzese comes here
Answer: January 7th 14:16, January 9th 14:12, January 14th 14:13, January 16th 16:56: Dylan Scozzese went to Abila Scrapyard and spent large amount of money. We believe this phenomenon is not something normal. Because all the other GAStech employees didn't have the consumption record here before, and they had never spent such big amount of money ($9482.09) as Dylan Scozzese did, and I make this event as level A.
Pattern 9:
Fig2.9.Hennie Osvaldo’s relationship of common consumption record
Answer: Many employees of Security Department share the common consumption record with some secretaries of executives. And people share the same last name often have breakfast together. Hennie Osvaldo (security) share the common consumption record with Mat Bramar (secretary of CEO); Linda Lagos(secretary of COO) and Varja Lagos have the same last name, and they often have breakfast together; Loreto Bodrogi (security) shares common consumption record with lots of employees. We think this pattern is important. If a kidnapper wants to kidnap someone, he/she would try to get more information about the targets. So the kidnapper can collect information from secretaries of executives, or someone close to the executives, and I make this event as level B.
Pattern 10:
Fig2.10.Ruscella Mies Haber’s relationship of common consumption record
Answer: Ruscella Mies Haber (Engineering) did not share the common consumption record with his own department. However, he shared common consumption record with Kanon Herrero (security) and Varja Lagos(security) at Brew've Been Served at around 8:00 on weekday mornings. As an Engineering manager, it's suspicious not to share common consumption record with his own department, and get so close to the security employees, and I make this event as level B.
Pattern 11:
Fig2.11.Isande Borrasca’s relationship of common consumption record
Answer: Isande Borrasca (Engineering) is an ordinary employee, but this person shares lots of common consumption records with executives like Ada, Campo-Corrente (CIO) and Ingrid Barranco (CFO). During these two weeks, they usually went to Brewed Awakenings for lunch at 12:00 for 9 times in total. We believe this is not normal. As an engineer(who is even not a manager), Isande Borrasca seems to act abnormally to have so many common consumption records with the executives, and I make this event as level A.
Pattern 12:
Fig2.12.Orilla Elsa’s GPS rountine on January 9th and common consumption record
Answer: We find that the car No.28, which is assigned to Orilla Elsa (Engineering), has abnormal GPS information. The GPS trace information usually appeared in areas of Upump and Jacks magic beans. At the same time, Kanon Herrero (security) often borrowed Elsa's loyalty card and the scores of loyalty card rised, while the credit card did not have related records to show that Kanon spent the money. We believe this is abnormal and Elsa may break down the GPS on purpose, and I make this event as level A.
MC2.3 Like most datasets, the data you were provided is imperfect, with possible issues such as missing data, conflicting data, data of varying resolutions, outliers, or other kinds of confusing data. Considering MC2 data is primarily spatiotemporal, describe how you identified and addressed the uncertainties and conflicts inherent in this data to reach your conclusions in questions MC2.1 and MC2.2.?Please limit your response to no more than five images and 300 words.
GPS data
We find that some car id is beyond car id that company assigned to staff, such as car id 101. We extracted records whose car id is in the range of [1, 35]. We handle exception data alone, trying to find whether there is a potential pattern.
Missing data: In cc_data.csv, there are some garbled data like symbol ? in location field, and some fields are missing because of misplacement, these kind of data is 212 items mainly about Katerina Caf.
Conflicting data: cc_data.csv contains 1491 items record, while loyalty_data.csv contains 1393 items, 1066 items is normal (record matches in location, price, etc), the other conflict items are in following kinds:
Fig3.1 Visual design about conflict data
1. 229 items of records are both in credit and loyalty card, but price is conflict, we use circle to display price, and ring around circle display conflict type, 1066 abnormal items is in blue ring, and these 229 items is in yellow ring (see Fig3.1).
2. There are 196 items of record only in credit card, these items are in red ring, 156 items of them means credit card owner has loyalty card but didn’t use, when we match time simply, we can see 109 items (contain duplication in case of omission) exist questionable borrow possibility, and price totally meet is 15 items, which are all that Elsa Orilla lent to Kanon Herrero.
3. There are 98 items of record only in loyalty card, we just draw them as outlier (solid circle) to consumption line.
Confusing data: Contrary to common sense, sometime we can see loyalty record is a day earlier than credit card record. We just draw them as what they are but regard them as concerned patterns.