Student Team: Yes
Approximately how many hours were spent
working on this submission in total?
Around 150
Hours
May we post your submission in the
Visual Analytics Benchmark Repository after VAST Challenge 2014 is complete?
Yes
Video:
https://www.youtube.com/watch?v=ELRsqLfHtgQ&feature=youtu.be
Link to the tool:
http://embers.cs.vt.edu:60051/
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Questions
Common Answer: Concept of “Points of
Interest (POIs)” |
A Point of Interest (POI) is defined as any location where a user
stops for more than 5 minutes. This location has a diameter of 50 meters. For
example: A shopping complex. Several
users will park their car for more than 5 minutes and will park within
vicinity (< 50 meters) of each other. Because no POI information was given
for any of the recreational points, we used this definition to identify all
such POIs. In total 129 POIs were identified. Out of these 64 POIs resulted
from faulty GPS associated with cars 9 and 28. After removing these POIs, we
were left with 64 POIs. These POIs were further classified as Home POIs
(location where a user spends most of his nights), Work POIs (location where
the user is present during office hours) and Recreational POIs (POIs which
are frequented during lunch and dinner hours). |
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.
Explanation |
Supporting
Evidence |
We analyzed POI Frequency of work, home and recreational POIs
over time for weekdays and weekends to find an answer to this question. POI
frequency over time represents the number of cars present at that particular
POI during a particular time accumulated for all the days chosen. For
example, left chart in figure 1 shows that cumulatively 181 cars were present
at POI 2 (GAStech building) at 8:45 am from Jan 13th
till Jan 17th. Figure 1 shows POI 2 (GAStech
building) frequency over time for both weekdays and weekends. As we can see
people come there during work hours and go out during lunch breaks and
weekends Figure 2 shows POI frequency distribution for Home POIs during
weekdays and weekends. Figure 3 shows POI frequency distribution of recreational POIs
during weekdays and weekends. Figure 4 shows POI Distribution for all the POIs across time for
weekdays. POI distribution shows that for the selected days, which cars were
present at which POIs during a particular time of the day. As we can see on
weekdays during office hours all the employees come to POI 2 and at other
times they either go to their home POIs or at other recreational POIs. Figure 5 shows POI Distribution over Time for all the cars for
weekends. Based on the information provided by these charts, a day in the
life of a typical GAStech employee can be
summarized in the table mentioned below: |
Figure 1: Work
POI’s (POI_ID 2) Frequency Distribution on Weekdays and Weekends Figure 2: Home
POIs Frequency Distribution on Weekdays and Weekends Figure 3:
Recreational POIs Frequency Distribution on Weekdays and Weekends Figure 4: POI
Distribution on Weekdays Figure 5: POI
Distribution on Weekends |
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.
S.
No |
Explanation |
Supporting
Image |
1 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Loreto Bodrogi (Car
15) -
Isia Vann (Car 16) -
Hennie Osvaldo (Car 21) -
Minke Mies (Car 24) c.
What
locations are involved? -
N Branda St 4131
3801 (POI 16) -
N Nia Ave / N Blant
St (POI 53) -
N Bairn St / N Utmana St (POI 65) -
N Utmana St / N Bilar St (POI 68) d.
When
does the pattern or event take place? -
POI 16 -
POI 53 -
POI 65 -
POI 68 e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
6: POI Distribution for each of the cars show their presence at home POIs of
Executives |
2 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Inga Ferro (Car 13) -
Loreto Bodrogi (Car
15) -
Hennie Osvaldo (Car 21) -
Minke Mies (Car 24) c.
What
locations are involved? -
N Tackan Ave / N Acera St (POI 59) -
N Gerantoni St / N Aveny St (POI 60) -
N Agentes St / N Maskin St (POI 61) -
S Evripidou Ave / S Eleftherias St (POI 62) -
N Camino St
5600 5652 (POI 63) d.
When
does the pattern or event take place? -
POI 59 -
POI 60 -
POI 61 -
POI 62 -
POI 63 e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
7: POI Distribution for each of the cars show a cluster near POI 59-63 and
around 11:30 – 12:30 |
3 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Hennie Osvaldo (Car 21) -
POI 57 is also the home POI for Inga Ferro,
Loreto Bodrogi, and Isia
Vann -
POI 64 is also the home POI for Lidelse Dedos and Birgitta Frante c.
What
locations are involved? -
N Edessis St / N Karanteg St (POI 57) -
N Karanteg St / N Agentes St (POI 64) d.
When
does the pattern or event take place? -
At POI 57 on Weekdays except Wednesdays -
At POI 64 on Weekends and Wednesdays e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
8: Hennie Osvaldo’s POI Distribution over different days. As can
be seen on some days he sleeps at POI 57 and on others at POI 64 |
4 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Hennie Osvaldo (Car 21) -
Lidelse Dados (Car
14) -
Birgitta Frente (Car 18) c.
What
locations are involved? -
N Karanteg St / N Agentes St (POI 64) d.
When
does the pattern or event take place? -
Every weekday between 17:45 and 19:30 e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
9: As can be seen, Cars 14, 18 and 21 are all present at POI 64 consistently
every weekday around 18:00 |
5 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Loreto Bodrogi (Car
15) -
Kanon Herrero (Car 22) -
Adra Nubarron (Car 25) -
Edvard Vann (Car 34) c.
What
locations are involved? -
Kronos Capitol / Abila Park (POI 66) d.
When
does the pattern or event take place? -
January 18th in the afternoon
around 13:30 e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
10: Map shows that Cars 15, 22, 25 and 34 were present at POI 66 on January
18th |
6 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Lucas Alcazar (Car 1) c.
What
locations are involved? -
GAStech Office d.
When
does the pattern or event take place? -
January 6th 22:30 – January 7th
01:15 -
January 8th 21:45 – 23:45 -
January 15th 22:45 – January 16th
00:15 -
January 17th 20:45 – 22:45 e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
11: Lucas Alcazar (Car 1) POIs Distribution over Time across Days |
7 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Kanon Herrero (Car 22) -
Lucas Alcazar (Car 1) -
Linnea Bergen (Car 6) -
Felix Resumir (Car
30) c.
What
locations are involved? -
Abila Zacharo -
Frydos Autosupply n more -
Guy’s Gyro -
Hippokampos -
Kalami Kafenion -
U-pump d.
When
does the pattern or event take place? -
Abila Zacharo -
Frydos Autosupply n more -
Guy’s Gyro -
Hippokampos Car 22 was present at GAStech
on Jan 17th, 13:38 -
Kalami Kafenion -
U-pump e.
Why
is this pattern or event significant? f.
What
is your level of confidence about this pattern or event? Why? |
Figure
12: Employees location at the time of credit card swipe |
8 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Lucas Alcazar (Car 1) 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? |
Figure
13: Lucas Alcazar’s Spending Analyzed in different ways |
9 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Axel Calzas (Car 9) -
Elsa Orilla (Car 28) 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? |
Figure
14: POI distribution of Cars 9 and 28 after hiding Home, Work and
Recreational POIs |
10 |
a.
What
is the pattern or event you observe? b.
Who
is involved? -
Lars Azada (Car 2) –
Home Owner -
Lucas Alcazar, Felix Balas,
Isak Baza, Linnea Bergen,
Isande Borrasca, Nils Calixto, Axel Calzas (doubtful,
car 9), Gustav Cazar, Lidelse
Dedos, Birgitta Frente, Vira Frente, Adra Nubarron, Marin Onda, and Brand
Tempestad 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? |
Figure
15: Map display of all the cars present at POI 8 on January 10th |
11 |
a.
What
is the pattern or event you observe? b.
Who
is involved? c.
What
locations are involved? -
Bean there done that -
Brewed Awakening -
Jack Magical Beans 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? |
Figure
16: Employees’ location at the time of credit card swipe |
12 |
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? g.
What
is your level of confidence about this pattern or event?_Why? |
Figure
17: Employee’s location at the time of credit card swipe at Kronos Mart |
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.
S. No |
Uncertainties and Conflicts |
Supporting Image |
1 |
Missing POI Data -
No “points of
interests’ were provided in the geo-spatial data which makes it difficult to
identify suspicious locations. A suspicious location is one, which is not a
recreational or home or office location. -
Hence there was
a need for the identification of Recreational POIs. -
Credit Card
Swipe times, POI Frequency over time, POI distribution over time and Visual
map of Abila were used to identify POIs |
Figure 18:
Identification of Recreational POIs |
2 |
Missing Truck Assignments to Truck Drivers -
No information
about which truck was assigned to which driver has been provided, thereby
making analysis of truck drivers difficult -
The trucks were
analyzed by mapping their POIs on the map and correlating them with Abila
tourist map and the credit card expenditure of the truck drivers -
Nothing
suspicious was found other than the fact that people from facilities spend
considerably larger as compared to other employees. Looking at their spending
trend, we can see that it is primarily for business transactions and that’s
why such high amounts. |
Figure 19:
Analysis of Truck POIs |
3 |
Corrupt GPS Data -
For cars 9 and 28,
the GPS appears to be broken, which led to the discovery of several false
POIs -
Once it was
determined that cars 9 and 28 have faulty GPS, all the POIs associated only
with these cars were discarded from further analysis. This led to the removal
of 64 POIs out of 129 total POIs. |
Figure 20: POI
Frequency data for discarded POIs resulting from Car 9 and 28 |
4 |
Data of Varying Resolution -
While credit
card data was available to the minute, the loyalty card information was
available only at day level. -
Several of the
transactions were missing either from credit card data or loyalty card data. -
Total
expenditure for a person was determined by finding the union set of
expenditure for each day for each establishment for each employee. |
Figure 21:
Analyzing Loyalty and Credit Card spending against Total spending |