Student
Team: YES
· &nb=
sp;
Graph Based Anomaly Detection (GBAD) URL:
· &nb=
sp;
Excel
Approximately how many hours were spent working=
on
this submission in total?
90 hours
May we post your submission in the Visual Analy=
tics
Benchmark Repository after VAST Challenge 2017 is complete? YES
Video
http://users.csc.tntech.edu/~
Questions
MC1. Question 1 – =
“Patterns
of Life” analyses depend on recognizing repeating patterns of activities by
individuals or groups. Describe up to six daily patterns of life by vehicles
traveling through and within the park. Characterize the patterns by describ=
ing
the kinds of vehicles participating, their spatial activities (where do they
go?), their temporal activities (when does the pattern happen?), and provid=
e a
hypothesis of what the pattern represents (for example, if I drove to a cof=
fee
house every morning, but did not stay for long, you might hypothesize I’m
getting coffee “to-go”). Please limit your answer to six images and 500 wor=
ds.
GBAD Introduction:
The Graph Based Anomaly Detection
(GBAD) system discovers both normative and anomalous patterns. GBAD uses the
minimum description length (MDL) principle to identify the normative pattern
that minimizes the number of bits needed to describe the input graph after
being compressed by the pattern, and then identifies three possible changes=
to
a graph: modifications, insertions=
and
deletions. Figure 1 demonstrates e=
ach of
the different types of structural changes.
Figure 1: Example Graph Showing Different Typ=
es of
Anomalies
For more detailed information
about GBAD, the reader can refer to:
[1] Eberle, W. and Holder, L., 2007. Anomaly detection in
data represented as graphs. Intelligent Data Analysis, =
11(6),
pp.663-689.
=
Patterns
1, 2, 3:
Figures 2-4 show visual representations of the top three normative patterns,
where each represents the most travelled routes by vehicles in the preserve=
.
Figure 2: Normative Path
Figure 3: Normative Path
Figure 4: Normative Path
Pattern 4 and 5: Figure 5 represents another n=
ormative
pattern, indicating routes travelled by 2-axle vehicles whereas Figure 6
represent the most common routes travelled by 4 axle vehicles.
Figure 5: Normative Path for 2 Axle Car
Figure 6: Normative Path for 4-Axle Truck
Pattern 6: Figure 7 shows a visualization=
of
a substructure normative pattern that indicates a route travelled by vehicl=
es
that leave the preserve through entrance 2. From this substructure, we infer
that most vehicles leave the preserve through entrance 2.
Figure 7: Normative Pattern
MC1. Question 2 Patterns of Life analyses may also
depend on understanding what patterns appear over longer periods of time (in
this case, over multiple days). Describe up to six patterns of life that oc=
cur
over multiple days (including across the entire data set) by vehicles trave=
ling
through and within the park. Characterize the patterns by describing the ki=
nds
of vehicles participating, their spatial activities (where do they go?), th=
eir
temporal activities (when does the pattern happen?), and provide a hypothes=
is
of what the pattern represents (for example, many vehicles showing up at the
same location each Saturday at the same time may suggest some activity occu=
rring
there each Saturday). Please limit your answer to six images and 500 words.=
Pattern
1: Figure 8 shows a
visualization of one of the normative patterns, indicating a route travelle=
d by
vehicles on Sundays.
Figure 8: Normative Pattern-Sundays
Pattern 2: Figure 9 shows a visualization o= f a normative pattern that indicates a route travelled by vehicles on Fridays.<= o:p>
Figure 9: Normative Pattern-Friday
Pattern 3: Fi=
gure
10 shows a visualization of a normative patterns that indicates a route
travelled by vehicles on Saturdays.
Figure 10: Normative Pattern- Saturday
Pattern 4:
Figure 11 shows a visualization of a normative patterns that indicates a ro=
ute travelled
by vehicles during Morning.
Figure SEQ Figure \* ARABIC 11: Normative Pattern- Morning
Pattern 5: Figure 12 shows a visualization =
of a
normative pattern that indicates a route travelled by vehicles during
Afternoon.
Figure 12: Normative Pattern-Afternoon
Pattern 6: Figure 13 shows a visualization of a normative pattern that
indicates a route travelled by vehicles in the month of July.
=
Figure 13: Normative Pattern-July
MC1. Question 3 – Unusual patterns may be patterns of activity that changes fr=
om
an established pattern, or are just difficult to explain from what you know=
of
a situation. Describe up to six unusual patterns (either single day or mult=
iple
days) and highlight why you find them unusual. Please limit your answer to =
six
images and 500 words.
Pattern 1: Figure 14 shows a
visualization of the most common substructure. For this substructure, the v=
ehicle
entered the preserve on 2015-07-17 and stayed at campsite 5, leaving campsi=
te 5
on 2015-07-18 as shown in Table 1.
Table 1: Vehicle- Location, Date
|
Location |
Date |
|
Entrance4 |
2015-07-17 |
|
General-gate5 |
2015-07-17 |
|
General-gate2 |
2015-07-17 |
|
Ranger-stop0 |
2015-07-17 |
|
Ranger-stop2 |
2015-07-17 |
|
General-gate1 |
2015-07-17 |
|
General-gate4 |
2015-07-17 |
|
General-gate7 |
2015-07-17 |
|
Camping 5 |
2015-07-17 to
2015-07-18 |
|
General-gate7 |
2015-07-18 |
|
General-gate4 |
2015-07-18 |
Figure 14: Vehicular Stay at Camping 5
Pattern 2: Figure 15 shows a visualization of an anomalous or infreq=
uent
substructure detected by GBAD-P. From this substructure, we infer that out =
of all
the vehicles only one vehicle left the preserve through General-gate4
(anomalous path).
Figure 15: Anomalous Path
Pattern 3: Figure 16 shows a visualization of a normative pattern th= at indicates a route travelled by vehicles on 2015-07-17.
Figure SEQ Figure \* ARABIC 16: Normative Path on 2015-07-17
Pattern 4: Figure=
17
shows a visualization of an anomalous or infrequent substructure detected by
GBAD-P. In this example, passing through General-gate0 is an anomalous
extension path.
Figure 17: Anomalous Path
MC1. Question 4 –– =
What
are the top 3 patterns you discovered that you suspect could be most impact=
ful
to bird life in the nature preserve? (Short text answer)
The
following is the summary of our observations:
·&nb=
sp;
Most vehicles
enter the preserve from entrance 3.
·&nb=
sp;
Most vehicles
leave the preserve through entrance 2. <=
/span>
·&nb=
sp;
Most common ty=
pe
of vehicle: 2 axle car (or motorcycle)
·&nb=
sp;
Most vehicles =
pass
through general-gate7.
·&nb=
sp;
Most vehicle
traffic in the preserve occurs on July 11th 2015
·&nb=
sp;
Vehicular
movements are higher during the months of July and next August.
·&nb=
sp;
Most vehicles =
stop
at ranger-stop0 in the preserve.
·&nb=
sp;
Most visitors =
stay
at camping-8.
·&nb=
sp;
More vehicular
movements are on Sunday.
·&nb=
sp;
The morning is=
the
busiest time of the day.
Based
on the above observations, we hypothesize the following as reasons for the
decrease in the number of nesting pairs of the Rose-Crested Blue Pipit:
Ø
The type of
vehicles traveling most are 2-axle cars. They cause more pollution compared=
to
other types of vehicles.
Ø
During the mon=
ths
of July and August, vehicular traffic is at its peak - the prime breeding
months for these birds.
Ø
Vehicle traffi=
c is
at its peak in the mornings, which due to atmospheric conditions, is when t=
he
birds sing. However, due to peak traffic in the mornings, atmospheric
conditions are changed, causing possible migration elsewhere.