Entry Name: "AD-Law-MC1"
VAST Challenge 2017
Mini-Challenge 1
Team Members:
Po-Ming Law, Adobe Research, Georgia Tech, plaw@adobe.com PRIMARY
Sriram Karthik Badam, Adobe Research, University of Maryland, College Park, srbadam@adobe.com
Student Team: YES
Tools Used:
Python: for data processing
D3: for custom visualizations
Approximately how many hours were spent working on this submission in total?
50 hours
May we post your submission in the Visual Analytics Benchmark Repository after VAST Challenge 2017 is complete? NO
Video
Attached with the submission
If it does not work, here is another link: https://drive.google.com/file/d/0B-CF7LsSE_WGMGdxNnE4Uk5LeW8/view?usp=sharing
Questions
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 describing the kinds of vehicles participating, their spatial activities (where do they go?), their temporal activities (when does the pattern happen?), and provide a hypothesis of what the pattern represents (for example, if I drove to a coffee 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 words.
1. Park service vehicles patrolling the preserve
The vehicle shown in the MapVis starts from the ranger base, moves to gate 8. It moves to gate 8 again and finally ends at ranger base. From the funnel visualization, we can see that this daily pattern is very common. Out of 12780 paths which last for a day, 998 paths start with ranger base, right after they go through the ranger base, the same 998 paths go through gate 8, after that, all the paths go through the ranger base eventually.
2. Round-trip camping
This pattern is characterized by having the same entrance and exit (For example, in the figure, the path starts with entrance 4 and ends with entrance 4). They also go through a camping site and stay in the camping site for a long period of time. It indicates that these people go to the preserve for camping and they go home by following the original path. By scrolling through the paths of different categories of vehicles using MapVis, we discover that only 2 axle car (or motorcycle), 2 axle truck and 3 axle truck are involved in this pattern. The funnel tree shows that 31 paths start with entrance4, go through camping6 and end with entrance4. There are around 100 paths with the form entrance4->camping->entrance4. It should that it is indeed common for people to do round-trip camping.
3. Camping with different entrance and exit
This pattern is characterized by having the difference entrances and exits (For example, in the figure, the path starts with entrance 3 and ends with entrance 4). They also stay in a camping site for a long period of time. We believe that these people have some other places to go after camping (probably a party!). This might explain why they do not go back by following the same path but instead exiting the preserve from a different entrance.
4. Sightseeing
These paths enter and exit the preserve with the same entrance. Although they might enter a camping site, they only stay for a very short period of time. Another characteristic is that these paths cover a large area of the preserve. Altogether, we predict that these people go to the preserve for sightseeing and go home by following the original path.
5. Pass by the preserve
This pattern has very short event sequences. It is likely that they only pass by the preserve to go to some other destination. From the funnel tree visualization, we can observe that it is instead a common pattern. There are 416 paths which go directly from entrance 2 to entrance 4, 386 paths which go directly from entrance 3 to entrance 0 and 415 paths which go directly from entrance 4 to entrance 2.
6. Bus routes
There are two kind of buses in the preserve: 2 axle bus and 3 axle bus. We observe from the MapVis that these two categories of vehicles follow paths which do not have any camping sites. Another characteristic of the paths of these two categories of vehicles is that they last for less than two hours. Therefore, we suspect that these are the bus routes.
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 occur over multiple days (including across the entire data set) by vehicles traveling through and within the park. Characterize the patterns by describing the kinds of vehicles participating, their spatial activities (where do they go?), their temporal activities (when does the pattern happen?), and provide a hypothesis 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 occurring there each Saturday). Please limit your answer to six images and 500 words.
1. Round-trip camping
Same as question 1 pattern 2, these paths have the same entrance and exit. From the funnel tree, we can see that this pattern is indeed very common. Out of the 5928 paths which last for more than a day, 410 paths start with entrance 3 and end with entrance 3, 182 paths start with entrance 0 and end with entrance 0, and 220 paths start with entrance 4 and end with entrance 4.
2. Camping with different entrance and exit
Same as question 1 pattern 3, these paths have different entrance and exit. The funnel tree again tells us that this pattern is common: 263 paths start with entrance 3 and end with entrance 4, and 416 paths start with entrance 0 and end with entrance 3.
3. Multiple camping sites
These paths go to more than one camping site and stay in each camping site for a long period of time. For example, the path in the following figure stay in both camping 1 and camping 4 for more than an hour.
4. Long camping (>10 days)
These paths are camping tours (going through a camping site) that last for more than 10 days. From the MapVis, these tours start more commonly during Fridays, Saturdays, and Sundays compared to other days.
5. Short-distance camping
These paths are characterized by camping close to the entrance where the visitors come from. They also do not travel a lot around the preserve. The example in the figure travels to camping 8 which is very close to entrance 2. The visitor also does not travel around the preserve a lot which can be seen from the short event sequence (only 6 events).
6. Long-distance camping
These paths are characterized by camping far away from the entrance where the visitors come from. As a result, these visitors need to travel a lot around the preserve. For example, in the figure, although there are other camping sites which are closer to entrance 1 (e.g. camping 2, camping 3 and camping 4), the visitor chooses to travel to 8 which is very far away from entrance 1.
3 – Unusual patterns may be patterns of activity that changes from 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 multiple days) and highlight why you find them unusual. Please limit your answer to six images and 500 words.
1. Not ending with an entrance or ranger-base
While exploring with the funnel tree, we observe the pattern shown in the figure: among 2462 paths which start with entrance2, 147 paths go to camping6 right after going through entrance2 but only 146 paths go to camping6 again. None of the 147 paths do not go through camping6 after going through camping6. Our question was where does the 1 path goes? The missing path is shown in the following map visualization. As you can see, it does not end with an entrance or the ranger base. We analyzed the dataset further and found that 34 paths do not end with an entrance or the ranger base.
2. Not park service vehicles but going through the "Gates"
From the figure, we can see that the vehicle travels through some “Gates” but it is a 4-axle truck rather than a park service vehicle. It means that the vehicle travels through the “Gates” without permission. We found that there are 23 paths like this and all of them have the same event sequence: entrance3->gate6->ranger-stop6->gate5->general-gate5->gate3->ranger-stop3->ranger-stop3->gate3->general-gate5->gate5->ranger-stop6->gate6->entrance3.
3. Too fast between two gates
As there is a speed limit in the preserve, we suspect that some of the visitors might exceed the speed limit. We compute the time taken between any two pairs of gates and define abnormal as less than the 5th percentile or greater than the 95th percentile. In the figure, each line represents the distribution of time between two gates. The numbers on a line from left are right are the shortest time, the 5th percentile, the 95th percentile and the longest time (all in seconds). The dots corresponding to the 5th percentile and the 95th percentile in the line are red in color. Between general-gate6 and general-gate5, the shortest time is 85s which is far shorter than the 5th percentile (294s). We compute the straight-line distance between the two gates and find that the speed corresponding to 85s is around 75 mph which is threefold of the speed limit.
4. Too long between two events
The figure shows that for those event pairs in which both events are camping, the longest time is far greater than the 95th percentile indicating that some people may stay in the camping sites for too long. For example, the longest time between camping0 and camping0 is 2789958s (32 days!!). It is very weird that somebody stay in a camping site for such a long period of time.
5. Few distinct events but many repeats
Using our map visualization, we select the path with 281 events from the bar chart. We observe that although the path has 281 events but only 10 of them are distinct. It might be due to some problems in the tracking device, causing the same gates to be recorded multiple times.
6. Too many distinct events in a sequence
Different from pattern 5, this pattern has many distinct events. Many of them start from the ranger base, we suspect that it might cause harmful impact to the reserve due to its large activity area. We discover this pattern by selecting gate_count = 49 from the bar chart and observing these paths from the map.
4 –– What are the top 3 patterns you discovered that you suspect could be most impactful to bird life in the nature preserve? (Short text answer)
We concluded that there are three patterns which are most likely to cause harmful impact to the preserve:
1. Too long camping
As mentioned in question 3 (pattern 4), some people same to stay in a camping site for a long period of time. Human activities are clearly a burden to the bird's habitat.
2. Too fast between two gates
As mentioned in question 3 (pattern 3), vehicle speed in the preserve can be as high as 75 mph. When a vehicle travels too fast, it will inevitable cause excessive noise to the bird's habitat.
3. Too many distinct events in a sequence
As mentioned in question 3 (pattern 6), the vehicles involved in this pattern are the park service vehicles. They start from the ranger base and cover a large area of the preserve. We believe that the large coverage will cause a lot of disturbances to the preserve (e.g. noise pollution) and harm the bird's habitat.
