1 CSE 6242 Fall 15 Capstone Project Team Advisor Matt Garvey Dr. - - PowerPoint PPT Presentation

▶
1
SMART_READER_LITE
LIVE PREVIEW

1 CSE 6242 Fall 15 Capstone Project Team Advisor Matt Garvey Dr. - - PowerPoint PPT Presentation

Feb 28, 2023 152 likes •496 views

1 CSE 6242 Fall 15 Capstone Project Team Advisor Matt Garvey Dr. Polo Chau Nilaksh Das Jiaxing Su Bhanu Verma Meghna Natraj 2 PROBLEM Atlanta is one of the most crime-ridden cities in U.S.A. Pedestrians are highly

slide-1
SLIDE 1

1

slide-2
SLIDE 2

Team

Matt Garvey Nilaksh Das Jiaxing Su Bhanu Verma Meghna Natraj

Advisor

  • Dr. Polo Chau

2

CSE 6242 Fall ‘15 Capstone Project

slide-3
SLIDE 3

○ Atlanta is one of the most crime-ridden cities in U.S.A. ○ Pedestrians are highly susceptible to crime, especially at night.

PROBLEM

3

slide-4
SLIDE 4

4

slide-5
SLIDE 5

OBJECTIVES

○ Enhance walking safety by providing routes with less crime risk ○ Provide risk-distance trade-off path choices to users ○ Enable safety alert to friends when user is in distress

5

slide-6
SLIDE 6

ANALYTICS BUILDING BLOCKS

6

Collection Cleaning Integration Analysis Visualization Presentation Dissemination

slide-7
SLIDE 7

7

Collection Cleaning Integration Analysis Visualization Presentation

slide-8
SLIDE 8

○ Atlanta Police Department website ○ 2009 → 2015 ○ ~ 250k crimes ○ All crime data in CSV format

CRIME DATA

8

slide-9
SLIDE 9

Legend

Class Count (2009 - 2015)

LARCENY-FROM VEHICLE 64345 LARCENY-NON VEHICLE 55902 BURGLARY-RESIDENCE 38277 AUTO THEFT 33256 AGG ASSAULT 16388 ROBBERY-PEDESTRIAN 12483 BURGLARY-NONRES 7243 ROBBERY-RESIDENCE 1632 ROBBERY-COMMERCIAL 1575 RAPE 789 HOMICIDE 592

> 20,000 > 5,000 AND < 20,000 < 5,000

9

CLASSES OF CRIMES

slide-10
SLIDE 10

MAP DATA

○ OpenStreetMap of Atlanta ○ Downloaded using Mapzen metro extracts

10

slide-11
SLIDE 11

11

Collection Cleaning Integration Analysis Visualization Presentation

slide-12
SLIDE 12

Data is usually messy!

12

slide-13
SLIDE 13

13

Collection Cleaning Integration Analysis Visualization Presentation

slide-14
SLIDE 14

Integration of 2 datasets

14

City Crime Data - available by coordinates and time of day City Map Data - in OpenStreetMap format

slide-15
SLIDE 15

MAP DATA

○ Converted to a graph using osm4routing ○ Graph consists of nodes on every road segment in the city ○ Nodes on the same road segment are successively connected by edges ○ Nodes: 111,380 ○ Edges: 141,656

15

slide-16
SLIDE 16

MAP DATA - EDGE LENGTH

Walkable Distance ○ Skewed left with a mean of ~215m ○ Majority of edges being under 150m ○ Maximum 400m-500m

16

slide-17
SLIDE 17

RISK OF EDGES

17 Map Node Crime Node

slide-18
SLIDE 18

RISK OF EDGES

18 Map Node Crime Node

slide-19
SLIDE 19

RISK OF EDGES

○ Assign risk values to nodes based on crime density

19 Map Node Crime Node

slide-20
SLIDE 20

RISK OF EDGES

○ Assign risk values to nodes based on crime density ○ Assign risk values to edges based on node values

20 Map Node Crime Node

slide-21
SLIDE 21

RISK OF EDGES

○ Assign risk values to nodes based on crime density ○ Assign risk values to edges based on node values ○ Each edge has a both a distance and risk value

21 Map Node Crime Node

slide-22
SLIDE 22

22

Collection Cleaning Integration Analysis Visualization Presentation

slide-23
SLIDE 23

OPTIMAL PATHS

Pulse algorithm

○ shortest distance, more risk → least risk, more distance ○ pruning algorithm ○

  • utputs all dominant paths

23

slide-24
SLIDE 24

TRADEOFF ANALYSIS

  • Left Plot:

○ Ratio of Least-Risk-Path’s distance to the Shortest-Distance-Path’s distance ○ mean: 1.13

  • Right Plot:

○ Ratio of Shortest-Distance-Path’s risk to the Least-Risk-Path’s risk ○ mean: 1.58

  • Takeaway

○ Going from SDP to LRP produces a larger proportional decrease in risk than the proportional increase in distance

24 d(LRP) / d(SDP) r(SDP) / r(LRP)

slide-25
SLIDE 25

RUNTIME ANALYSIS

400 recorded runtime instances Statistics (seconds)

mean 1.22 SD 0.51 max 6.8 (not shown) min 1.15

25

slide-26
SLIDE 26

TECHNOLOGY

26

  • MongoDB
  • Apache Spark
  • Python 2.7
  • Node.js
  • Phonegap - HTML/JS

(Storing graph data, geospatial indexing) (Preprocessing) (Preprocessing / Back-end) (Back-end) (Front-end)

slide-27
SLIDE 27

27

Collection Cleaning Integration Analysis Visualization Presentation

slide-28
SLIDE 28

28

slide-29
SLIDE 29

29

Collection Cleaning Integration Analysis Visualization Presentation

slide-30
SLIDE 30

DEMO

slide-31
SLIDE 31

31

slide-32
SLIDE 32

32

slide-33
SLIDE 33

Team Passage: Matt Garvey Nilaksh Das Jiaxing Su Meghna Natraj Bhanu Verma

PASSAGE

Advisor:

  • Dr. Polo Chau