Difference between revisions of "Report Group7"

From Visual Analytics and Applications
Jump to navigation Jump to search
Line 53: Line 53:
  
 
[[Image:Pic3_calendar.png |1000px|center|float]]
 
[[Image:Pic3_calendar.png |1000px|center|float]]
 +
<br/>
 +
This gave us the final table for the required for calendar.
 +
 +
[[Image:Pic1_calendar.png |1000px|center|float]]
 
<br/>
 
<br/>
 
''' Network Plot '''
 
''' Network Plot '''

Revision as of 23:15, 6 August 2017

ISSS608: Visual Analytics and Applications
- Anuthama Murugesan
- Krutika Balveer Choudhary
- Sumalika Kodumuru

Proposal

Poster

Application

Report

 

Geo-spatial analysis of vehicle movement data to uncover patterns and detect anomalies


Objective:

Data Pre-Processing:


Calendar Plot
Calendar heat maps can be used to understand the peak and non-peak cycles across months, day of the weeks and hour of the day. It gives us an overview of the dataset. We need the time-stamp to be separated in months, day of the week and hour of the day. The following code was used to extract hour, day of the week and month from the timestamp to make the calendar plot

float


We then aggregated the data by month and day of the week.

float


This gave us the final table for the required for calendar.

float


Network Plot
Network plot is constructed to examine the structure of relationship between gates in terms of how frequently people/users have taken the path. From this it is easy to identify which road is more popular. In order to construct this network plot, source and target gates need to be identified and the data should be prepared to plot path from source to target gate. However since the popularity of the path is an important insight that can be drawn from the network plot, frequency of the source and target needs to be extracted from the data.
The prepared data consists of two tables:
1. Node Table: Consists of all the available gates in the preserve mapped to a unique identifier and with their respective location details. X and Y co-ordinates represent the location of the gate in the preserve. A snippet of the actual data table is shown as follows:

float


2. Edge Table: Consists of Source and Target variables denoting Gate names but mapped to the node table with a common ID. Frequency and the time at which it occurred is noted for analysis. Here X1,X2, Y1 and Y2 represent the X-coordinate and Y-coordinate of Source and Target variable respectively. A snippet of the actual data table is shown as follows:

float



Methodology: