ISSS608 2017-18 T3 Assign Tan Yong Ying Data Overview and Cleaning

VAST Challenge 2018: Suspense at the Wildlife Preserve

Overview

Data Overview and Cleaning

Application Design

Insights

Conclusion

Data Overview

For this challenge, we were provided with the following data:

Dataset Description	Screenshot and Metadata
1. Calls and songs from known birds in the Boonsong Lekagul Wildlife Preserve. They are in MP3 format with varying lengths. The file name contains the species name and an integer that can be referenced with AllBirdsv4.csv for its metadata.	Screenshot of file names provided by Mistford College
2. “AllBirdsv4.csv” contains metadata for the calls and songs from known birds in eight variables.	Description of each variable in AllBirdsv4.csv Variable Name Description File ID The integer index of the file names for calls and songs from known birds English_name The common English name for the recorded bird Vocalization_type Type of bird sound, typically a “call” or a “song” Quality Quality of the recorded bird sound (A, B, C, D, E or no score). A recording with loud background noise or low bird sound purity has a lower quality. Time Time the recording was captured Date Date the recording was captured X X coordinate of the location the recording was captured at Y Y coordinate of the location the recording was captured at
3. Fifteen bird sound files provided by Kasios to support their claim that Rose-crested Blue Pipits are still plentiful across the Preserve. They were recorded over the past few months and are in MP3 format. The file names are integers between 1 and 15, and its metadata can be referenced with Test Bird Locations.csv.	Screenshot of file names provided by Kasios
4. “Test Bird Locations.csv” contains metadata for the Kasios files in three variables.	Description of each variable in Test Bird Locations.csv Variable Name Description ID The integer index of the file names provided by Kasios X X coordinate of the location the Kasios recording was captured at Y Y coordinate of the location the Kasios recording was captured at Screenshot of Test Birds Locations.csv
5. “Lekagul Roadways 2018.bmp” is a 200 by 200 pixel map of the Preserve which outlines the roadways through the site. The bottom left of the map has coordinate (0,0) and the top right of the map has coordinate (199,199). Note that the alleged dumping site was centered around coordinate (148, 159) and its extent has not been thoroughly studied.	Map of Lekagul Preserve with roadways

Data Cleaning

Out of the 5 pieces of data listed above, only AllBirdsv4.csv requires data cleaning to remove values that cannot be imputed or replaced manually through guessing or inference. The data cleaning outcome for each variable in AllBirdsv4.csv is as follows:

1. File ID

Banner image credit to: Marshal Hedin