Difference between revisions of "IS428 2017-18 T1 Assign Cheryl Chiam"

From Visual Analytics for Business Intelligence
Jump to navigation Jump to search
Line 119: Line 119:
  
 
|-
 
|-
−
| Results || [[Image:ccc.png|400px|center]]
+
| Results || [[Image:ccwinddirection.png|400px|center]]
 
|}
 
|}
  

Revision as of 23:45, 8 October 2017

Dataset Analysis & Transformation Process

Activity #1 Clean and Prepare Meteorological Data for Interpolation
Steps
  1. Created Date and Time Column with the appropriate format (for interpolation later)
  2. Removed column D and E because we are not focusing on Elevation.
  3. Removed empty row 460.
  4. Filled in row 445 with 0s for empty fields.
  5. Change Date header to Date Time for column A.
  6. Converted file to MD.csv.

(We will be using MD.csv for interpolation.)

Created files:

  • MD.csv
Results
ClearnPrepareforInterpolate.png
Activity #2 Interpolate Meteorological Data
Steps

As the records are in three hour intervals, the hours in between are not present. To aid the investigation, I hold the assumption that the next 2 hours after each wind reading follows the same readings taken at the original hour. Eg. 00:00’s wind speed and wind direction will be the same for 01:00 and 02:00.

I carried out this interpolation by writing an Interpolation java program from scratch after editing the fields in MD.csv to a suitable Date/Time format and removing header.

Basically, it reads through each line in the csv and then separates each field by looking at the delimiter. It will focus on the Time field, carrying out an additional separation of hours and minutes. If the hours are any of the 3 hour intervals, it will add a that timing’s record into the newly created csv called MeteorologicalData(Interpolate).csv along with two more records for the next two hours with the copied data.

After csv is generated, I changed the Date Time format to follow Sensor Data’s format and removed columns Date and Time since it’s not needed anymore.

Created files:

  • Interpolate.java
  • Interpolate.class
  • MeteorologicalData(Interpolate).csv
Results
InterpolateMeteorData.png

Direct Results of the program shown in MeteorologicalData(Interpolate).csv:

Cc3.png

After editing the csv generated to follow Sensor Data’s format:

Cc4.png
Activity #3 Create Locations.xlsx
Steps

From the description document, we can record the factorys’ X,Y coordinates and combined with the sensors’ coordinates.

Created files:

  • Locations.xlsx
Results
Cc5.png
Activity #4 Inverse Map Colour
Steps

Used external tool to inverse map colour for clearer and more focused display if showing diagrams on it.

Created files:

  • InverseMap.png
Results
Cc6.png
Activity #5 Created new calculated field for Compass Direction Images
Steps

Used this formula: IF [Wind Direction] < 22.5 Then "N" ELSEIF [Wind Direction] <45 Then "NE" ELSEIF [Wind Direction] <67.5 Then "NE" ELSEIF [Wind Direction] <90 Then "E" ELSEIF [Wind Direction] <112.5 Then "E" ELSEIF [Wind Direction] <135 Then "SE" ELSEIF [Wind Direction] <157.5 Then "SE" ELSEIF [Wind Direction] <180 Then "S" ELSEIF [Wind Direction] <202.5 Then "S" ELSEIF [Wind Direction] <225 Then "SW" ELSEIF [Wind Direction] <247.5 Then "SW" ELSEIF [Wind Direction] <270 Then "W" ELSEIF [Wind Direction] <292.5 Then "W" ELSEIF [Wind Direction] <315 Then "NW" ELSEIF [Wind Direction] <337.5 Then "NW" ELSEIF [Wind Direction] <=360 Then "W" ELSEIF [Wind Direction] == NULL Then "N"

End


Results
Ccwinddirection.png

Files that we would be using for our visualization:

  1. MeteorologicalData(Interpolate) (created)
  2. Sensor Data.xlsx (original)
  3. Locations.xlsx (created)
  4. InverseMap.png (created)
Retrieved from ‘https://wiki.smu.edu.sg/1718t1is428g1/index.php?title=IS428_2017-18_T1_Assign_Cheryl_Chiam&oldid=3011’