Difference between revisions of "IS428 AY2019-20T1 Assign Chew Hui Ling Visualization"

From Visual Analytics for Business Intelligence
Jump to navigation Jump to search
(Created page with "<div style=background:#D3D3D3 border:#696969> 165px <b><font size = 6; color="#8B4513">   VAST Challenge 2019 MC 1: Crowdsourcing for Situational Aw...")
 
 
(6 intermediate revisions by the same user not shown)
Line 11: Line 11:
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
 
;
 
;
−
[[IS428_AY2019-20T1_Assign_Chew_Hui_Ling_Data_Preparation|<b><font size="4"><font color="#8B4513">Data Analysis and Preparation</font></font></b>]]
+
[[IS428_AY2019-20T1_Assign_Chew_Hui_Ling_Data_Preparation|<b><font size="2"><font color="#BC8F8F">Data Analysis and Preparation</font></font></b>]]
  
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
 
;
 
;
−
[[IS428_AY2019-20T1_Assign_Chew_Hui_Ling_Visualization|<b><font size="2"><font color="#BC8F8F">Visualization</font></font></b>]]
+
[[IS428_AY2019-20T1_Assign_Chew_Hui_Ling_Visualization|<b><font size="4"><font color="#8B4513">Visualization</font></font></b>]]
  
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
 
| style="font-family:Century Gothic; font-size:100%; solid #1B338F; background:#D3D3D3; text-align:center;" width="20%" |  
Line 29: Line 29:
 
<br/>
 
<br/>
  
−
<font size="5"><font color="#8B4513">'''Data Analysis'''</font></font>
+
<font size="5"><font color="#8B4513">'''Interactive Visualization'''</font></font>
−
 
+
{| class="wikitable"
−
Before the analysis began, the dataset given is analysed to identify its respective format and attributes. There was only 1 csv file provided in the assignment which meant that there was minimal amount of information that can be analysed. Hence, I will be taking the shp file from mini challenge 2 as well as creating one csv file called connection to do my transformation process for each dataset.  
+
|-
 +
! Features !! Reasons
 +
|-
 +
| [[File:CH10-2019.jpg|300px]]|| I used a range picker in most of my dashboard so that the user can zoom in on a certain range of date. The user can also still click on the date itself and select a day which makes it multi functional.
 +
|-
 +
| [[File:CH11-2019.jpg|300px]]|| In order to be more user-friendly, instead of having multiple charts that visualise the same variables such as time and value. I decided to create parameters such as variable type and activity type. This is to allow the user to choose what they want in just one chart.
 +
<br/>
 +
For example, if the user wants to look at count instead of average. The user simply must click on count within the activity type.
 +
<br/>
 +
<br/>
 +
Code is as follows:
 +
<br/>
 +
CASE [Activity Type]
 +
<br/>
 +
WHEN "Average" THEN Avg([Damage value])
 +
<br/>
 +
WHEN "Count" THEN COUNT([Damage value])
 +
<br/>
 +
END
 +
<br/>
 +
|-
 +
| [[File:CH12-2019.jpg|300px]]|| Just like before I decided to give more choices for the users where there would be a standard chart that is available, and the user can choose to put in the extra variable or not.  
 +
<br/>
 +
For example, if the user wants to add damage area for a more in-depth visualisation, the user can just click on damage area within choice and it will automatically be placed into the “row” field
 +
<br/>
 +
<br/>
 +
Code is as follows:
 +
<br/>
 +
CASE [Damage Area]
 +
<br/>
 +
WHEN 'Damage Area'
 +
<br/>
 +
THEN [Damage area]
 +
<br/>
 +
ELSE ''
 +
<br/>
 +
END
 +
|-
 +
| [[File:CH13-2019.jpg|300px]]|| If the user is more into visualisation, we have this feature that will plot the different variables chosen such as average damage value or count for shake intensity. It will then slowly plot to allow the user to view how is the reporting like during the 4 days.
 +
|-
 +
| [[File:CH345-2019.jpg|100px]]|| I have also created something for the tables so that it will allow the user to view the top 10 and bottom 10 counts immediately. This is to make it more user friendly because without it, they have to slowly find which are the highest and lowest values which makes it also very time consuming. Hence, with this it will change the color for the numbers that are top 10 and bottom 10. Special thing to note is that it is calculated based on table across then down so that it will calculate for the whole table instead of one column or row.
 +
<br/>
 +
Code is as follows:
 +
<br/>
 +
IF RANK_UNIQUE(sum([Number of Records]), 'desc') <= 10
 +
<br/>
 +
THEN "Highest Count"
 +
<br/>
 +
ELSEIF RANK_UNIQUE(sum([Number of Records]),'asc') <= 10
 +
<br/>
 +
THEN "Lowest Count"
 +
<br/>
 +
ELSE"Neither"
 +
<br/>
 +
END
 +
<br/>
 +
|}
  
 
<br/>
 
<br/>
−
<font size="3"><font color="#8B4513">'''Mini-challenge 1 Reports'''</font></font>
+
<font size="5"><font color="#8B4513">'''Dashboard Explanation'''</font></font>
−
<br/><br/>
 
−
[[File:C1-2019.jpg|500px]]
 
 
<br/>
 
<br/>
 +
The dashboard link is as follows: https://public.tableau.com/profile/chew.hui.ling#!/vizhome/Assignment_15710532491460/BackgroundPage
  
−
Within the report, it shows the time, damage taken by the different components, shake intensity and location. This provides minimal information for data visualization. Hence, in order to improve the current data that we have, I will be doing the following changes:
+
The following image shows the dashboard that I have created (missing background page). In each image, it shows a paragraph of what the image/figure/table is supposed to describe and the filters available.  
−
# For each component: power, medical buildings. It would be hard to see all the values together. This is because either you must put 5 graphs together or you will have to create a calculated field to add all the 5 values together. Instead of doing that, pivoting will help to generate 1 column for the component name and 1 column for the component value. And with filter it will allow the user to view all or the components separately.
 
−
# For each value within the component, there is no specific name attached. The user only knows that 10 is the highest and 0 is the lowest. But for user experience, it is much better to attach something to the value such as 0 = Weak, etc.
 
−
# Within the report, there is no specific name for each location and to solve the problem. I will be using the shp file to integrate the location number and name together.
 
  
 +
<font size="3"><font color="#8B4513">'''Within Earthquake Reports'''</font></font>
 +
<br/>
 +
[[File:CH14-2019.jpg|1200px]]
 +
<br/>
 +
[[File:CH15-2019.jpg|1200px]]
 +
<br/>
 +
[[File:CH16-2019.jpg|1200px]]
 +
<br/>
 +
[[File:CH17-2019.jpg|1200px]]
 +
<br/>
 +
[[File:CH18-2019.jpg|1200px]]
 
<br/>
 
<br/>
−
<font size="5"><font color="#8B4513">'''Data Preparation'''</font></font>
+
[[File:CH19-2019.jpg|1200px]]
 
<br/>
 
<br/>
−
Through the 3 points that was mentioned, I have created this flow within tableau prep.
 
−
 
−
[[File:C2-2019.jpg|500px]]
 
−
 
−
<font size="3"><font color="#8B4513">'''Idea 1'''</font></font>
 
 
<br/>
 
<br/>
−
For the first idea for pivoting the table. Using tableau prep, click on the “Plus” icon next to the mc1-reports document and click on “Pivot”. The following screen will be shown, drag the fields that needs to be pivoted into the “Pivoted Fields” column. It will then automatically pivot those field, and after pivoting rename the columns within the “Pivoted Fields” column as shown below.
+
<font size="3"><font color="#8B4513">'''Within Map Comparisons'''</font></font>
−
 
 
−
[[File:C3-2019.jpg|500px]]
 
−
 
 
−
<font size="3"><font color="#8B4513">'''Idea 2'''</font></font>
 
 
<br/>
 
<br/>
−
The second idea is to attach the value for both damage and intensity to a description. In order to do this, we will be using the connection csv that I have created previously as shown below to join each value. I will be using a left join because within the data I realised that for both columns there were null values which could not be attached to a description. This makes sense because for null values, it means that there is no damage or shake.
+
[[File:CH20-2019.jpg|1200px]]
−
 
 
−
[[File:CH4-2019.jpg|500px]]
 
−
 
 
−
Hence, in order to solve this problem, I have created a field to replace the damage description that was created previously in the left join. What I have done was to first check whether the value was null, if it was I will replace the damage description from null to none but if it is not null, I will continue using the damage description. This is done to both shake intensity and damage.
 
−
 
 
−
[[File:CH5-2019.jpg|500px]]
 
−
 
 
−
After idea 2, this will be the output for the following csv after you have generated it.
 
−
 
 
−
[[File:CH6-2019.jpg|800px]]
 
−
 
 
−
<font size="3"><font color="#8B4513">'''Idea 3'''</font></font>
 
 
<br/>
 
<br/>
−
For the third idea, instead of using tableau prep. I will be using tableau itself to join a shp file to the output that was created in the previous 2 ideas. I will inner join both location number and id together to get the location name and geometry which will be used to create the maps for visualisation.
+
[[File:CH21-2019.jpg|1200px]]
 
<br/>
 
<br/>
−
[[File:CH8-2019.jpg|500px]]
 

Latest revision as of 23:46, 14 October 2019

MC1-2019.jpg   VAST Challenge 2019 MC 1: Crowdsourcing for Situational Awareness

Background

Data Analysis and Preparation

Visualization

Insights

Conclusion

 


Interactive Visualization

Features Reasons
CH10-2019.jpg I used a range picker in most of my dashboard so that the user can zoom in on a certain range of date. The user can also still click on the date itself and select a day which makes it multi functional.
CH11-2019.jpg In order to be more user-friendly, instead of having multiple charts that visualise the same variables such as time and value. I decided to create parameters such as variable type and activity type. This is to allow the user to choose what they want in just one chart.


For example, if the user wants to look at count instead of average. The user simply must click on count within the activity type.

Code is as follows:
CASE [Activity Type]
WHEN "Average" THEN Avg([Damage value])
WHEN "Count" THEN COUNT([Damage value])
END

CH12-2019.jpg Just like before I decided to give more choices for the users where there would be a standard chart that is available, and the user can choose to put in the extra variable or not.


For example, if the user wants to add damage area for a more in-depth visualisation, the user can just click on damage area within choice and it will automatically be placed into the “row” field

Code is as follows:
CASE [Damage Area]
WHEN 'Damage Area'
THEN [Damage area]
ELSE
END

CH13-2019.jpg If the user is more into visualisation, we have this feature that will plot the different variables chosen such as average damage value or count for shake intensity. It will then slowly plot to allow the user to view how is the reporting like during the 4 days.
CH345-2019.jpg I have also created something for the tables so that it will allow the user to view the top 10 and bottom 10 counts immediately. This is to make it more user friendly because without it, they have to slowly find which are the highest and lowest values which makes it also very time consuming. Hence, with this it will change the color for the numbers that are top 10 and bottom 10. Special thing to note is that it is calculated based on table across then down so that it will calculate for the whole table instead of one column or row.


Code is as follows:
IF RANK_UNIQUE(sum([Number of Records]), 'desc') <= 10
THEN "Highest Count"
ELSEIF RANK_UNIQUE(sum([Number of Records]),'asc') <= 10
THEN "Lowest Count"
ELSE"Neither"
END


Dashboard Explanation
The dashboard link is as follows: https://public.tableau.com/profile/chew.hui.ling#!/vizhome/Assignment_15710532491460/BackgroundPage

The following image shows the dashboard that I have created (missing background page). In each image, it shows a paragraph of what the image/figure/table is supposed to describe and the filters available.

Within Earthquake Reports
CH14-2019.jpg
CH15-2019.jpg
CH16-2019.jpg
CH17-2019.jpg
CH18-2019.jpg
CH19-2019.jpg

Within Map Comparisons
CH20-2019.jpg
CH21-2019.jpg

Retrieved from ‘https://wiki.smu.edu.sg/1920t1is428g1/index.php?title=IS428_AY2019-20T1_Assign_Chew_Hui_Ling_Visualization&oldid=7015’