Difference between revisions of "1718t1is428T3"

From Visual Analytics for Business Intelligence
Jump to navigation Jump to search
m
m
Line 191: Line 191:
  
 
<div style="background: #006768; padding: 15px; font-weight: bold; line-height: 0.3em; text-indent: 15px; font-size: 20px;margin-top:25px"><font color=#FFFFFF>Project Timeline</font></div>
 
<div style="background: #006768; padding: 15px; font-weight: bold; line-height: 0.3em; text-indent: 15px; font-size: 20px;margin-top:25px"><font color=#FFFFFF>Project Timeline</font></div>
[[File:TimelineVA.jng|center| 850px]]
+
 
 +
[[File:TimelineVA.jng|center|850px]]
 +
 
 +
 
 
<div style="border-left: #008080 solid 12px; padding: 0px 30px 0px 18px; font-family: Arial, Helvetica;"><hr />
 
<div style="border-left: #008080 solid 12px; padding: 0px 30px 0px 18px; font-family: Arial, Helvetica;"><hr />
 
{| class="wikitable"
 
{| class="wikitable"

Revision as of 18:16, 1 November 2017

Logo.jpg

Proposal

 

Project Presentation

 

Project Application

 

Research Paper

 

Poster


Project Introduction and Motivation

With the continually rising prices of car prices in Singapore and the increasing polarisation of incomes in the country, our group intends to discover what have been the changes in the market demand for cars. Going deeper into the market, we will also show how the tastes for car brands have waxed and waned over the years. Demand for cars like Datsun and Supra have disappeared off the roads, while Fords and Audis have taken their places. Are Singaporeans starting to prefer more American and German cars than before? From the visualisations, we hope our users will be able to clearly segment the car market in Singapore and target the right group of buyers for their businesses.

Target Audience

We aim to deliver a highly interactive web application that allows someone to easily navigate between a large amount of time-series data and geographical data.

Our intended audience are:

  1. Car dealers to make better forecast for car stock
  2. Prospective car buyers
  3. Potential Car Suppliers wanting to enter the Singapore vehicle market
Objectives

In this project, we will be focusing on the following:

  • To discover the trend of registered cars and the COE for different brands across time
  • To help consumers discover current and future COE trends for different cars categories (A & B)
  • Trends in Singapore car market by brand (geography, market segmentation)
Dataset
Dataset/Source Data Attribute Rationale of Usage

New Registration of Cars (Jan 2002 to Aug 2017)

https://insights-ceicdata-com.libproxy.smu.edu.sg/Untitled-insight/myseries

  • Car brands
  • Month and Year
  • Number of registered cars
This dataset will be used to understand the current trend of the car market in Singapore

COE Price Data

https://insights-ceicdata-com.libproxy.smu.edu.sg/Untitled-insight/myseries

  • Month
  • Bidding Round
  • Vehicles Class
  • Quota Amount
  • No. of Successful Bids
  • Premium Amount (S$)
This dataset will be used to track the car demand of the consumers and their willingness to buy a car in Singapore

Car Prices by Model

One Motoring By Car Make

  • Serial No.
  • Make
  • Model
  • Bidding Round
  • Total Basic Cost (S$)
  • Total Basic Cost (With COE) (S$)
  • Average Selling Price by Authorised Dealer (S$)
  • Average Selling Price by Authorised Dealer (With COE) (S$)
This dataset will be used to track the car demand of the consumers and their willingness to buy a car in Singapore
Background Survey of Related Works
Related Works What We Can Learn

An interactive treemap indicating importers' revenue of different countries

Pic1.jpg

Source: http://atlas.media.mit.edu/en/profile/hs92/8708/

  • Colour indicates different continents, eg. Asia, Europe, North America etc.
  • Size indicates proportion of sales amount in the market
  • Hovering over the area indicates more detailed information
  • Provides a high level view of our data and displaying the item details at the same time. It allows us to see patterns quickly when our eyes visually aggregate rectangles in the same group.

An analysis of the trend of coe in a bar chart representation

Pic2.jpg

Source: http://coe.sgcharts.com/

  • Use line graph to compare the changes of COE prices over the years.
  • Bar chart indicates the different COE prices over time

Sunburst Diagram to visualize hierarchies in no. of registered cars by brand

Pic3.PNG

Source: https://accessanalytic.com.au/the-top-3-new-charts-in-excel-2016/

  • Provides an overview of the breakdown of data in categories hierarchically.
  • Colour can be used to highlight hierarchical groupings or specific categories.
  • Shows car brand proportion accurately


Proposal of StoryBoard
Proposed Visualization Explanation

Sunburst Diagram

Sunburst-new1.jpg

This type of visualisation shows hierarchy through a series of rings, that are sliced for each category node. Each ring corresponds to a level in the hierarchy, with the central circle representing the root node and the hierarchy moving outwards from it. Colour can be used to highlight hierarchal groupings or specific categories. We believe that a sunburst diagram would be effective as we would be able to visualise the number of cars breaking down from region to country to car brand. We would be using chart since it is easy to implement and interactive. These are some variations of the sunburst that we are considering too.

Treemap with Geographic Heatmap

Treemap2.jpg
Treemap1.jpg

Treemaps are used to display hierarchical data. Treemaps are economical in that they can be used within a limited space and yet display a large number of items simultaneously. We chose to use a tree map, because it would give an overview of the amount spent on different car brands, car brands of similar regions would be grouped with same colour. The cars would be grouped by continent.

When the cursor hovers over the continent, the geographic heatmap will be displayed to show the country distribution for that continent.

Radial Stacked Barchart

Stacked.jpg

We may want to effectively compare the COE prices. As there would be many years and months of COE, a radial stackbar chart will allow the prices to be presented in a single view and not across the screen. This is a non traditional take of the bar chart.

Technical Challenges
Technical Challenges Mitigation Plan
Obtaining datasets
  1. Go through databases provided by the library
  2. Enlist help from Professor Kam for CIEC access to data
Acquiring Data, Data Cleaning
  1. Plan the cleaning process and work closely to clean and analyse all data sources
Using d3.js and Highchart to display the visualisations which we want
  1. Attend D3 Programming Workshop
  2. Designated time and roles on learning on D3 and HighCharts Libraries & Technical Tools
  3. Peer Learning
Unfamiliarity in Implementing Interactivity and Animation Tools/Techniques in Visualization App
  1. Develop a Storyboard
  2. Different specialisation on Interactivity/Animation Techniques
  3. Referred to existing data visualisations for implementation
Project Timeline



Tasks Duration Resource Status
1. Iteration 1
1.1 Brain Storm of Topics Week 6-7 All Completed
1.2 Decide and search for availability of data Week 8 All In Progress
1.3 Data Gathering/Cleaning and Dashboard Skeleton Week 9 All Not yet started
2. Iteration 2
2.1 Dashboard & Technical Learning (First Draft) Week 10 Sunburst: Kang Li, TreeMap: Sarah, Radial Bar Chart: Cheryl Not yet started
2.2 Dashboard Integrated (Second Draft) Week 11 All review one another's portion Not yet started
2.3 Finalize Dashboard, obtain feedback from users Week 12 All Not yet started
3. Iteration 3
3.1 Designing of Poster, Update Wiki Pages, User Guide & Research Paper Week 13 Poster/Wiki: Cheryl, User Guide: Sarah, Research Paper: Kang Li Not yet started
3.2 Completion of Research Paper, Final Amendments and Update Wiki Week 14 All Not yet started
Technologies and Tools

Our team has decided to focus on these few tools and libraries in order to showcase our product.

  • D3.js
  • HighChart.js
  • Tableau
  • Notepad++
  • Adobe PhotoShop
References
Comments

Please share with us your feedback! :)

Prof In class review: Project intro and motivation need review, get a full appreciation of data, data rich - information poor, need effort to consolidate. Technically this data is someone who collected through various sources to put it through LTA, consolidated info in LTA repo, one of this web data service provider, they actually subscribe to LTA, they distribute through their paid portal, unless you know a query search for vertical form. Need effort to organize it to do a proper analysis. Understand how the structure of data looks like, to create data visualization. The current data structure of what you download from CEIC, dont have hierarchical structure, so need to reorganize the data in hierarchy ways to have hierarchical structure. Organize data to be in line with visualization else cant realise it. Be more specific in the motivation. Data sets are generic cant see the real data, need provide screenshot to see how it looks like. Reviewing diff is good. Timeline need to be updated. Tech and tools need review.