Two Eyes One Pizza

From Visual Analytics for Business Intelligence
Revision as of 21:51, 11 October 2019 by Kelvin.chia.2017 (talk | contribs)
Jump to navigation Jump to search
2e1p2.png

Back to Project Home

HOME

 

PROPOSAL

 

POSTER

 

APPLICATION

 

RESEARCH PAPER


Problem and Motivation

To build an exploratory analysis tool that allows

Business profiling of the company’s outlet to determine point of interests that can generate insights such as: Highest earning outlets Relative performance of outlets Outlet’s profile patterns Item sales information

Create dynamic visualisations to make informed business decisions: Eg. Determining locations for new outlet opening with matching POIs of high sales outlets Clearly visualise the sales of different outlets and insights on what can be improved


Scalable program to incorporate future data to generate current information (Using data from other countries besides Taiwan)


Information about Data

This project DATA IS:

PLACE HOLDER

Background Survey of Related Works

Visualizations Explaination


Data source: http://bbvatourism.vizzuality.com/?nationality=US#tourism

This visualisation enables the viewer to find out the countries that the tourists are coming from. This, combined with the thickness of the line to represent the relative number of tourists compared to the other countries will enable us to have a quick and clear overview. Additionally, the team will look into customising this current chart, to allow for a more detailed view of each countries' actual proportion of tourists that are coming into Singapore compared to the other countries being visited.


Data source: https://www.stb.gov.sg/statistics-and-market-insights

This visualisation comprises of two charts in one - a time-series line chart, and a bar chart, which compares the international visitor arrival with the year on year change. Despite the limitations with this visualisation (seemingly confusing), the team feels that it is necessary to give a broad overview of the overall trend in relation to the total number of visitors. Additionally, it would be good to subsequently break this down to enable zooming into a certain country's monthly changes.


Data source: https://www.stb.gov.sg/statistics-and-market-insights/Documents/STB%20Q1%202016%20Tourism%20Sector%20Performance%20Report.pdf

The bar chart allow user to compare data between the countries. Next to each bar, we will indicate whether the number of visitors increases or decreases from the previous year, and how much different is the percentage. This chart is to let our user get a rough gauge of the number of visitors to Singapore.


Data source: https://www.stb.gov.sg/statistics-and-market-insights/Documents/STB%20Q1%202016%20Tourism%20Sector%20Performance%20Report.pdf

The used of stacked horizontal bar charts is an effective way of displaying the expenditure of visitors by country. The columns allow users to see the expenditure breakdown of visitors from the given country, as well as compare this breakdown against other visitor countries. This helps with understanding the distribution of expenditure by overseas visitors to Singapore, which will help with targeted marketing efforts such as market segmentation.

Design Inspirations

The following interactive data visualisations were used as inspiration for our application:

Visualization example What was learnt Desired objective to meet


World Map Visualisation

We learnt how to display the geographical locations through a world map, to showcase the physical distance that travelers travel to get to Singapore. This visualisation also shows the use of scale and colors, to convey the number of visitors coming to Singapore from a given country, compared to the other top 13 countries.

#1: Countries who are sending the most tourists to Singapore

#2: Proportion of tourists coming to Singapore against the total outbound residents of the country


Fisheye Zoom

We learnt how to make use of the fisheye zoom to allow for better viewing of clustered data. Many of the top 13 visiting countries are in the ASEAN region, and due to their close proximity to Singapore, many arrows and labels makes it difficult for a viewer to make active selections or comparisons. The fisheye zoom allows for finer mouse control by the user, without compromising on the overall view of the world map or distorting the sense of scope and scale.

#1: Countries who are sending the most tourists to Singapore

#2: Proportion of tourists coming to Singapore against the total outbound residents of the country


Radar Chart Visualisation

We learnt how to use a radar chart to demonstrate the breakdown of expenditure into different categories. This allowed for an insightful expenditure distribution to be displayed, along with being aesthetically pleasing. The ability to overlay multiple such breakdowns helps with comparison via area and skew easily. We also learnt that too many overlays will make the radar chart too cluttered and lose its prime value in its pure simplicity.

#3: Comparison of their expenditure and income patterns

#4: Breakdown of tourists’ expenditures to find out which industries they are spending most on

Proposed Storyboard

The following storyboard was designed:


#1: Title Screen

The title screen indicates the objectives that the data visualisation application seeks to accomplish on the topic of Visitors to Singapore. The use of a video background is meant to capture the different sights and places of interest in Singapore, playing to the theme of tourism. We liken our application user to that of a Singapore tourist, with the landing page akin to landing in Singapore.

The screens are assembled as part of a single-page website design, where each screen occupies the entirety of a screen display and page navigation is accomplished via scrolling or arrow keys.

#2: Visitor Arrivals

The Visitor Arrivals visualisation consists of a geomap, line graph and bar chart in order to visualise different aspects of Visitor Arrivals, due to the multifaceted nature of the data. These visualisations are all linked by year, with the dropdown menu dictating the data that is presented on all charts to present a unified dashboard view of Visitor Arrivals.

Geomap

The geomap visualises the travel pattern of visitors from Singapore’s Top 15 visitor countries between 2010–2016. The geomap allows the viewer to see the geographical locations of visiting countries on a scaled world map, depicting, the number of visitors, the distance they travelled to Singapore, and how geographically close these countries are to each other.

The following controls have been added to enhance the analysis of the geomap:

  • Dropdown menu to select year to display
  • Directed weighted arrows from visiting country to Singapore, with the weight being proportional to the absolute number of visitors visiting Singapore
  • Tooltip when hovering the mouse over a labelled country to provide
    • Number of visitor arrivals to Singapore
    • Total number of departures from selected country
    • Proportion of total departures to Singapore
  • Fisheye zoom on mouse cursor for better navigation and viewability

Bar Chart

The bar chart compares the number of Singapore visitors across the Top 13 visitor countries per year. These are ranked in descending order, showing the absolute number of visitors to Singapore per country. Following which, a toggle can be done to depict the absolute number of outbound visitors per country, allowing for a comparison by proportion of total outbound visitors arriving in Singapore.   The following controls have been added to enhance the analysis of the bar chart:

  • Bar chart
  • Dropdown menu to select year to display
  • Toggle to display “Arrived in Singapore” (default) or “Departed from Home Country”
  • Tooltip when hovering the mouse over a bar to provide
    • Number of visitor arrivals to Singapore
    • Total number of departures from selected country

Line Graph

The line graph depicts a time-series of Singapore visitors from the Top 13 visitor countries between 2010-2015. This allows for a quick yet meaningful comparison of the visitor arrival trend over time, along with a detailed drilldown which allows the user to see the individual country's visitor trends over time.

The following controls have been added to enhance the analysis of the line graph:

  • Toggle to display individual country or all countries (default)
  • Tooltip when hovering the mouse over each point of the line to provide
    • Year of visitor arrival
    • Number of visitor arrivals to Singapore

#3: Visitor Expenditure

The Visitor Expenditures visualisation consists of a radar chart and line graph, to compare the yearly expenditure by country and their detailed breakdown against the average. These visualisations are all linked by country and year, with the dropdown menu dictating the country and the pill buttons dictating the year to render data that is presented on all charts to present a unified dashboard view of Visitor Expenditures.

Radar Chart

The radar chart provides the breakdown of visitor expenditure per country, with a comparison against the average tourist expenditure. This allows for easy comparisons to be made, visualising any skews in the data clearly.

The following controls have been added to enhance the analysis of the radar chart:

  • Dropdown menu to select country to display
  • Toggle to select year to display
  • Highlighting of country or average expenditure on hovering mouse over a bounded area

Line Graph

The line graph provides the time series of visitor expenditure per country, with a comparison against the average tourist expenditure between the period of 2013-2015. This allows for easy comparisons to be made, visualising any fluctuations in the expenditure pattern clearly.

The following controls have been added to enhance the analysis of the line graph:

  • Toggle to display individual country, average or both (default)
  • Tooltip when hovering the mouse over each point of the line to provide
    • Year of visitor arrival
    • Number of visitor arrivals to Singapore

#4: Key findings and conclusion

The key findings and conclusion screen display the important insights gathered to meet the objectives of the Visitors to Singapore topic, along with a conclusion on what this dataviz application accomplishes and how might STB benefit from such findings. The use of the video background wraps up the theme of tourism, with a flying plane used to invoke a sense of completion and departure. We further the metaphor used for our application user, gaining much from the experience and leaving the application in a similar fashion as leaving Singapore: fulfilled.

Tools / Libraries

The following tools and libraries are used in the development of the web-based data visualisation application:

  • Microsoft Excel
  • D3.js
  • Chart.js
  • HighCharts.js
  • JQuery
  • Github
  • Tableau
  • Bower

Architectural Diagram

The architecture diagram depicts the design implementation of the web-based data visualisation application:

Datasets

We have chosen the following datasets to do data discovery:

Dataset Rationale
Singapore Tourism Board Master Plan
  • To understand STB's strategy for tourism - a key focus area of Singapore economy
CEIC Outgoing Tourist Data
  • Study the outgoing pattern of residents from a given country
  • Obtain the proportion of these outgoing residents as incoming tourists to Singapore
STB Tourist Expenditure Data
  • Study the spending pattern of Singapore tourists
  • Drilldown into different aspects

Technical Challenges

We encountered the following technical challenges throughout the course of the project. We have indicated our proposed solutions, and the outcomes of the solutions.

Key Technical Challenges Proposed Solution Outcome
Data Unavailability
  • Countries may disclose their inbound tourism data sets but not their outbound tourism data due to their political nature in data confidentiality, for example China. (Source: Contemporary Perspectives on China Tourism)

The use of the CEIC database was used for the most accurate information. For information that simply could not be sourced, or are from less credible sources, consultation with Professor Kam was arranged to seek his guidance on the matter.

Unfamiliarity in D3.js
  • Initial hands-on experience during D3.js workshop.
  • Independent learning on D3.js.
  • Peer learning and sharing

The independent learning and team sharing greatly aided the team with our competence and confidence in using D3.js and other programmatic visualisation tools. The D3.js workshop that was organised was excellent in ensuring that the right skills were learnt.

Data Cleaning & Transformation Proposed Solution
  • Having a systematic process while working together in order to maximise efficiency e.g. taking turns to clean, transform and perform checks on the data to ensure accuracy

The adopted process was having clear instructions issued to each member in the team, along with maintaining constant communication with each other. In the event that the dataset is deemed too dirty to be usable, it was dropped along with sourcing for new data that would be a suitable replacement.

Integrating Relevant Data from Multiple Sources Proposed Solution
  • Working together to decide on what data to extract or eliminate

The constant communication and discussion helped with deciding on which data should be integrated and displayed. Professor Kam's advice was also sought, with his guidance serving most helpful throughout the integration process.

Determining the Most Effective Ways in Visualising the Data
  • Proposed solution: Gain exposure to various forms of data visualisations - revisit course materials, assess existing libraries to gain inspirations

The design inspirations phase and background research greatly aided the team in understanding the most effective ways in visualising the data. Understanding the type of data that we worked with, along with adopting the best design practices helped with creating meaningful visualisations. Professor Kam was also consulted to provide feedback on ideal visualisations for the given dataset.

Roles & Milestones

  • Project Timeline
  • Gantt Chart

References

Ideation Drafts

In the process of completing this research project, an iterative approach was adopted. This comprised of various ideation drafts, where whiteboarding and draft prototyping was conducted.

Comments

Feel free to leave comments / suggestions!