COMP6214 Open Data Innovation Coursework 1
Instructions
This coursework has two parts, the first requires you to clean the provided dataset, and the
second requires you to create one visualisation using D3.
You must use the provided dataset for both cleaning and generating your visualisation. The
dataset should not be considered "real" data. The dataset has been heavily modified in order to
evaluate your ability to clean, manipulate and visualise such data. The data is provided in CSV
and can be found via the course website.
Part 1: Clean the dataset
You will be required to clean the dataset and then perform. simple manipulations to prepare it for
creating your visualisation. You will be assessed on your ability to identify and handle a number
of different types of errors in the dataset. These errors should be accounted for these either
through pre-processing (using tools such a Open Refine). You must provide the names of the
tools you used, which errors you found, and how you cleaned the error. There are 5 marks
available for this suggesting you should look for 5 or more different types errors. It is not
necessary to find all the errors.
Part 2: Create and host your visualisation
Using this dataset you must build a website to show your visualisation of the data. You should
aim to develop multidimensional (greater than 2 dimensions) visualisations that enable rich
exploration of the data. The visualisation should be appropriate to the dataset and appropriate
for the target audience or use case of your choosing. The web page must include a description
of the visualisation. This description should detail:
- What the visualisation shows
- How the visualisation is interactive
- Who the intended audience is
You must use the D3 visualisation library for this task, and your code must dynamically generate
the visualisation not just present a pre-rendered image. Your visualisation should have suitable
interactivity that allows for manipulation, filtering, and detailed analysis of the visualisation.
You must also host a working copy of your visualisation somewhere online (e.g. your ECS web
space) and ensure it is accessible between the handin and feedback date outlined at the top of
this document. It is your responsibility to keep this visualisation accessible.
In the submission you will include all code for the visualisation, so that another running
installation could be easily established. Please do not include the D3 library. The submission
will also include a text file, which should be kept concise (it is not a report) and should not be
longer than 500 words. The accompanying text file should contain:
- A short description of how to host a copy of your visualisation.
- An overview of the audience and use case for each visualisation and why your
visualisation is appropriate both to this audience and the data. (or a pointer to this data
on your website)
- A description of the interaction each visualisation provides and why this interaction is
appropriate both to the audience and the data. (or a pointer to this data on your website)
- Any details of any other operation you feel exceptional, such as data enrichment.
Submission
You must submit a zip archive containing:
1) Your cleaned csv file (part 1)
2) A text file, detailing tool the tools, which errors you found and how you cleaned them.
(part 1)
3) Source code for your D3 visualisation, including any accompanying CSS or JavaScript.
files (part 2)
4) A text file, containing the URL and instructions to run your code. (part 2)
Your zip file will be submitted electronically via handin.ecs.soton.ac.uk. We recommend you
host your solution on your ECS web space (as no marks are lost for ECS being offline).
Relevant Learning Outcomes
1) Identify innovation opportunities for open data.
2) Be able to apply appropriate validation, cleaning and transformation to use, reuse and
combine a multitude of complex datasets.
3) Critically evaluate a large range of Infographics and interaction techniques suitable for
different tasks.
Marking Scheme
Criterion Description Outcomes Total
Visualisation
Choice
Each visualisation was appropriate to the data
and audience. Each visualisation was well
presented showing careful design. The
visualisation presented multi-dimensional data.
1,2,3 5
Implementation Good use was made of an appropriate library
for presenting dynamically loading
visualisations. The implementation is
considered elegant and maintainable, is robust
and performs smoothly and is without errors
2, 3 4
Interactivity The visualisation provides interaction that is
appropriate for the audience. The interactivity
allows powerful filtering, selection and analysis
of the data.
2, 3 4
Cleaning and
Manipulation
The student has identified a number of
different types of errors in the dataset. The
student has applied suitable techniques to fix
errors and manipulate the dataset ready to be
visualised.
2 5
Completion The student has shown exceptional ability to
provide complex, yet highly suitable
visualisations that have potential to be used
commercially.
1, 2, 3 2
Further Resources
• Getting Started with D3, Mike Dewar, O'Reilly Media, 2012
• Information Visualization: Perception for Design, Colin Ware, Morgan Kaufmann, 2004
• Visualising Data, Ben Fry, O'Reilly Media, 2007