This is the repository for the AvivaQuantum Data Science hackathon organised with the Cambridge University Data Science Society (CUDSS). You will find below some useful hints on how to get started. We describe the challenge in the presentation "Cambridge Data Science Hack.pdf".
The data itself is available in the ".json" format and split into training data and hold out (test) data. The notebook in "Cambridge Data Science Hack Notebook.ipynb" provides some hints on how to load data, etc. If you prefer not to use python but rather R or another language of your choice, this is fine as well. We recommend having a look at the description in "Cambridge Data Science Hack Notebook.pdf " then, so that you get a feeling for the data.
Please send your predictions before 4.00pm to Jonty (jonty.haberfield@aviva.com).
Subject line in the email should be – ‘Aviva Challenge: ’ If multiple team members, then use surnames e.g. Hayes Dalglish Haberfield Only one submission should be made.
Here some hints on how to best go about the presentation for the final day:
The audience will be very mixed: from undergraduates to senior management staff from Aviva. Therefore aim to explain your work at a high level and don't assume any detailed knowledge of the algorithms or technologies used. Tell a story. Structure the presentation with an introduction, development and conclusion.
Try to ensure you address what you see as the key problem and your solution to it, putting yourself in the shoes of a data scientist working for Aviva. Explain your methods and solution, but make sure to do so in a way that is clear, simple, and as accessible as possible.
Slides should be very visual with powerful, relevant images. Ensure any text is clear and readable from a distance. If you need text prompts to remind you what to say, don't put this as text on the slides, but keep them in notes not visible to the audience.
Note that you should limit your presentation to 5 minutes.
Good luck! Aviva and the CUDSS