A datathon is an event, usually organized over one or more days, where you are given a set amount of time to complete a data-based challenge. These challenges are different depending on the datathon, but the underlying goal is to have a fun, creative, and innovative event where you can show off and improve your data analysis skills.
Behavioral data is all about about consumers' online behaviors and habits, collected across their devices: personal computers, smartphones and tablets. Behavioral data includes viewed webpage URLs, search behavior, and which applications are used on smartphones and tablets. Behavioral data has been around for a while – but there is still a wealth of untapped potential ready to be used to help brands better understand their consumers.
You can register on ourEventbrite page, where you will be asked to provide information about yourself and any additional teammates.
Since you won't be seeing the data until the day of the datathon, we don't expect you to have an idea in mind in advance.
For the predictive track we will provide a training set and a test set. We don't provide additional tasks to be solved for the insights and visualization tracks. You are free to explore and analyze the data in your own way and answer your own questions to find meaningful insights.
The venue will be open as a work space all night, but if you would like to get some sleep, we recommend you find somewhere else to stay and come back the following morning.
No, we do not compensate travel expenses or accommodation costs.
The competition is open to everybody with an interest or background in data. Students, young professionals, and even experienced data scientists are welcome. However, we recommend that you have some working experience in analyzing data. For more information read the ‘REQUIREMENTS' section on the homepage.
If you already have teammates in mind, you can of course compete together – all you need to do is register as a team. However, we place a 4 person limit on the teams. Please note that if you have formed a team of less than four people, you do have the option of indicating whether or not you are open to adding new people to your team when you register.
When you register using the individual ticket, you will be asked whether or not you would like to form a team or join an existing one. We will have a networking session before the competition begins, during which you can find other teammates to work with. You do have the option to compete on your own – however, you might be at a disadvantage against teams of more people.
- The winner of the predictive track will be chosen based upon the performance and accuracy of their model in comparison to the test set.
- The insights track will be evaluated by the jury based on the actionability of the insights, their level of innovation as well as the potential business value of the results.
- The jury will evaluate the data visualization track based on the best design and content of the visualized information and insights from the data.
Yes. Not everybody within your team has to present, but at least one person has to show the judges your work in order to be eligible to win. At the end of the 24 hour work period, your team will have up to 7 minutes to present their findings.
By definition, everyone competes in all the tracks. You don't have to indicate to us which one you would like to participate in. However, the tracks are there to allow you to play to your strengths. For example, even if you are not the strongest in visualization or insights, you can still win the predictive track by creating the most accurate model.
The winners will receive their prizes atWHAT Conference on April 26 after the presentation of their findings to the audience.
When you register, you will be asked whether or not you are available/interested in an internship interview. Note: This information will not affect your chances of winning.
We will have 10-15 teams of 1-4 people each.
During the datathon, there will be supervisors on hand at the venue to help you with any questions about the day's events or to solve issues with the data, but not to help with analysis. We do not offer mentorship.
Yes. Since the data we provide is real data from real people, who have given us permission to collect their data, it's necessary to protect the data and only let people work with it who have signed a NDA. With your signature you ensure that you won't use the data for any other purpose, you won't share it and that you will delete it after the datathon.