How do you apply data science to your projects?
I am the data scientist of the ARISE project, monitoring demonstration sites (MDS) team. The MDS is primarily concerned with demonstrating the automated monitoring of Dutch biodiversity and developing knowledge and practical experience of the deployment of digital sensors for use in ecological research and biodiversity monitoring. As such, they run a few different demonstration sites using several different sensor types (wildlife cameras, audio loggers, insect cameras, etc.). These produce an awful lot of data, which it is my job to deal with.
Is there a project from this past year that you are most proud of?
Since I started this position, a lot of my time has been spent building and developing a system to ingest and organise the different data coming in from all these different sensors in a number of different ways. Some devices transmit automatically. Others still require the collecting of SD cards from the field. We needed to keep track of all this data and push it into cold storage on tapes, which is the most cost effective way of keeping such a large dataset. We also wanted to be able to monitor all this data as it comes in, keeping an eye on the performance of sensors and what sort of data they are collecting. Finally, we also wanted these monitoring tools to be available to stakeholders, meaning we have to be able to control access to the sensors and data.
All in all this took a fair amount of back end and front end development, during which I feel I really learnt a lot. And the system seems to be working, which is nice!
What do you like most about being a DSC member?
I very much enjoy seeing how data science is applied in very different fields from my own, such as the humanities. We have had some great workshops too!
What is your favourite data science method?
Not sure if it is a method per se, but I am very much a fan of building user friendly tools. Even for things that seem like relatively simple data management operations, removing the need to write a single line of code will vastly increase the likelihood of people using them!
Are you camp Python/R/or something else?
Very much both. Occasionally running python code from within R even. I am most likely to fire up R if I need to do some analysis or make a plot, but then tend to use Python for most other things.