Internship experiences Puck Mulders

In my master Data science in engineering the opportunity is offered to do an internship. I chose to take this great opportunity to gain experience in analyzing datasets in practice. The data that Jacob has collected over the years lend themselves perfectly to this. I myself have no background in the agricultural sector and I noticed that I really enjoyed learning a lot about cultivation and potatoes.

The purpose of my internship was to explain on the basis of data which variables in practice cause a difference in yield between the plots. I modeled the growth per plot for the various components of the potato plant. I did this for the foliage weight, the tuber weight, but also for the nitrate content in the plant. I then clustered the profiles I estimated for each plot. You can clearly see that there are groups of plots where, for example, the weight development in tubers lags behind, but also that there are groups of plots that do a lot better. Then, based on those groups, I looked at which variables are likely to influence this: is this due to the type of manure, the amount of nitrogen in the soil, or the leg spacing and the size of the seed? This way you can create a model that can predict the yield.
  
During this internship I learned a lot about agriculture and how data can gain meaning in this world. I have been able to talk to potato experts about the growth, the thousands of factors that influence the yield. I am very happy that I got the opportunity to analyze his data with Jacob van den Borne.