So, week three just passed and we’re moving forward. Slowly forward, but yes, forward. I actually thought we would be way ahead of where we’re standing at the moment and it’ kind of frustrating. As I wrote last time, I’m very aware of the fact that this is a part of the thesis, to figure out what to do and how to do it. But some days, when you’re full of energy, you just want run. And I’m like that most of the time which, I guess, makes this part of the thesis a very good experience for me.
I haven’t written about what we’re going to do yet. I thought I’d try to save it for later, when I would actually know what we’re doing. But I don’t know exactly when that time is going to be, so I will give a short description now.
As I wrote this summer, I was very determined to write my master thesis within the field of data analysis or data mining. Now we’re at a group at Scania working with operational data from vehicles. Our thesis will examine if we can find patterns in trouble codes and we will try to link these to reparation data. We will work with clustering methods and the dream senario would be if we use the data to help us predict reparations (sounds like a dream, yes). It doesn’t sound simple and I can tell you, it’s not.
At uni, when you study data mining, someone will give you nice and neat data, ready for analyzing. That is not the case IRL. I will tell you more about my experiences pre-processing data next week… Until then, take care!