GMMevents

So a while back I reported I was going a tad insane translating code from Matlab to R. I’ve since written a guide that I hope will help others with similar issues. The code I translated is now available as part of the ASNIPE package for R. ASNIPE stands for Animal Social Network Inference and Permutations for Ecologists. The code I translated is gmmevents, which runs a Gaussian mixture model on time series data. A gaussian mixture model is a form of unsupervised machine learning, sorting one dimensional data into clusters, similar to k-means clustering. To illustrate:

Adapted from Fig. 2 – Psorakis et al., J. R. Soc. Interface  (2012)

This obviously has a lot of applications in social network analysis, as we can infer that individuals present in the same event likely have some interaction. I hope people find it useful, and that it was worth me going slightly mad over!

Chickadees are hard to take photos of

image

See?

I think I’ve been in Canada almost two months now. It’s hard to say as the initial turmoil of moving to a new country slowly changes into everyday routine. This generally involves getting up, trying to do some science until some time in the evening and drinking far too much tea.

I’ve met my study species properly now. A few weekends ago we went for a walk at a place called Muddy Lake (currently frozen and definitely not muddy). When walking along the paths you will generally be followed by a cloud of chickadees, who live in hope that you will be one of the many people that feed them. They whiz around your head, dancing from branch to branch, waiting for the food to be provided. If you DO have food, they will quite cheerfully eat out of your hand.

chickadee

Despite their tameness taking a photo of them in a more natural setting is challenging, due to their dislike of sitting still for more than a second.

IMG_3046

As well as having met my study species properly, I’ve also been working to get to grips with both social network analysis and the study of personality in animals. I had a passing interest in both of these topics before, but now I’m having to rapidly learn about how these analyses are done.

IMG_3050

Chickadees are highly social and tend to move about in small flocks. We have information about which birds were with which other birds at feeders.This is generated by special feeders which can identify individual birds fitted with RFID readers.

Picture1An RFID feeder, photo by Teri Jones

We also have information about how certain birds reacted to personality tests as well as which birds are dominant and which are subordinate. There are quite a few interesting questions we could answer by bringing these datasets together. Wrangling the various files together is and working out how to analyse them is the main thing I’m currently dealing with. This involves spending a lot of time shuffling data about in R. I’m also trying to automate and improve some of the workflow for generating this type of data.

We’ve  had a few quite heavy snowfalls. This was my building the day after a particularly severe one:

12734285_10153563559802054_2903874915728631094_n

and this was the window to our office on the same day as the sun was steadily blotted out:

12718241_10153577600197054_8652879118166444072_n

A few weeks ago I gave a departmental seminar on my ENTIRE PHD. This involved talking for longer than I ever had before, which made it very easy to lose track of time. I was extremely afraid of speaking for too long and boring everyone to death.

Picture2So many slides!

I felt it went quite well. My main mistake was towards the end. I glanced at a clock (the WRONG clock) as it turned out and was convinced I’d gone overtime. I sped up for the last few slides (which is unfortunate as they are the most interesting) and then apologised at the end for speaking for so long. As it turned out, I’d looked at the wrong clock and came in under time. At least it left lots of time for the many questions, which hopefully indicate that people are interested in what I spent about four years of life doing.

I have been finding time to try and do some more CANADIAN things. For example, I’ve now tried Poutine which according to Wikipedia is Canada’s national food. It’s basically chips, cheese and gravy. I think that undersells it, it’s warm, tasty, cheesy, potatoey goo.

12742628_10153550946112054_1518076879411992022_n

I also ambled up to Winterlude (a winter festival thing held in Ottawa), where the international ice sculpture championship:IMG_0759OWL

IMG_0758almostashag?

This included a demonstration of ice carving, which is apparently mostly done with powertools now!

IMG_0771Swan?

I clearly need to find some more CANADIAN things to do. Someone mentioned something called Nanaimo bars..

 

More new articles

Just a heads up that two new articles added to the article section of the main site, both  summaries of recently released marine  science results:

skull

  • Hanging out with the Right dolphins

A study of how early socialising can have significant effects on the later life of bottlenose dolphins in Shark Bay, Australia.

Funnily enough the day after I wrote this (these articles sometimes take a while to get published on the main site) Celine Frere who has worked on the shark-bay dolphins for some time and recently started work at our university, gave a talk on “Incorporating social interactions into our understanding of evolution”. Which was rather a fun coincidence.

I have a few friends in the department who work on social networks. If this article interests you, I recommend checking out their blogs: Flock Mentality and The Irish Brent Goose Project for various goose based social network things.

  • Leopard seals look after their teeth

A description of the rather unique dual feeding strategy of the Leopard seals.

You can find these artciles under the article section the main website at fourth element.

I’m playing with the idea of directly linking to articles from these blog posts, but unfortunately this would just open the one article in a new tab without any of the surrounding website.How the article section on the site will be organised as more articles are added is also something to think about. If anyone has any feedback or suggestions into how any aspect of the site is run please e-mail me or comment and I’ll pass it on to the Fourth Element webmaster.

 Image courtesy of Wikicommons