This week, I am working to get the dynamic time warping functionality into my program. The process of doing so includes re-processing the features to include the time series, putting each series back together when we construct sequences, and then performing the DTW to generate a number that will be used to compute the kNN of each sequence which can then be used for predictions with the models. The processing time of these activities has gone up significantly since we have been using five different metrics with each of the F phase datasets. I am returning to school next week, and once I've completed the DTW processing all that will remain before we put together our second paper (The date for the reach journal we would like to submit it to is October 1), I am hoping I will have time to look again into the Agglomerative Hierarchical Clustering concept, which I did not successfully complete when we explored it earlier in the summer and then changed focus to the paper. We heard back
The bar charts I created showing the relationships between sleep minutes prior to and following a day and its sum of activity counts are much easier to understand than my previous scatter plots, they are still not entirely coherent.
In addition to the charts, I calculated the correlation coefficients between lights on activity and sleep minutes during lights out of each of the nights surrounding it. These numbers were quite interesting because in some cases, there was almost no relationship. In others, there was a relatively strong positive correlation, and in others still, there was a relatively strong negative one.
This week, Dr. Skornyakov replied to our questions about sedentary activity classification and how we might be able to connect sleep bouts. In addition to answering our questions, Dr. Skornyakov notified us that we may want to look at differences between the first 26 subjects and the rest, as there was a change in the actigraphs used between subjects 26 and 27. We are interested in looking at how this may be affecting our data and, if it is, how we can account for this difference in our future model.
On the subject of sleep bouts: Dr. Skorynakov has suggested that we may be able to merge sleep bouts that are disjoint by ten or fewer minutes of wake.
Going forward, I want to look more closely at the activity counts across subjects and the way that they are distributed. I am also going to re-calculate sleep bouts with the new 10-minute rule.
Comments
Post a Comment