How does the SAX works, I have the code for the same in java but how to decide the sliding window size for a time series. My time series is not continuous and there is lot of noise in the data.
I'm newbie using SAX but I was wondering if you could use an adaptive version of the PAA under which the SAX alphabet is performed. This'd provide shorter sliding window sizes when the variability of the series is higher and longer "steps" when the changes are steady.
I attached a couple of references about "variance-wise segmentation" and also about APCA, in addition to the main references about SAX already pointed out by George Appiah Sarfo. Hope this helps.