Dear friends,


Android software update's official release will come around early next week. We thank all our beta user feedback which helped us identify an array of bugs / problems.


Quirk of machine learning and taming it


We know that quite a number of our backers have a research degree and we get very interesting discussions. There is one such discussion which could prove a great case to understand the quirk of FOCI’s machine learning and how to tame it.


The question is how would physical pain (neck pain, lower back pain or a headache etc) affect FOCI’s data?


-I just woke up with a neck pain this morning and have perfect data sample.




Physical pain can cause chronic stress, so there is a greater chance of seeing more stress, fatigue streaks orbs.


If machine learning is incomplete at this time, machine learning may learn too many physiological characteristics of breathing under chronic stress, and this reduces the sensitivity to physiological characteristics of stress, fatigue.


If the machine learning is performed during the period of physical pain, it is recommended that the machine learning be re-learned after the body recovers or (better) you can use ‘interactive machine learning’ to tell Machine learning that distraction is more like stress. As otherwise machine learning would not never have known that you have been in chronic stress because of physical pain.




That is to say, if you have been working with chronic stress, i.e. fast-paced financial service/startup/exam prep, etc, you can use similar principle, use 'interactive machine learning' and tell it that distraction is more like stress.


Best wishes,

Mick and the FOCI team