Condition Effect Learning

Conditon Effect Learning

This page reflects my idea about the Condition Effect Learning system. If you have comments, please dont delete text but add comments so I can reflect. See this as a first draft, which can be used as a basis for the Cheetah condition effect learning system.

As said in Advisory System, Cheetah needs a system that learns about the effect of certain conditions. As you can read in Advisory System, I think there are three kinds of conditions:

1. Certain conditions;
2. Predicted uncertain conditions;
3. Yet unpredicted conditions.

There are some comlications regarding learning about food effects:

• Since each human and each body is different, conditions dont always have a fixed certain effect. Food for instance has GI (Glucose Index) that tells about the effect of the food on BG (Blood Glucose) levels. Responses vary between individuals and between days as much as 20%. So, the food effect can be expressed as an effect range. E.g. a minimum and maximum BG effect. This also counts for responses to insuline and the effect of activities (like sports).

Therefore, to account for individual and temporal differences, I think it is a good thing to generally express condition effect by a range instead of just a single number.

The effect of conditions types 2 and 3 should be induced by the learning system of Cheetah. Lets assume that at each blood glucose (BG) measurement, Cheetah starts the Condition Effect Learning (CEL) system. Then, the ΔG is calculated: the difference between predicted and measured BG levels: <math>\Delta G = G_{measured} - G_{predicted}</math> <math>\sqrt{1-e^2}</math>

Simply said, the system looks at ΔG: the difference between a predicted and measured blood glucose level). At the end of each day, Cheetah looks at the BG (measured blood)