Hey there! Sorry to jump in to the forum like this, but I have a question that I've been struggling with for many months, and was wondering if someone would be able to help. Basically, what I am looking for is a program or a simple tool that will help me to identify significant correlations among any set of data fields. I need it for a medical application. For example, if the patient recorded an increase in symptoms during the summer, we would love it if the program could recognise this increase and report these findings to the user. The program would not need to understand what the fields are, or what they mean, it would just need to pick up on statistically significant correlations between the data points. So, for example, if you entered in someone's symptoms on an hourly basis, including their diet, the system might be able to deduce that the person's symptoms always increase 3 hours after eating cheese (an observation that might be missed by the user's, since it is somewhat 'hidden' in the data) and therefore allow the user to interpret the evidence that the patient might be lactose intolerant.
I would also like to have an option to compare on or off values with range values. For example: the patient had stomach pain of level 8, and they did take their medication that day. Would it be easiest to simply list the medication value at 100% and the pain value at 80% and compare? Or is there a better way?
I hope this description has made sense, and that there's someone out there that can provide some advice as to where to look for a software or manual solution.