6 Replies Latest reply on Feb 1, 2018 9:12 PM by Paul Fowler

# Logic Puzzle - OPA Determination with probability

A probability problem comes up in policy that needs frequent solving:

Do we immediately send police to a residence of suspected activity based on probability the situation will escalate?  Will a person likely skip bail? How likely is a benefit to be applicable after only a preliminary screening?

The challenge is that sometimes probability is involved.  We can't immediately send police to every event all the time.  We must sometimes be selective and our policy needs to allow for policy probability in some parts of our determinations.

Puzzle #5

The local department of criminal justice has provided the following statistics in determining the person’s probability of missing court (these statistics are made-up).  The overall probably of missing a court appointment is 45%.  Of those who miss court, 40% are in the local community, 80% had an outstanding warrant, and 10% had a job.

Allowing for uncertainty and collecting 1) whether a person is a member of the community, 2) has an outstanding warrant, and/or 3) has as job, write an OPA policy that implements the following:

As any FYI, my collection screen looks like this:

Good Luck!  As always, I will mark helpful and one correct solution.

P.S. After solved by the community, I will provide my quick and generic OPA pattern using Excel and Entity logic to solve for an arbitrary number of base attributes so that each attribute modifies the final result.

• ###### 1. Re: Logic Puzzle - OPA Determination with probability

Any takers?

My other unsolved puzzle on this forum, I can keep unsolved for years...  This puzzle, however, introduces patterns that have wide value to policy automation.

I want to introduce probabilistic determinations, so I can introduce the next step of merging AI and OPA determination at a conceptual level.

AI can go through our data and give us probable insights.  Whether decisions are made directly by AI or whether AI interpretation / guidance is weighted within a policy is itself a determination.  Someone has to decide (usually in policy) what to do with AI provided probabilistic determinations.

If no takers on this puzzle by the end of January, I will post a solution and a pattern to solve these problems on my blog, and get to the next problem -> high-level merging of AI (applied on top of big data) with human policy in determinations.

• ###### 2. Re: Logic Puzzle - OPA Determination with probability

Anyone want a final go at this, before I post a solution at the end of next week?

• ###### 3. Re: Logic Puzzle - OPA Determination with probability

No time to implement it, but looking forward to seeing Bayesian inference in OPA (perhaps a hint to others on one way to implement a solution).

• ###### 4. Re: Logic Puzzle - OPA Determination with probability

Fair question!  Good idea.

Note for those interested.  This is one of the several mathematical approaches used in AI.  It is also a good thing to understand in many real-life scenarios.  I have been to many blogs where people say understanding this makes them stronger reasoners in life.

• ###### 5. Re: Logic Puzzle - OPA Determination with probability

So, I had to provide the solution to this puzzle.  Apologies. I may post an AI / Machine Learning blog later today.

Solution:

Using this, we can take known statistics and combine it with OPA to have a probabilistic determination.  I am sure this can be improved upon.

We can use OPA for everything from horse betting (based on horse attributes such as "first-time turfer") to security risk analysis.  We just need the general stats.  Those stats may be continually updated with big data, machine learning, etc.

• ###### 6. Re: Logic Puzzle - OPA Determination with probability

Closing the loop on this.

I added an AI blog entry here: Artificial Intelligence, Machine Learning, and Oracle Policy Automation