# Is there a way to predict permutations that consist of sets?

SequencePredict is a function that predicts the next value from the arranged data.

But I want to use the data for the combination to predict the next combination(set).

For example,

1. There are numbers 1-9 in your pocket.

2. Choose three by a rule.

3. There is data listing the results of the five repeated trials.

data = {{1,2,3}, {2,3,4}, {3,5,4}, {5,6,4}, {6,7,8}}

In fact, I will use 800 sets of data.

Can we design an algorithm that predicts the three combinations we can expect from the next run?

Is there a way to predict permutations that consist of sets?

• Are the triplets independently generated by the rule or each triplet depends on the previous one? Oct 3, 2019 at 4:18
• Unknown It looks like an array of arbitrary sets, but the idea is, "Can you find a rule?" The data above is just a sample of my own to make the rules visible. Oct 3, 2019 at 9:07
• I feel that the question is unanswerable by SequencePredict if each of the permutations are independent, because you need a sequence, not an instance of a rule. But maybe you can clarify this in the question? (I'm not exactly sure what are you asking...) Oct 3, 2019 at 15:15
• More specifically, Five people play rock-paper-scissors game. Scissors are 1, rocks are 2, the other is 3. Currently there are 10 games played. data = {{1,2,2,2,1}, {3,3,2,1,1}, ..., {2,1,3,2,2}} In this case, can you predict how the next five rock paper scissors will be served? Oct 3, 2019 at 15:28
• Another example is a winning number consisting of five pieces of data, like a lottery ticket. If we have 10 of these data, can we predict the next one? Oct 3, 2019 at 15:36