Numerical series density - page 12

 

Yes, and you never said how you propose to group the numbers.

 
Vyacheslav Kornev:
You know that without constants, nothing works.
A man, when he looks at the sky, still
Wouldn't mistake a cluster of 2-3 stars for a cluster. A 5 run is it? And at what distance. Roughly, at least, the brain will accept

It's all relative - you can set limits and restrictions with constants... Anyway, my algorithm works without constants.

 
OK at least this way, do you agree that we know how to find the centre of mass? What was that a couple of pages ago?
 
Vyacheslav Kornev:
OK at least this way, do you agree that we know how to find the centre of mass? What was that a couple of pages ago?

About that spreadsheet https://www.mql5.com/ru/forum/163871/page9#comment_4130691 ?

The centre - where on average it is fastest to get to from different points - yes.

 
My algorithm also works without constants.
You can get them.
 
If you're looking for clusters.
Then just start with the smallest delta you have. This will mark you the densest clusters.

Then take the next delta. It'll show you a larger cluster.
More specifically, a cluster.
And that's it.
 
You don't have to make things up.
 
Vyacheslav Kornev:
Yes, what is the problem. So we have a row of only 2 numbers.
You have set the density within 6
You can set a minimum. Do not take into account the numbers that have a minimum delta of less than 2 or 3.

That's where I don't understand you. By the way, I corrected there - also the number 65 is missing in the results according to my method.

I set the density within 6 - let's say we see the numbers in which this condition was met - how do these numbers group with each other? How many clusters would that be and with what density?

 
And that's what you're talking about.
Briefly
For number 10, write down all numbers with which 10 has a delta <=6. We get a series of digits.

Then for number 13 we do the same.
Then up to 15 and all the way to the end.


We have a series for each number.

Now we remove duplicate rows.

And we got clusters.
 
Vyacheslav Kornev:
And you have this.
Briefly
For number 10, write down all numbers with which 10 has a delta <=6. We get a series of numbers.
Then for number 13 we do the same.
Then up to 15 and all the way to the end.
We have a series for each number.
Now we remove duplicate rows.
And we got clusters

Here is a table - the numbers that meet the condition are highlighted by a unit - what do you propose to do with it?

Reason: