0
$\begingroup$

I have a knowledge base, where each row represent a class and columns represent features of classes:

|Feature 1|Feature 2| Feature 3|...|Feature n|Class|
----------------------------------------------------
|  1 or 0 |  1 or 0 |  1 or 0  |...| 1 or 0  | "A" |
|  1 or 0 |  1 or 0 |  1 or 0  |...| 1 or 0  | "B" |
|  1 or 0 |  1 or 0 |  1 or 0  |...| 1 or 0  | "C" |
|  1 or 0 |  1 or 0 |  1 or 0  |...| 1 or 0  | "D" |

And I have input where I know only part of features' values e.g:

|Feature 1|Feature 2| Feature 3|Feature 4|Feature 5|
----------------------------------------------------
|    0    |  NaN   |     1     |    1    |   NaN   |

How can I calculate probability of each class being the result? e.g:

"A": 0.2132
"B": 0.1654
"C": 0.5342
"D": 0.4234
$\endgroup$
3
  • $\begingroup$ Hi & welcome to CV! Your question doesn't provide enough information. How does your data look? $\endgroup$
    – cherub
    Mar 22 '19 at 18:30
  • $\begingroup$ @cherub The table in the question is pretty much how it looks. It's a table of "deseases x symptoms", where rows correspond to deseases and columns to symptoms. If desease has particular symptom - the value in corresponding column would be 1, otherwise - 0. I'm trying to understand how can I calculate probability of a desease being result with given info about symtoms. So that value of one of desease's symptoms is 1 and it's the same in input than probability would increase, otherwise - decrease. And I don't know how to properly calucalte such probability if I only know a part of symptoms. $\endgroup$
    – HadJower5
    Mar 23 '19 at 11:56
  • $\begingroup$ @cherub thank you for commenting by the way :) $\endgroup$
    – HadJower5
    Mar 23 '19 at 11:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.