1
$\begingroup$

I am doing some work about entity disambiguation.

Please suppose, there are some candidates for each entity, e.g. e1 has three candidates c1, c2, c3.

Each candidate has two values: v1 and v2

We know that higher values are better but I do not know about their weights. I am looking for an approach to find the best candidates (there exist some training data, the suitable candidate of the entity is available).

Could you please advise me which one should be used: classification, regression or otherwise?

$\endgroup$

1 Answer 1

0
$\begingroup$

Are both the candidates and values to be estimated? Or is it only the candidates.

If it's only the candidates, then try using a classification model. You can then use the predicted class probabilities as weights on the candidate values to arrive at a final values for your entities.

$\endgroup$
1
  • $\begingroup$ yes, the both ones can be estimated $\endgroup$
    – dinan
    Apr 21, 2021 at 13:08

Not the answer you're looking for? Browse other questions tagged or ask your own question.