0
$\begingroup$

I am using a random forest classifier for binary classification with 977 records and class proportion of 77:23.

I am using Lime explainer to explain the predictions made by the model.

However, I see that my Lime exp score is only 20-40 for 80pc of my observations.

But I like the idea that Lime discretizes continuous variable into bins for model explanations. Ex: Age is divided into bins 3 bins. <30, >30 and <=78 and >78.

So, positive and negative classes have different bins. Meaning, positive class has only two bins (bin 1 and 2) and negative class has only bin 3.

So, instead of relying on LIME feature coefficients (which may not be reliable due to poor explanation score), I plan to just compute the number of times under each class, a specific bin appears and use that to plot bars. So, I take the advantage of lime discretization (for each class) but use my method to show importance of a feature.

But, Do you think poor lime exp score indicates poorly computed bins for continuous variable? Can I rely on lime computed bins of continuous variable?

$\endgroup$
5
  • 1
    $\begingroup$ What question are you trying to answer? What problem are you trying to solve, and how does your proposed scheme for bins using Lime relate to that problem? Please edit to clarify. $\endgroup$
    – Sycorax
    Commented Mar 29, 2022 at 15:55
  • $\begingroup$ I dont propose bins. Lime computes bins. I am trying to solve binary classification using random forest. All of this is mentioned in post $\endgroup$
    – The Great
    Commented Mar 29, 2022 at 15:58
  • $\begingroup$ I want to know whether I can rely on lime discretization of continuous variables if the lime explanarion score is poor $\endgroup$
    – The Great
    Commented Mar 29, 2022 at 16:00
  • 1
    $\begingroup$ Lime doesn't do classification, the random forest does. "Do you think is this the right approach?" Right approach to do what? "Does lime score matter for continuous variable discretization?" Matter in what sense? What bad thing are you worried about? $\endgroup$
    – Sycorax
    Commented Mar 29, 2022 at 16:00
  • $\begingroup$ edited and updated. I thought lime has a local model and it does classification. Are you sure it doesn't do classification? $\endgroup$
    – The Great
    Commented Mar 29, 2022 at 16:03

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.