if domain_model poorly classified text, we should retrain that model based on LLM response, so in future it could process similar texts better.
initial implementation plan:
-
if confidence of domain_model < threshold, store to db following:
text, domain_model confidence, domain_model result (grade), LLM confidence, LLM result
-
prepare a scheduled job, that will take all failed classifications from DB, and retrain domain_model based on it
if
domain_modelpoorly classified text, we should retrain that model based on LLM response, so in future it could process similar texts better.initial implementation plan:
if confidence of
domain_model< threshold, store to db following:text,
domain_modelconfidence,domain_modelresult (grade), LLM confidence, LLM resultprepare a scheduled job, that will take all failed classifications from DB, and retrain
domain_modelbased on it