Can we add a feedback to the given output to improve the upcoming outputs


I am using gpt-3.5-turbo model
llm = ChatOpenAI(model_name=“gpt-3.5-turbo”, temperature=0)

do we have any option after getting the answer from the gpt-3.5-turbo model if we give a feedback that the answer is not relevant to the question then from next time it show not show that answer it should give a new answer.

actually my requirement is
reading data from the PDFs and storing it in the chroma vector db using OpenAIEmbeddings()

Next when we ask a question by using the data from the chroma vector db and gpt-3.5-turbo it is giving the answers

till here everything is fine

now would like to add some sort of feedback system to this
based on that feedback it show show the results

is it possible to implement this
Please let me know