[Retracted] MKA: A Scalable Medical Knowledge-Assisted Mechanism for Generative Models on Medical Conversation Tasks

Algorithm 2

Pseudocode of the topic detector.

Algorithm 2: the detection of the question topic

Input:

Key phrase set . ( is the set for disease topic, is the set for symptom topic, is the set for drug topic, is the set for check topic, is the set for the positive food topic, and is the set for the negative food topic.)

The patient question in conversation turn .

Similarity coefficient for checking whether the phrase is inside of the patient question

Output:

Question topic tuple in conversation turn

Main:

1: ={}

2: for in do

3:   for in do

4:       ifthen

5:         ifthen

6:            Append the “disease” topic in

7:         else ifthen

8:            Append the “symptom” topic in

9:         else ifthen

10:            Append the “drug” topic in

11:         else ifthen

12:            Append the “check” topic in

13:         else ifthen

14:            Append the “recommended food” topic in

15:         else ifthen

16:            Append the “not recommended food” topic in

17:       end if

18:   end for

19: end for