There are new students which should automatically be assigned to a coach (which than teaches the students).
The students fill out a survey with 30 multiple choice questions beforehand from which the answers should be used as our features.
My idea is to handle this as a multiclass classification problem where every coach is a class and we can match a student based on their survey answers.
One problem is, that the training data is very small with only a few hundred student-coach matches with survey answers made by hand which can be used as the training data.
Does this approach can yield good results for this problem or is there a better solution/approach for it?
question from:
https://stackoverflow.com/questions/65945178/which-approach-to-use-for-matching-new-students-to-a-coach-based-on-survey-answe 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…