C
C
cawadead2022-03-16 11:23:00
Mathematics
cawadead, 2022-03-16 11:23:00

How to distribute documents among people by keywords?

There is a set of people, each of which is assigned a set of some keywords. There is also a set of documents, each of which has a set of keywords. I need to distribute these documents among people by keywords. Of the possible options, I got only distribution by clustering, setting the number of clusters equal to the number of people and then predicting which cluster the document will belong to. To what extent is this correct? Perhaps something else should be done?

Answer the question

In order to leave comments, you need to log in

1 answer(s)
W
Wataru, 2022-03-16
@wataru

Define some metric. Or the number of common words for a person and a document, or somehow it is normalized, like what part of their words they have in common. After building the distance matrix, you can already solve your problem.
If documents can be distributed irregularly, then just give each document to the nearest person.
If it is necessary to somehow normalize the load, then the assignment problem arises, which can be solved by the minimum cost maximum flow algorithm. You can introduce penalties for overloading each person, and then the algorithm can issue documents to non-nearest people who have little work.

Didn't find what you were looking for?

Ask your question

Ask a Question

731 491 924 answers to any question