Clustering method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a clustering method and device, equipment and a storage medium, and the method comprises the steps: obtaining question and answer data on a question and answer platform when group division is carried out for questioning parties on the question and answer platform, wherein the question and answer data comprises a questioning party identification for proposing a question and an answering party identification for answering the question; for each questioner, according to the corresponding relation between the questioner identifier and the answerer identifier, determining questioning behavior characteristics of the questioning party, wherein the questioning behavior characteristics are embodied by an answerer set for answering questions raised by the questioning parties, the subdivision aspects concerned by different questioning parties may be different, the subdivision aspects skilled by different answering parties are also different, the questioning parties can be clustered according to the questioning behavior similarity determined by the questioning behavior characteristics, a target number of questioning party group class clusters are obtained, and fine-grained division of the questioning parties is achieved. In addition, the method does not take the type of the object as a division basis, and can avoid the problem of difficult classification.

Description

technical field

[0001] The present application relates to the computer field, in particular to a clustering method, device, equipment and storage medium. Background technique

[0002] With the rapid development of Internet medical care, various online medical question-and-answer communities are becoming more and more popular among patients, such as "seeking medical advice" and "Lilac Garden". With the increasing number of patients, it is necessary for the platform to automatically cluster patients with similar diseases through technical means in order to establish sub-communities for specific types of patients, or to further provide some specific recommendation services for similar patients.

[0003] In the related art, the method for clustering patients often uses a classification model to identify the type of disease to which the question raised by the patient belongs, and then divides the patients into groups according to the disease, that is, clusters.

[0004] However,...

Images