A role permission-aware based PLM knowledge hierarchical retrieval method and system

Abstract

The application discloses a PLM knowledge hierarchical retrieval method based on role permission awareness, which takes user permission information as a built-in constraint condition of a retrieval process, so that the retrieval process is executed within the permission boundary. The method first extracts semantic features of documents, constructs partition indexes, and analyzes and caches user permission data through preprocessing. When a user initiates retrieval, a query statement is subjected to semantic enhancement and a feature vector is generated, similar candidate documents are screened out in the classified index, and then, according to the user permission data set, the documents that the user has no access to are removed from the candidate results, and finally, the remaining results are sorted according to the classified index and output for display. The application can solve the technical problem that the existing traditional retrieval method based on keyword matching only depends on the literal matching of words, cannot understand the deep semantic association between the query intention and the document content, and results in low recall rate and accuracy of the retrieval results.