Knowledge graph entity relationship mining method based on path sorting
The invention discloses a knowledge graph entity relationship mining method based on path sorting. The method comprises the steps of obtaining a path and a path partition mapping table of a to-be-mined file; determining a lexicographical order of a path of the to-be-mined file; obtaining a minimum a...
Saved in:
Main Author | |
---|---|
Format | Patent |
Language | Chinese English |
Published |
08.11.2019
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a knowledge graph entity relationship mining method based on path sorting. The method comprises the steps of obtaining a path and a path partition mapping table of a to-be-mined file; determining a lexicographical order of a path of the to-be-mined file; obtaining a minimum absolute path lexicographical order; determining partition information to which the target absolute path belongs; taking all files in a partition corresponding to the target partition information as a file set to which the mining file belongs; analyzing a data mining process; determining a node for providing physical resources for each execution step; determining the maximum data volume of the input data capable of being processed by the distributed system; and performing mining processing on thedata to be mined according to each execution step. According to the method, related files can be mined in a large-scale file storage system, the storage space is remarkably reduced, the file mining performance is improved, t |
---|---|
Bibliography: | Application Number: CN201910499161 |