Distributed computing platform based multisource vertical knowledge graph classified integration query method

Abstract

The invention provides a distributed computing platform based multisource vertical knowledge graph classified integration query method. The distributed computing platform based multisource vertical knowledge graph classified integration query method comprises an integration control management system, a graph retrieval service system and a query control management system; the integration control management system builds a multisource vertical knowledge graph classified integration catalog model, carries out classified integration management for vertical knowledge graph sets stored at different positions according to logic category and is responsible for receiving a query task, splitting the query task into sub-tasks and sending the sub-tasks; the graph retrieval service system carries out query sub-tasks and realizes keyword group based knowledge graph related entity and relation query; and the query control management system manages the query task and executes the concurrent query sub-tasks. The distributed computing platform based multisource vertical knowledge graph classified integration query method can realize parallel query of multiple knowledge graphs, so that a user can finish the keyword group based knowledge graph related entity and relation information query without carrying out global mode data integration; and the distributed computing platform based multisource vertical knowledge graph classified integration query method can be used for building a cross-domain question-answering system and an application system recommendation platform.

Description

Technical field [0001] The invention belongs to the field of information processing and can be used to construct a cross-domain question answering system and an application system recommendation platform, and specifically relates to a multi-source vertical knowledge graph classification integrated query technology based on a distributed computing platform. Background technique [0002] The knowledge graph is essentially a knowledge base called a semantic network, that is, a knowledge base with a directed graph structure, where the nodes of the graph represent entities or concepts, and the edges of the graph represent Various semantic relationships between entities / concepts. The concept of Knowledge Graph was formally proposed by Google in 2012. It aims to realize a smarter search engine. It has been popularized in academia and industry since 2013. It has been widely used in chat robots and Q&A systems. To assist in the in-depth understanding of human language and support reasoni...

AI Technical Summary

Problems solved by technology

For the knowledge fusion of heterogeneous knowledge graphs, it will involve the merging of synonymous entities and relationships in different knowledge graphs, as well as the resulting entity / relationship inconsistency detection problems, some of which can be solved automatically by machine learning methods, and some can only be solved manually Merge, as the amount of data continues to increase, the knowledge fusion process is more complicated and time-consuming

[0006] 2. In terms of data quality: knowledge graphs in different fields are constructed by experts or institutions in their respective fields, and the quality of the data is relatively high. However, in the process of knowledge fusion, there will still be inconsistent descriptions of the same knowledge in different knowledge graphs. Sometimes manual review, proofreading, and merging are required, which consumes a lot of manpower

Therefore, when only part of the domain data needs to be calculated, all the split knowledge graphs still need to be processed, wasting computing resources

Images