Knowledge graph-based knowledge credibility measurement method

Abstract

The invention discloses a knowledge graph-based knowledge credibility measurement method, which comprises the following steps of: firstly, vectorizing entities and relationships in a knowledge graph by utilizing a knowledge representation learning algorithm, and performing calculating by utilizing an energy function and a sigmoid function to obtain relationship credibility; then, according to the attribute standard set, evaluating entity attribute values in the knowledge graph, and obtaining the attribute value credibility through calculation in a mode of combining lexical similarity and editing distance; then, constructing a question-answering system, and performing question solving credibility calculation on the knowledge in the knowledge graph by using the question-answering system and the question-answering standard set; and finally, in combination with the relationship credibility, the attribute value credibility and the problem solving credibility, performing weighted summation according to different weights to obtain the credibility of the knowledge graph. According to the method, error knowledge in the knowledge graph can be corrected, and the accuracy of the knowledge in the knowledge graph is improved.

Description

technical field [0001] The invention relates to knowledge graph knowledge credibility measurement in the field of knowledge graph improvement, in particular to a knowledge graph-based knowledge credibility measurement method. Background technique [0002] After Google proposed the concept of "knowledge graph" in 2012, knowledge graphs have been widely used in fields such as semantic search, intelligent question answering, and decision-making assistance. Use the knowledge graph to recommend search results, etc. [0003] Early knowledge graphs were basically constructed manually, which was not only very labor-intensive, but also unable to update new knowledge in the real world in a timely manner. Therefore, many researchers have been working on how to automatically extract knowledge from unstructured data, so that knowledge can be continuously obtained from the network to complement the existing knowledge graph. However, although automated construction technology saves a lot...

AI Technical Summary

Problems solved by technology

However, although automated construction technology saves a lot of manpower and can supplement new knowledge in a timely manner, such an approach will inevitably introduce a lot of wrong knowledge

At present, knowledge graphs are widely used in the fields of question answering, recommendation, and search, and these applications based on knowledge graphs assume that the knowledge in the knowledge graph is correct, which will inevitably have some potential problems, such as in auxiliary decision-making. knowledge can lead to wrong decisions with serious consequences

Images