Triple credibility evaluation method based on multi-source knowledge graph
A knowledge graph and triplet technology, which is applied in the field of trusted evaluation of triplet data in knowledge graphs, can solve the problems of insufficient use of semantic information and graph topology characteristics.
Pending Publication Date: 2021-03-16
SOUTHEAST UNIV
View PDF1 Cites 7 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] 1) First of all, it does not make full use of semantic information and graph topology features;
[0006] 2) The method of the above-mentioned patent is oriented to a single knowledge graph, and does not deal with the situation of multi-source knowledge graph
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0081] see figure 1 , the present invention is based on a multi-source knowledge graph triples credible assessment method, comprising the following steps:
[0082] S1. For a given multi-source knowledge graph, put the entity, entity category, relationship and attribute information of the multi-source knowledge graph into a common space by means of continuous numbering. At the same time, the attributes of the multi-source knowledge graph are represented and learned to obtain the embedded information of the attributes;
[0083] S101. For a given multi-source knowledge map, under the unified numbering space, uniformly number the entity name, entity type, attribute key value and relationship name with the source of the map, and generate an index from the unified numbering space such as the entity name The mapping relationship to the data source name; in the common space, the attribute value is divided to obtain the text collection of all attributes;
[0084] S102. Based on the t...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a triple credibility evaluation method based on a multi-source knowledge graph, and the method comprises the steps: S1, respectively obtaining the corresponding vector representations of an entity name, an entity type, an attribute key value and a relation name for a given multi-source knowledge graph; and obtaining vector representation of attribute values of the multi-source knowledge graph in an embedding space; s2, evaluating the credibility of the triad in a single knowledge graph; s3, evaluating the credibility of a single knowledge graph data source in the multi-source knowledge graph; s4, enabling credibility evaluation of the triples in the common space under the condition of the multi-source knowledge graph to depend on the credibility obtained by the single knowledge graph and the mutual influence between the multi-source knowledge graphs, and enabling the mutual influence between the multi-source knowledge graphs to depend on by the triples, namely interactive triples, influencing different data sources; and S5, carrying out a training step based on multi-source knowledge graph credibility evaluation. The method provided by the invention improvesthe accuracy of the triad credibility evaluation.
Description
technical field [0001] The invention belongs to the technical field of credible evaluation of triplet data by knowledge graphs, and in particular relates to a credible evaluation method based on multi-source knowledge graphs. Background technique [0002] With the wide application of knowledge graphs, knowledge graphs related to various fields have been established accordingly. How to analyze and select data with high reliability in the case of multiple data sources is a difficult problem at present. First of all, the construction of the knowledge map cannot completely avoid the introduction of wrong information: the source text of the information extraction has errors, and the construction process has not passed the quality assessment to screen out such wrong information; another example is the deviation of the automatic extraction technology itself that leads to errors in the process of extracting information information. Error information will greatly affect the perform...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/36
CPCG06F16/367
Inventor 王萌秦旭陆保国漆桂林刘哲一姚茜
Owner SOUTHEAST UNIV