US 2023/0401461 A1 · Intelligent Collaborative Decision Generation System with Link Prediction Assisted by User Feedback

Abstract

This application relates generally to intelligent and explainable link prediction in knowledge graph systems that automatically incorporate user feedback. In one aspect, this application discloses an iterative process for predicting a link set as a group of links in a knowledge graph in an embedding space by expanding the knowledge graph with predicted and validated single links in each iteration such that a final set of links are predicted with each one being added to the set depending on previously added predicted links. In another aspect, this application also discloses automatically extracting rules from user feedback of link predictions and generating a user feedback knowledge graph from the extracted rules, which in combination with an original knowledge graph are used for the generation of the link predictions.

Claims20

Dependent claims

A system comprising at least one memory for storing computer instructions and a processing circuitry in communication with the at least one memory, the processing circuitry, when executing the computer instructions, is configured to:

ingest a ground-truth knowledge graph comprising a plurality of nodes and a plurality of ground-truth edges;
generate a feedback knowledge graph comprising at least a subset of the plurality of nodes and a plurality of predicted edges based on the ground-truth knowledge graph and user feedback of the predicted edges;
combine the ground-truth knowledge graph and the feedback knowledge graph into a training knowledge graph comprising training nodes and training edges;
convert the training nodes and the training edges into a set of embedding vectors in a multi-dimensional embedding space;
and
generate the plurality of predicted edges using a link prediction procedure based on the set of embedding vectors in the multi-dimensional embedding space.

3 more elements

A method performed by a system comprising at least one memory for storing computer instructions and a processing circuitry in communication with the at least one memory, the method comprising:

ingesting a ground-truth knowledge graph comprising a plurality of nodes and a plurality of ground-truth edges;
generating a feedback knowledge graph comprising at least a subset of the plurality of nodes and a plurality of predicted edges based on the ground-truth knowledge graph and user feedback of the predicted edges;
combining the ground-truth knowledge graph and the feedback knowledge graph into a training knowledge graph comprising training nodes and training edges;
converting the training nodes and the training edges into a set of embedding vectors in a multi-dimensional embedding space;
and
generating the plurality of predicted edges using a link prediction procedure based on the set of embedding vectors in the multi-dimensional embedding space.

3 more elements

Dependencies are inferred from the claim text.

Intelligent Collaborative Decision Generation System with Link Prediction Assisted by User Feedback

Life of this patent

Intelligent Collaborative Decision Generation System with Link Prediction Assisted by User Feedback

Life of this patent

Abstract

Claims20