| CPC G06F 16/9024 (2019.01) [G06F 16/3344 (2019.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] | 25 Claims |
|
1. A method for building a new knowledge graph, the method comprising:
receiving at least one existing knowledge graph;
sampling random walks through the at least one existing knowledge graph;
determining embedding vectors for vertices and edges of the sampled random walks;
training of a machine-learning model thereby building a trained machine-learning model enabled to predict sequences of terms, taking as input sequences of the embedding vectors of the random walks;
providing a set of documents;
determining sequences of terms from phrases from documents from the set of documents;
building sequences of embedding vectors from the determined sequences of terms from the phrases;
using the built sequences of embedding vectors from the determined sequences of terms from the phrases as input for the trained machine-learning model for predicting second sequences of terms; and
merging the predicted second sequences of terms thereby building the new knowledge graph.
|