Relation extraction using full dependency forests
AssigneeTencent America LLC
InventorLinfeng Song
Life of this patent
Estimating the term…
Abstract
A method, computer program, and computer system is provided for extracting relations between one or more entities in a sentence. A forest corresponding to probabilities of relations between each pair of the entities is generated, and the generated forest generated forest is encoded with relation information for each of the pairs of entities. One or more features are extracted based on the generated forest and the encoded relation information, and a relation is predicted between the entities of each pair of entities based on the extracted features.
Claims20
A method of extracting relations between one or more entities in a sentence by a computer, the method comprising:
- generating a forest comprising a three-dimensional tensor and corresponding to probabilities of relations between each pair of the entities;
- encoding the generated forest with relation information for each of the pairs of entities;
- correlating representation vectors associated with target mentions corresponding to the entities;
- generating convolution kernels parameterized with the target mentions;
- applying the generated convolution kernels to the correlated representation vectors;
- and
- extracting features based on the generated forest, the encoded relation information, and the correlated representation vectors to which the generated convolution kernels have been applied;
- and
- machine-learning a relation between the entities of each pair of entities based on the extracted features and the three-dimensional tensor,
- wherein first and second dimensions of the tensor comprise the entities, and
- wherein a third dimension of the tensor comprises the target mentions each of which indicating ones of relationships between ones of the entities and a plurality of others of the entities.
A computer system for extracting relations between one or more entities in a sentence, the computer system comprising:
- one or more computer-readable non-transitory storage media configured to store computer program code;
- and
- one or more computer processors configured to access said computer program code and operate as instructed by said computer program code, said computer program code including:
- generating code configured to cause the one or more computer processors to generate a forest comprising a three-dimensional tensor and corresponding to probabilities of relations between each pair of the entities,
- encoding code configured to cause the one or more computer processors to encode the generated forest with relation information for each of the pairs of entities;
- correlating code configure to cause the one or more computer processors to correlate representation vectors associated with target mentions corresponding to the entities;
- further generating code configured to cause the one or more computer processors to generate convolution kernels parameterized with the target mentions;
- applying code configured to cause the one or more computer processors to apply the generated convolution kernels to the correlated representation vectors;
- extracting code configured to cause the one or more computer processors to extract features based on the generated forest the encoded relation information, and the correlated representation vectors to which the generated convolution kernels have been applied;
- and
- machine-learning code configured to cause the one or more computer processors to machine-learn a relation between the entities of each pair of entities based on the extracted features and the three-dimensional tensor,
- wherein first and second dimensions of the tensor comprise the entities, and
- wherein a third dimension of the tensor comprises the target mentions each of which indicating ones of relationships between ones of the entities and a plurality of others of the entities.
A non-transitory computer readable medium having stored thereon a computer program for extracting relations between one or more entities in a sentence, the computer program configured to cause one or more computer processors to:
- generate a forest comprising a three-dimensional tensor and corresponding to probabilities of relations between each pair of the entities,
- encode the generated forest with relation information for each of the pairs of entities;
- correlate representation vectors associated with target mentions corresponding to the entities;
- generate convolution kernels parameterized with target mentions;
- apply the generated convolution kernels to the correlated representation vectors;
- and
- extract features based on the generated forest the encoded relation information, and the correlated representation vectors to which the generated convolution kernels have been applied;
- and
- machine-learning a relation between the entities of each pair of entities based on the extracted features and the three-dimensional tensor,
- wherein first and second dimensions of the tensor comprise the entities, and
- wherein a third dimension of the tensor comprises the target mentions each of which indicating ones of relationships between ones of the entities and a plurality of others of the entities.
Dependencies are inferred from the claim text.