Subject-specific data set for named entity resolution
AssigneeAmazon Technologies, Inc.
InventorsChristos Christodoulopoulos, Arpit Mittal, Andrew Christopher Graham Hopkinson
Life of this patent
Estimating the term…
Abstract
A method comprising receiving subject data indicative of a subject entity. First entity data is obtained from a knowledge database using the subject data, the first entity data representative of a first related entity related to the subject entity. First text data is obtained from the knowledge database using the first entity data, the first text data representative of first text associated with the first related entity. A subject entity data set is generated. The subject entity data set comprises first related entity data based on the first entity data, and the first text data.
Claims20
A method of at least partially populating a first look-up table with first look-up table data related to a topic, the method comprising:
- receiving topic data indicative of the topic, the topic corresponding to a particular instance of a named entity;
- processing a knowledge database using the topic data to identify particular entity data representative of the particular instance of the named entity, the knowledge database comprising:
- entity data representative of a plurality of instances of named entities comprising the particular instance of a named entity;
- text data representative of a plurality of phrases, each of the plurality of phrases associated with an instance of the plurality of instances, respectively;
- and
- relationship data representative of relationships between respective pairs of instances of the plurality of instances,
- identifying first related entity data from the knowledge database, the first related entity data representative of a first instance of a named entity related to the particular instance of a named entity by a relationship which satisfies a relationship condition;
- obtaining first text data from the text data of the knowledge database the first text data representative of a first phrase associated with the first instance of a named entity;
- and
- storing the first related entity data and the first text data as first look-up table data in the first look-up table so as to indicate an association between the first related entity data and the first text data.
A method comprising:
- receiving topic data indicative of a topic corresponding to a particular instance of a named entity;
- obtaining first entity data from a knowledge database using the topic data, the first entity data representative of a first instance of a named entity related to the particular instance of a named entity;
- obtaining first text data from the knowledge database using the first entity data, the first text data representative of a first phrase associated with the first instance of a named entity;
- and
- generating a data set related to the topic, the data set comprising:
- first related entity data based on the first entity data;
- and
- the first text data.
A system comprising:
- at least one processor;
- at least one memory comprising computer program instructions,
- the at least one memory and the computer program instructions operable to, with the at least one processor:
- receive topic data indicative of a topic corresponding to a particular instance of a named entity;
- obtain first entity data from a knowledge database using the topic data, the first entity data representative of a first instance of a named entity related to the particular instance of a named entity;
- obtain first text data from the knowledge database using the first entity data, the first text data representative of a first phrase associated with the first instance of a named entity;
- and
- generate a data set related to topic, the data set comprising:
- first related entity data based on the first entity data;
- and
- the first text data.
Dependencies are inferred from the claim text.