System and method for user modeling to enhance named entity recognition
First Claim
Patent Images
1. A named entity (NE) list generator, comprising:
- a NE identifier configured to receive information from a plurality of different sources and to identify NEs in the information, at least one of the plurality of different sources of the information including a plurality of NEs referred to by a given group of users; and
an analyzer calculating probabilities associated with the identified NEs to obtain a user group-specific NE list, specific to the given group of users, with associated probabilities.
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention employs user modeling to model a user'"'"'s behavior patterns. The user'"'"'s behavior patterns are then used to influence named entity (NE) recognition.
25 Citations
30 Claims
-
1. A named entity (NE) list generator, comprising:
-
a NE identifier configured to receive information from a plurality of different sources and to identify NEs in the information, at least one of the plurality of different sources of the information including a plurality of NEs referred to by a given group of users; and an analyzer calculating probabilities associated with the identified NEs to obtain a user group-specific NE list, specific to the given group of users, with associated probabilities. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21)
-
-
22. A method of generating a named entity (NE) model, comprising:
-
receiving information from a source indicative of NE usage patterns of a given user, the NE usage patterns being indicative of a set of NEs referred to by the given user, wherein receiving information comprises receiving information from a plurality of different sources; and generating a user-specific NE model, specific to the given user, based on the information received, wherein generating comprises identifying NFs in the information received and calculating a probability associated with each NP based on the information received, and wherein calculating the probability comprises calculating by weighting and information received differently based on the source from which it was received. - View Dependent Claims (23, 24, 25, 26, 27, 28)
-
-
29. A computer readable medium having stored thereon computer readable instructions which, when executed by the computer cause the computer to perform steps of:
-
receiving information from a plurality of different information sources, the information being indicative of a specific user'"'"'s usage of named entities (NEs); generating a list of NEs in the information; calculating probabilities associated with each NE in the list based on the information; and integrating the NE list and probabilities into a computer component. - View Dependent Claims (30)
-
Specification