System and method for detecting a search intent
First Claim
1. A computerized method for detecting a search intent of a user, comprising:
- receiving an input search query from a user device;
tokenizing the input search query to at least one token to create at least one tokenized query;
processing the at least one tokenized query by a plurality of engines, wherein each engine of the plurality of engines configured to compute a certainty score that indicates a probability that the at least one tokenized query is mapped to at least one entity, wherein each engine of the plurality of engines is further configured to correspond to at least one entity indicating a topic of interest, thereby the plurality of engines are configured with different entities;
receiving from a set of at least two engines of the plurality of engines their respective entities and computed certainty scores, wherein the set of engines output computed certainty scores above a predefined threshold; and
analyzing the received certainty scores and the respective entities to determine the search intent.
2 Assignments
0 Petitions
Accused Products
Abstract
A method and system for detecting a search intent of a user are provided. The method includes receiving an input search query from a user device; tokenizing the input search query to at least one token to create at least one tokenized query; processing the at least one tokenized query by a plurality of engines, wherein each engine of the plurality of engines configured to compute a certainty score that indicates a probability that the at least one tokenized query is mapped to at least one entity, wherein each engine of the plurality of engines is further configured to correspond to at least one entity indicating a topic of interest, thereby the plurality of engines are configured with different entities; receiving from a set of engines of the plurality of engines their respective entities and computed certainty scores, wherein the set of engines output computed certainty scores above a predefined threshold; and analyzing the received certainty scores and the respective entities to determine the search intent.
84 Citations
18 Claims
-
1. A computerized method for detecting a search intent of a user, comprising:
-
receiving an input search query from a user device; tokenizing the input search query to at least one token to create at least one tokenized query; processing the at least one tokenized query by a plurality of engines, wherein each engine of the plurality of engines configured to compute a certainty score that indicates a probability that the at least one tokenized query is mapped to at least one entity, wherein each engine of the plurality of engines is further configured to correspond to at least one entity indicating a topic of interest, thereby the plurality of engines are configured with different entities; receiving from a set of at least two engines of the plurality of engines their respective entities and computed certainty scores, wherein the set of engines output computed certainty scores above a predefined threshold; and analyzing the received certainty scores and the respective entities to determine the search intent. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. An apparatus for detecting a search intent of a user, comprising:
-
a processor; an interface configured to receive an input search query from a user device over a network; a tokenizer configured to tokenize the input search query to at least one token to create at least one tokenized query; a plurality of engines, wherein each engine of the plurality of engines configured to compute a certainty score that indicates a probability that the at least one tokenized query is mapped to at least one entity, wherein each engine of the plurality of engines is further configured to include at least one entity indicating a topic of interest, thereby the plurality of engines are configured with different entities; an analyzer configured to receive from a set of at least two engines of the plurality of engines their respective entities and computed certainty scores, wherein the set of engines output computed certainty scores above a predefined threshold, wherein the analyzer is further configured to determine the search intent using the received certainty scores and the respective entities to determine the search intent; and a memory for saving at least the tokenized queries and the determined search intent. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18)
-
Specification