Generating a user-specific ranking model on a user electronic device

Patent metadata

Field	Value
Publication number	US-2016299899-A1
Application number	US-201515100337-A
Country	US
Kind code	A1
Filing date	Jan 27, 2015
Priority date	Aug 19, 2014
Publication date	Oct 13, 2016
Grant date	—

Abstract

There is disclosed a method of generating a user-specific ranking model on an electronic device associated with a user. The method is executable on the electronic device. The method comprises: receiving, from a search engine server, via a communication network, an indication of a resource-specific feature; appreciating a user interaction with the web resource performed by the user using the electronic apparatus; based on the user interaction, determining a value parameter for the web resource; generating the user-specific ranking model on the basis of the value parameter and the resource-specific feature.

First claim

1 . A method of generating a user-specific ranking model on an electronic device associated with a user, the method executable on the electronic device and comprising:
receiving, by the electronic device, from a search engine server, via a communication network, an indication of a resource-specific feature associated with a web resource; appreciating, by the electronic device, a user interaction with the web resource performed by the user using the electronic device; based on the user interaction, determining, by the electronic device, a value parameter for the web resource; generating, by the electronic device, the user-specific ranking model on the basis of the value parameter and the resource-specific feature.

2 . The method of claim 1 , further comprising:
transmitting to the search engine server the user-specific ranking model, the user-specific ranking model to be combined with a generic ranking model available on the search engine server to create a combined ranking model, the combined ranking model for ranking search results generated on search queries originated from the user.

3 . The method of claim 1 , further comprising:
transmitting to the search engine server the user-specific ranking model, the user-specific ranking model to be used with a generic ranking model available on the search engine to rank search results generated on search queries originated from the user.

4 . The method of claim 3 , wherein said user-specific ranking model and said generic ranking model are for:
ranking search results using the generic ranking model to generate a generically ranked search results list; and re-ranking generically ranked search results list using the user-specific ranking model to generate a user-centric ranked search results list.

5 . The method of claim 1 , wherein said resource-specific feature is the same resource-specific feature that is used by a generic ranking model used by the search engine server for ranking search results.

6 . The method of claim 1 , wherein said resource-specific feature comprises a plurality of resource-specific features each of said plurality of resource-specific features being the same resource-specific feature used in a generic ranking model used by the search engine server for ranking search results.

7 . The method of claim 1 , wherein said user interaction is indicative of a user preference associated with the web resource.

8 . The method of claim 1 , wherein said generating comprises applying a machine learning algorithm to create the user-specific ranking model using the resource-specific feature and the value parameter as a training set.

9 . The method of claim 1 , wherein said resource-specific feature is selected so that to optimize tracking of the user interaction on the electronic device.

10 . The method of claim 1 , further comprising:
receiving, from the user, a search request; sending the search request to the search engine server; transmitting to the search engine server the user-specific ranking model, receiving from the search engine a ranked search result list, the ranked search result list having been ranked by the search engine server by executing the steps of:
ranking search results using the generic ranking model to generate a generically ranked search results list; and
re-ranking generically ranked search results list using the user-specific ranking model to generate a user-centric ranked search results list.

11 . The method of claim 1 , further comprising:
receiving, from the user, a search request; sending the search request to the search engine server; transmitting to the search engine server the user-specific ranking model, the user-specific ranking model to be combined with a generic ranking model available on the search engine server to create a combined ranking model; receiving from the search engine a ranked search result list, the ranked search result list having been ranked by the search engine server by applying the combined ranking model.

12 . The method of claim 1 , wherein said resource-specific feature is selected on the basis of a generic ranking model used by the search engine server to rank search results.

13 . The method of claim 12 , wherein said generic ranking model is in itself personalized based on a server-tracked user interaction associated with the resource-specific feature.

14 . The method of claim 13 , wherein said server-tracked user interaction associated with resource-specific feature is of a different type than said user interaction tracked by the electronic device.

15 . A method of generating a user-centric ranking model, the method executable at a search engine server, the search engine server being coupled to an electronic device via a communications network, the electronic device being associated with a user, the method comprising:
transmitting to the electronic device, via the communication network, an indication of a resource-specific feature; receiving, by the electronic device, from the communication device, via the communication network, a resource-specific feature ranking model, the resource-specific feature ranking model having been generated by the electronic device by executing the steps of:
appreciating, by the electronic device, a user interaction with the web resource performed by the user using the electronic device;
based on the user interaction, determining, by the electronic device, a value parameter for the web resource;
generating, by the electronic device, the user-specific ranking model on the basis of the value parameter and the resource-specific feature.

16 . The method of claim 15 , the method further comprises selecting the resource-specific feature for transmission to the electronic device.

17 . The method of claim 16 , wherein said resource-specific feature is the same resource-specific feature that is used in a generic ranking model used by the search engine server for ranking search results.

18 . The method of claim 15 , wherein said resource-specific feature comprises a plurality of resource-specific features each of said plurality of resource-specific features being the same resource-specific feature used in a generic ranking model used by the search engine server for ranking search results.

19 . The method of claim 15 , wherein said selecting comprises choosing the resource-specific feature so that to optimize tracking of the user interaction on the electronic device.

20 . The method of claim 15 , wherein said user interaction is indicative of a user preference associated with the web resource.

21 . The method of claim 15 , further comprising:
receiving, from the electronic device, a search request; generating a set of search results responsive to the search query; ranking the set of search results using a generic ranking model to generate a generically ranked search results list; and re-ranking generically ranked search results list using the user-specific ranking model to generate a user-centric ranked search results list; sending the user-centric ranked search result list the electronic device.

22 . The method of claim 15 , further comprising:
combining user-specific ranking model with a generic ranking model available on the search engine server to create a combined ranking model.

23 . The method of claim 22 , further comprising:
receiving, from the electronic device, a search request; generating a set of search results responsive to the search query;

Assignees

• Yandex Europe Ag

Inventors

• Logachev Alexander Alexandrovich

Classifications

• G06F17/3053 Primary

  Physics · mapped topic

• G06N99/005

  Physics · mapped topic

• G06F17/30864

  Physics · mapped topic

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06N20/00

  Machine learning · CPC title