Ranking of search results

Patent metadata

Field	Value
Publication number	US-10331680-B2
Application number	US-201514981763-A
Country	US
Kind code	B2
Filing date	Dec 28, 2015
Priority date	Dec 28, 2015
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

Methods for ranking search results are disclosed. The methods include obtaining an unranked set of functionally similar search that all correspond to an action. The methods further include retrieving a ranking vector corresponding to the action. The ranking vector identifies a plurality of software applications that perform a function corresponding to the action and a ranking range for each of the plurality of software applications. The methods further include iteratively ranking the functionally similar search results of the unranked set to obtain a ranked set of functionally similar search results. The ranking is based on the retrieved ranking vector and a random number generator that generates random numbers between the minimum value and the maximum value of the ranking vector.

First claim

What is claimed is:

1. A method comprising:
receiving, by a search system, a search query from a remote device;
identifying, by the search system, an unranked set of functionally similar search results based on the search query, the unranked set of functionally similar search results collectively linking to states of two or more software applications, each search result linking to a specific state of a respective software application, and the unranked set of functionally similar search results corresponding to an action;
retrieving, by the search system, a ranking vector corresponding to the action, the ranking vector identifying a plurality of software applications that perform a function corresponding to the action and a ranking range for each of the plurality of software applications, each ranking range having a respective upper value and a respective lower value, the ranking ranges of the ranking vector being non-overlapping and collectively defining a total range having a minimum value of the ranking vector and a maximum value of the ranking vector;
generating, by the search system, a random number between the minimum value and the maximum value of the ranking vector;
iteratively ranking, by the search system, the functionally similar search results of the unranked set, the ranking being based on the retrieved ranking vector and the random number;
obtaining, by the search system, a ranked set of the functionally similar search results; and
transmitting, by the search system, the ranked set of the functionally similar search results to the remote device.

2. The method of claim 1 , wherein the retrieving of the ranking vector includes querying a ranking vector data store with an action identifier of the action, the ranking vector data store outputting a global ranking vector corresponding to the action, the global ranking vector being indicative of collective preferences of a group of users when leveraging functions of applications corresponding to the action.

3. The method of claim 2 , wherein the querying of the ranking vector data store includes:
querying the ranking vector data store with the action identifier of the action and a region identifier indicating a region associated with the remote device, the ranking vector data store outputting:
a regional ranking vector when regional ranking vector corresponding to the region identifier and the action exists in the ranking vector data store, the regional ranking vector being indicative of collective preferences of a group of users when leveraging functions of applications corresponding to the action from the region indicated by the region identifier, and
the global ranking vector corresponding to the action when the ranking vector data store does not include a regional ranking vector corresponding to both the region identifier and the action identifier.

4. The method of claim 2 , further comprising:
receiving, by the search system, engagement statistics corresponding to the ranked set of functionally similar search results, the engagement statistics indicating a user selection of one of the functionally similar search results; and
updating, by the search system, the ranges in the global ranking vector based on the engagement statistics.

5. The method of claim 1 , wherein the iterative ranking of the functionally similar search results of the unranked set includes:
determining an application of the plurality of software applications based on the random number and the ranking vector;
removing a search result from the unranked set that links to the determined application; and
appending the removed search result to the ranked set of search results.

6. The method of claim 5 , wherein the removing of the search result from the unranked set includes:
when more than one search result link to the determined application, removing the search result from the unranked set with the highest relevance score.

7. The method of claim 1 ,
wherein the iterative ranking of the functionally similar search results of the unranked set includes:
instantiating the ranked set of search results, the ranked set of search results initially being an empty set,
determining an application of the plurality of software applications based on the random number and the ranking vector,
removing a search result from the unranked set that links to the determined application, and
appending the removed search result to the ranked set of search results, and

wherein the determining of the application, the removing of the search result, and the appending of the removed search result are iteratively performed until the unranked set is empty.

8. The method of claim 7 , wherein the removing of the search result from the unranked set includes:
when more than one search result link to the determined application, removing the search result from the unranked set with the highest relevance score.

9. The method of claim 1 , wherein the identifying of the unranked set of functionally similar search results includes:
identifying a plurality of search results based on the search query, the plurality of search results including the functionally similar search results, and each search result linking to a state of a respective software application and including an action identifier corresponding to a function performed at the linked to state; and
grouping the plurality of search results based on the action identifiers defined therein to obtain a plurality of unranked sets of functionally similar search results, including the unranked set of functionally similar search results, each unranked set corresponding to a different action.

10. The method of claim 9 , further comprising:
for each other unranked set of functionally similar search results of the plurality of unranked sets:
retrieving, by the search system, another ranking vector corresponding to the action of the other unranked set,
iteratively ranking, by the search system, the functionally similar search results of the other unranked set to obtain another ranked set of functionally similar search results, the ranking being based on the retrieved other ranking vector and the random number generator, and
transmitting, by the search system, the other ranked set to the remote device.

11. A portable device comprising:
a transceiver configured to receive a search query from a remote device;
a storage configured to store a ranking vector identifying a plurality of software applications that perform a function corresponding to an action and a ranking range for each of the plurality of software applications, each ranking range having a respective upper value and a respective lower value, the ranking ranges of the ranking vector being non-overlapping and collectively defining a total range having a minimum value of the ranking vector and a maximum value of the ranking vector; and
a processor configured to:
identify an unranked set of functionally similar search results based on the search query, the unranked set of functionally similar search results collectively linking to states of two or more software applications, each search result linking to a specific state of a respective software application, and the unranked set of functionally similar search results corresponding to an action,
retrieve, from the storage, a ranking vector corresponding to the action,
generate a random number between the minimum value and the maximum value of the ranking vector,
iteratively rank the functionally similar search results of the unranked set, the ranking being based on the retrieved ranking vector and the random number,
obtain a ranked set of the functionally similar search results, and
transmit, via the transceiver

Assignees

• Samsung Electronics Co Ltd

Inventors

• Glover Eric
• La Victoire Patrick

Classifications

• G06F16/9537

  Spatial or temporal dependent retrieval, e.g. spatiotemporal queries · CPC title

• G06F16/24578 Primary

  using ranking · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title