Device-Specific Search Results

• US 20160188730A1
• Filed: 04/10/2015
• Published: 06/30/2016
• Est. Priority Date: 12/29/2014
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Quixey, Inc.
• Inventors: Delli Santi, James, Glover, Eric J.
• Granted Patent: US 10,366,127 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/235 Update request formulation
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A method includes receiving, by a processing system, a search query and a device type identifier from a user device and identifying, by the processing system, a consideration set of application records based on the search query and the device type identifier. Each application record of the consideration set corresponds to a respective software application and has an initial score associated therewith indicating a degree to which the application record matches the search query. The method also includes determining, by the processing system, a result score of the application record based on the initial score and features corresponding to the application record for each application record. The method further includes generating, by the processing system, device-specific search results based on the consideration set of application records and the result scores thereof and transmitting, by the processing system, the device-specific search results to the user device.

Claims

1. A method comprising:

receiving, by a processing system, a search query and a device type identifier from a user device, the search query including one or more query terms and the device type identifier being indicative of a device type of the user device;

identifying, by the processing system, a consideration set of application records based on the search query and the device type identifier, each application record of the consideration set corresponding to a respective software application and having an initial score associated therewith indicating a degree to which the application record matches the search query, wherein at least one of the application records contains a device-specific review corresponding to the device type of the user device that includes at least one keyword that matches to at least one of the query terms;

for each application record, determining, by the processing system, a result score of the application record based on the initial score and features corresponding to the application record;

generating, by the processing system, device-specific search results based on the consideration set of application records and the result scores thereof, the device-specific search results indicating one or more software applications that are relevant to the search query; and

transmitting, by the processing system, the device-specific search results to the user device.

2. The method of claim 1, wherein identifying the consideration set comprises:

determining one or more tokens with which to query a search index based on the query terms, wherein the search index is keyed by a plurality of keywords, each keyword being associated with one or more application records that contain the keyword, and each association between the keyword and a particular application record of the one or more application records being tagged with a general tag and/or one or more device-specific tags, the general tag indicating that the keyword is found in a device-independent portion of the particular application record and each device-specific tag respectively corresponding to a known device type of a plurality of known device types and indicating that the keyword is found in a device-specific portion of the particular application record corresponding to the known device type;

for each token, querying the search index using the token and the device type identifier, and receiving, from the search index, application identifiers associated with the keyword represented by the token and tagged with at least one of a general tag or a device-specific tag corresponding to the device type identifier; and

adding the application identifiers in the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.

3. The method of claim 2, wherein for each application record containing a device-specific portion corresponding to a particular device type, the device-specific portion is based on one or more device-specific reviews of the software application that indicate the particular device type.

4. The method of claim 3, wherein for each application record containing the device-specific portion corresponding to the particular device type, the device-specific portions are obtained from one or more scraped feedback units, each scraped feedback unit corresponding to the particular device type and including text that includes the device-specific review.

5. The method of claim 4, wherein the device-specific reviews of the software application identified in the application record are determined by:

obtaining, by the processing system, one or more documents relating to the software application from one or more data sources, at least one of the one or more documents including a plurality of feedback units corresponding to the software application, the feedback units each including text representing a review of the software application;

identifying, by the processing system, a subset of feedback units from the plurality of feedback units that pertain to the particular device type; and

for each feedback units in the subset of feedback units;

identifying, by the processing system, the text representing the review of the software application; and

associating, by the processing system, the text representing the review of the software application with the particular device type in the application record of the software application.

6. The method of claim 5, wherein a device-specific tag corresponding to the particular device type, when applied to an association between a keyword of the search index and the application record, indicates that the keyword is found in at least one device-specific review corresponding to the particular device type.

7. The method of claim 1, wherein identifying the consideration set comprises:

determining one or more tokens based on the query terms of the search query;

selecting a device-specific search index from a plurality of device-specific search indexes based on the device type identifier, the device-specific search index corresponding to the device type indicated by the device type identifier and being keyed by a first set of first keywords, each first keyword being associated with one or more application records that contain the first keyword in a device-specific portion of the one or more applications, the device-specific portions corresponding to the device type indicated by the device type identifier;

querying the device-specific search index with each of the one or more tokens, and receiving, from the device-specific search index, application identifiers of application records that match one or more of the tokens; and

adding the application identifiers to the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.

8. The method of claim 7, wherein for each application record indexed in the device-specific search index corresponding to the device type indicated by the device type identifier, the device-specific portion is based on one or more device-specific reviews of the software application that indicate a make and a model.

9. The method of claim 8, wherein for each application record indexed in the device-specific search index, the device-specific portions are obtained from one or more scraped feedback units, each scraped feedback unit corresponding to the device type and including text that includes the device-specific review.

10. The method of claim 9, wherein the device-specific reviews of the software application identified in the application record are determined by:

obtaining, by the processing system, one or more documents relating to the software application from one or more data sources, at least one of the one or more documents including a plurality of feedback units corresponding to the software application, each feedback unit including text representing a review of the software application;

identifying, by the processing system, a subset of feedback units from the plurality of feedback units that pertain to the device type; and

for each feedback unit in the subset of feedback units;

identifying, by the processing system, the text representing the review of the software application; and

associating, by the processing system, the text representing the review of the software application with the device type in the application record of the software application.

11. The method of claim 7, wherein identifying the consideration set further comprises:

querying a general search index with each of the one or more tokens, the general search index being keyed by a second set of second keywords, each second keyword being associated with one or more application records that contain the second keyword in a device-independent portion of the application record;

receiving, from the general search index, application identifiers of application records that match to one or more of the tokens; and

adding the application identifiers to the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.

12. A search system comprising:

a storage system that stores;

a plurality of application records, each application record respectively corresponding to a software application and including a plurality of attributes of the software application; and

a search index keyed by a plurality of keywords, each keyword being associated with one or more application records that contain the keyword, and each association between the keyword and a particular application record of the one or more application records being tagged with a general tag and/or one or more different device-specific tags, the general tag indicating that the keyword is found in a device-independent portion of the application record and each device-specific tag respectively corresponding to a known device type of a plurality of known device types and indicating that the keyword is found in a device-specific portion of the application record corresponding to the known device type; and

a processing system including one or more processors executing computer-readable instructions, the computer-readable instructions when executed by the processing system causing the processing system to;

receive a search query and a device type identifier from a user device, the search query including one or more query terms and the device type identifier being indicative of a device type of the user device;

identify a consideration set of application records based on the search query, the device type identifier, and the search index, each application record of the consideration set corresponding to a respective software application;

for each application record, determine a result score of the application record based on features corresponding to the application record;

generate device-specific search results based on the consideration set of application records and the result scores thereof, the device-specific search results indicating one or more software applications that are relevant to the search query; and

transmit the device-specific search results to the user device.

13. The search system of claim 12, wherein identifying the consideration set comprises:

determining one or more tokens based on the search query;

for each token, querying the search index using the token and the device type identifier, and receiving, from the search index, application identifiers associated with the keyword represented by the token and tagged with at least one of a general tag or a device-specific tag corresponding to the device type identifier; and

adding the application identifiers in the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.

14. The search system of claim 12, wherein for each application record containing a device-specific portion corresponding to a particular device type, the device-specific portion is based on one or more device-specific reviews of the software application that indicate the particular device type.

15. The search system of claim 14, wherein for each application record containing the device-specific portion corresponding to the particular device type, the device-specific portions are obtained from one or more scraped feedback units, each scraped feedback unit corresponding to the particular device type and including text that includes the device-specific review.

16. The search system of claim 15, wherein the device-specific reviews of the software application identified in the application record is determined by:

obtaining one or more documents relating to the software application from one or more data sources, at least one of the one or more documents including a plurality of feedback units corresponding to the software application, the feedback units each including text representing a review of the software application;

identifying a subset of feedback units from the plurality of feedback units that pertain to the particular device type; and

for each feedback units in the subset of feedback units;

identifying the text representing the review of the software application; and

associating the text representing the review of the software application with the particular device type in the application record of the software application.

17. The search system of claim 16, wherein a device-specific tag corresponding to the particular device type, when applied to an association between a keyword of the search index and the application record, indicates that the keyword is found in at least one device-specific review corresponding to the particular device type.

18. A search system comprising:

a storage system that stores;

a plurality of application records, each application record respectively corresponding to a software application and including a plurality of attributes of the software application;

a plurality of device-specific search indexes, each device-specific search index respectively corresponding to a different make and model of a user device and being keyed by a respective set of first keywords, each first keyword being associated with one or more application records that contain the first keyword in a device-specific portion of the one or more applications, the device-specific portions pertaining to the make and model corresponding to the device-specific search index; and

a general search index being keyed by a set of second keywords, each second keyword being associated with one or more application records that contain the second keyword in a device-independent portion of the application record; and

a processing system including one or more processors executing computer-readable instructions, the computer-readable instructions when executed by the processing system causing the processing system to;

receive a search query and a device type identifier from the user device, the search query including one or more query terms and the device type identifier being indicative of a make and model of the user device;

select a device-specific search index from the plurality of device-specific search indexes based on the device type identifier;

identify a consideration set of application records based on the search query, the selected device-specific search index, and the general search index;

for each application record, determine a result score of the application record based on features corresponding to the application record;

generate device-specific search results based on the consideration set of application records and the result scores thereof, the device-specific search results indicating one or more software applications that are relevant to the search query; and

transmit the device-specific search results to the user device.

19. The search system of claim 18, wherein for each application record indexed in the selected device-specific search index, the device-specific portion is based on one or more device-specific reviews of the software application that indicate the make and model indicated by the device type identifier.

20. The search system of claim 19, wherein for each application record indexed in the selected device-specific search index, the device-specific portions are obtained from one or more scraped feedback units, each scraped feedback unit corresponding to the make and model indicated by the device identifier and including text that includes the device-specific review.

21. The search system of claim 20, wherein the device-specific reviews of the software application identified in the application record are determined by:

obtaining one or more documents relating to the software application from one or more data sources, at least one of the one or more documents including a plurality of feedback units corresponding to the software application, the feedback units each including text representing a review of the software application;

identifying a subset of feedback units from the plurality of feedback units that pertain to the make and model; and

for each feedback units in the subset of feedback units;

identifying the text representing the review of the software application; and

associating the text representing the review of the software application with the make and model in the application record of the software application.

22. The search system of claim 18, wherein identifying the consideration set comprises:

determining one or more tokens based on the query terms of the search query;

querying the device-specific search index with each of the one or more tokens, and receiving, from the device-specific search index, application identifiers of application records that match one or more of the tokens; and

adding the application identifiers to the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.

23. The search system of claim 22, wherein identifying the consideration set comprises:

querying the general search index with each of the one or more tokens, and receiving, from the general search index, application identifiers of application records that match to one or more of the tokens; and

adding the application identifiers to the consideration set, each application identifier added to the consideration set respectively indicating an application record of the consideration set.