Analysis of mobile application reviews based on content, reviewer credibility, and temporal and geographic clustering

Patent metadata

Field	Value
Publication number	US-2016283497-A1
Application number	US-201514671082-A
Country	US
Kind code	A1
Filing date	Mar 27, 2015
Priority date	Mar 27, 2015
Publication date	Sep 29, 2016
Grant date	—

Abstract

A method by a network node includes generating logically associated clusters of reviews submitted by users regarding an application program executed on user equipments operated by the users, generating a separate summary for each of the clusters of the reviews, and communicating the summaries to a network node.

First claim

1 . A method comprising:
performing operations as follows on a processor of a computer system: generating logically associated clusters of reviews submitted by users regarding an application program executed on user equipments operated by the users; generating a separate summary for each of the clusters of the reviews; and communicating the summaries to a network node.

2 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
for each of a plurality of the users, determining credibility of the user; and sorting the reviews among the clusters based on the credibility determined for the users who submitted the reviews.

3 . The method of claim 2 , wherein the determining credibility of the user, comprises:
computing a level of similarity between one of the reviews by the user to any of the other reviews by identifying a level of matching between keywords contained in the one of the reviews by the user to content of the other reviews; and generating a credibility score that indicates a reduced level of credibility based on determining a threshold level of matching between keywords contained in the one of the reviews by the user to content of a threshold number of the other reviews.

4 . The method of claim 2 , wherein the determining credibility of the user, comprises:
identifying an IP address of a user during submission of a review; and generating a credibility score based on whether the IP address is associated with a location within a defined geographic region.

5 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
obtaining information indicating timing of different versions of the application; correlating timing of the different versions of the application to timing of the reviews; and sorting the reviews among the clusters based on the correlating.

6 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
determining geographic locations of the users who submitted the reviews; and sorting the reviews among the clusters based on the geographic locations determined for the users.

7 . The method of claim 6 , wherein:
the determining geographic locations of the users who submitted the reviews, comprises:
identifying an IP address of a user during submission of a review; and
sorting the review among the clusters based on the IP address.

8 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
sorting the reviews among the clusters based on when the reviews were submitted to a network node of the computer system.

9 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
selecting a template among a set of templates based on an identifier for the application program, each of the templates defining a set of keywords and being associated with a different identifiers for application programs; identifying a level of matching between individual keywords of the template and content of one of the reviews; sorting the review among the clusters based on the level of matching.

10 . The method of claim 9 , wherein:
the keywords relate to terms that users will use to describe problematic operation of the application program; and wherein the generating a separate summary for each of the clusters of the reviews, comprises
generating a summary of reviews containing a threshold number of the keywords relating to terms that users will use to describe problematic operation of the application program.

11 . The method of claim 9 , wherein:
the keywords relate to terms that users will use to describe features of the application program; and wherein the generating a separate summary for each of the clusters of the reviews, comprises
generating a summary of reviews containing a threshold number of the keywords relating to terms that users will use to describe features of the application program.

12 . The method of claim 9 , wherein:
the keywords relate to terms that users will use to describe existing features of the application program and other terms that users will use to describe new features of the application program; and wherein the generating a separate summary for each of the clusters of the reviews, comprises:
generating one summary of reviews containing a threshold number of the keywords relating to terms that users will use to describe existing features of the application program; and
generating another summary of reviews containing a threshold number of the keywords relating to terms that users will use to describe new features of the application program.

13 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
identifying characteristics of the user equipments that executed the application program; and sorting the reviews among the clusters based on the characteristics of the user equipments.

14 . The method of claim 13 , wherein the identifying characteristics of the user equipments that executed the application program, comprises:
matching keywords, among a set of keywords defining characteristics of different user equipment hardware and software configurations, to content of the reviews.

15 . The method of claim 13 , wherein the identifying characteristics of the user equipments that executed the application program, comprises:
analyzing content of communications with the user equipments during submission of the reviews to identifying characteristics of the user equipments.

16 . The method of claim 1 , wherein the generating logically associated clusters of reviews submitted by users regarding the application program executed on user equipments operated by the users, comprises:
determining credibility of the users who submitted the reviews; obtaining information indicating timing of different versions of the application; correlating timing of the different versions of the application to timing of the reviews; determining geographic locations of the users who submitted the reviews; determining when the reviews were submitted to a network node of the computer system; selecting a template among a set of templates based on an identifier for the application program, each of the templates defining a set of keywords and being associated with a different identifiers for application programs; identifying level of matching between individual keywords of the template and content of the reviews; identifying characteristics of the user equipments that executed the application program; and sorting the reviews among the clusters based on the credibility of the users who submitted the reviews, the correlating between the timing of the different versions of the application to the timing of the reviews, the geographic locations of the users who submitted the reviews, when the reviews were submitted to the network node of the computer system, the leve

Assignees

• Ca Inc

Inventors

• Singh Satnam

Classifications

• G06F16/35 Primary

  Clustering; Classification · CPC title

• G06F17/3087 Primary

  Physics · mapped topic

• G06F17/30719

  Physics · mapped topic

• G06F17/30663

  Physics · mapped topic

• G06F17/30705

  Physics · mapped topic