Brand personality comparison engine

Patent metadata

Field	Value
Publication number	US-10387894-B2
Application number	US-201514839120-A
Country	US
Kind code	B2
Filing date	Aug 28, 2015
Priority date	Aug 28, 2015
Publication date	Aug 20, 2019
Grant date	Aug 20, 2019

Abstract

Mechanisms are provided to implement a brand comparison engine. The mechanisms receive a request to compare brand personalities of a first specified brand and a second specified brand and obtain a first brand personality scale associated with the first specified brand and a second brand personality scale associated with the second specified brand. The mechanisms calculate at least one gap value indicating a difference between at least one personality trait in the first brand personality scale and a corresponding at least one personality trait in the second brand personality scale. The mechanisms also output an output indicating an aspect of the at least one gap based on the calculation.

First claim

What is claimed is:

1. A method, in a data processing system comprising a processor and a memory, the memory comprising instructions which when executed by the processor cause the processor to implement a brand comparison engine, the method comprising:
generating, by a feature engine executing in the data processing system, one or more brand personality computer models, by performing natural language processing on natural language crowdsource information obtained from one or more crowdsource computing systems to extract features that are indicative of brand personality traits and relationships between the features and the brand personality traits;
training, by the feature engine, the one or more brand personality computer models, using a machine learning computer operation, to identify selected features as indicative of the brand personality traits, based on first crowdsource information obtained from the one or more crowdsource computing systems, and based on one or more ground truth data structures generated from one or more computer executed surveys of persons regarding the brand personality traits of a plurality of brands;
configuring the brand comparison engine to execute the trained one or more brand personality computer models to evaluate brand personalities of one or more brands;
receiving, by the brand comparison engine, a request to compare a first brand personality of a first specified brand and a second brand personality of a second specified brand;
generating, by the brand comparison engine, a first brand personality scale associated with the first specified brand and a second brand personality scale associated with the second specified brand based on an execution of the trained one or more brand personality computer models on features extracted from second crowdsource information obtained from the one or more crowdsource computing systems;
calculating, by the brand comparison engine, at least one gap value indicating a difference between at least one personality trait in the first brand personality scale and a corresponding at least one personality trait in the second brand personality scale; and
outputting, by the brand comparison engine, an output indicating an aspect of the at least one gap value based on the calculation, wherein calculating the at least one gap value comprises determining, based on a calculated degree of overlap between a first set of providers providing a first portion of crowdsource information associated with the first specified brand, and a second set of providers providing a second portion of crowdsource information associated with the second specified brand, whether or not the second specified brand is a competitor of the first specified brand, and wherein the output comprises an indication of whether or not the second specified brand is a competitor of the first specified brand.

2. The method of claim 1 , further comprising:
obtaining, by the brand comparison engine, from a crowdsource information source, crowdsource information, wherein the crowdsource information comprises natural language content submitted by a plurality of providers to the crowdsource information source;
identifying, by the brand comparison engine, for each of the first specified brand and the second specified brand, providers in the plurality of providers providing portions of the crowdsource information directed to the corresponding first specified brand or second specified brand to thereby generate a first set of providers associated with the first specified brand and a second set of providers associated with the second specified brand; and
calculating, by the brand comparison engine, a first global personality representation of the providers in the first set of providers and a second global personality representation of the providers in the second set of providers.

3. The method of claim 2 , wherein calculating the at least one gap value comprises calculating the at least one gap value between the first global personality representation and the second global personality representation.

4. The method of claim 1 , wherein determining whether or not the second specified brand is a competitor of the first specified brand comprises calculating a ratio of a number of providers present in the first set of providers that are also present in the second set of providers, to a total number of unique providers in the first set of providers and the second set of providers, wherein in response to the ratio having a value less than a threshold value, the second specified brand is determined to be a competitor of the first specified brand.

5. The method of claim 1 , further comprising: determining, based on the calculated degree of overlap between the first set of providers and the second set of providers, whether or not the second specified brand is a potential partner of the first specified brand; an outputting, as part of the output, an indication of whether or not the second specified brand is a potential partner of the first specified brand.

6. The method of claim 5 , wherein determining whether or not the second specified brand is a potential partner of the first specified brand comprises calculating a ratio of a number of providers present in the first set of providers that are also present in the second set of providers, to a total number of unique providers in the first set of providers and second set of providers, wherein in response to the ratio having a value equal to or greater than a threshold value, the second specified brand is determined to be a potential partner of the first specified brand.

7. The method of claim 1 , further comprising grouping a plurality of brands into a plurality of groupings based on the brands being associated with similar entities such that brands in a same group are comparable or complementary brands, wherein the first specified brand and the second specified brand are in a same identified group and are comparable or complementary brands.

8. The method of claim 7 , further comprising:
calculating, by the brand comparison engine, a degree of brand personality singularity for the first specified brand in the identified group, wherein the degree of brand personality singularity for the first specified brand in the identified group is a measure of a first variance between brand personality traits of the brand personality scale associated with the first identified brand and a second variance between the brand personality scale of the first identified brand and a brand personality scale of the second identified brand.

9. A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on a data processing system, causes the data processing system to implement a brand comparison engine that operates to:
generate, by a feature engine executing in the data processing system, one or more brand personality computer models, by performing natural language processing on natural language crowdsource information obtained from one or more crowdsource computing systems to extract features that are indicative of brand personality traits and relationships between the features and the brand personality traits;
train, by the feature engine, the one or more brand personality computer models, using a machine learning computer operation, to identify selected features as indicative of the brand personality traits, based on first crowdsource information obtained from the one or more crowdsource computing systems, and based on one or more ground truth data structures generated from one or more computer executed surveys of persons regarding the brand personality traits of a plurality of brands;
configure the brand com

Assignees

• IBM

Inventors

• Akkiraju Rama Kalyani T
• Gou liang
• Liu Haibin
• Mahmud Jalal U
• Sinha Vibha S
• Xu Anbang

Classifications

• G06F3/01

  Input arrangements or combined input and output arrangements for interaction between user and computer (G06F3/16 takes precedence) · CPC title

• G06Q30/0269

  based on user profile or attribute · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06F15/16

  Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs {(coordinating program control therefor G06F9/52; in regulating and control system G05B)} · CPC title

• G06Q10/063

  Operations research, analysis or management · CPC title