Customer feedback analyzer

• US 9,799,035 B2
• Filed: 09/26/2013
• Issued: 10/24/2017
• Est. Priority Date: 09/26/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Cama, Karl J., Herman, Norbert, Lambert, Daniel
• Primary Examiner(s): DELICH, STEPHANIE ZAGARELLA
• Application Number:
  US14/037,968
• Publication Number:
  US 20150088608A1
• Time in Patent Office: 1,489 Days
• Field of Search: None
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/01 Customer relationship services
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A method and system for analyzing customer feedback is provided. The method includes accessing a keyword and word mapping database and receiving consumer feedback data associated with a product or service. The consumer feedback data includes feedback data groups. Each group is divided into segments based on word analysis. Each segment is analyzed with respect to the keyword and thesaurus database. A score is generated for each segment and a composite score is generated for each feedback data group. Each composite score is stored.

Claims

1. A social media network stream feedback improvement method comprising:

accessing, by a computer processor of a computing system comprising a feedback analysis system, a keyword and word mapping database comprising keywords and phrases indicating positive and negative sentiments with respect to products and services;

identifying a source social media data stream via a fixed URL structure and associated hashtags;

receiving, by said computer processor, said source social media data stream comprising consumer feedback data associated with a product or service, wherein said consumer feedback data comprises feedback data groups, each of the feedback data groups being associated with a data source, wherein said consumer feedback data comprises structured data and unstructured data, and wherein said consumer feedback data comprises voice to text converted data;

mapping, by said computer processor, said structured data and unstructured data into a consumer data feedback hub;

dividing, by said computer processor based on word analysis, each said group into a plurality of segments, wherein each segment of each said plurality of segments for each said group is associated with a category of a list of categories;

analyzing, by a feedback analyzer executing a psycholinguistic scoring engine with respect to said keyword and word mapping database, each segment of each said plurality of segments for each said group, wherein said analyzing comprises;

performing a text analysis process with respect to text within each segment of each said plurality of segments for each said group; and

generating via a machine learning process, by a machine learning circuit, a preprogrammed radius around a keyword within said text, wherein said preprogrammed radius is generated based on;

a fixed number of words associated with said keyword, punctuation associated with said keyword, a context change associated with said keyword, or spacing associated with said keyword;

comparing, by social media analytics circuits, multiple Website reviews from multiple users;

generating, by said computer processor, a quantitative scoring natural language based table comprising a first plurality of rows and a first plurality of columns intersecting said first plurality of rows to define social media based attributes associated with said Website reviews with respect to quantitative scores associated with said consumer feedback data, wherein said table configures storage of said social media based attributes associated with said Website reviews with respect to said quantitative scores;

determining, by said computer processor executing said psycholinguistic scoring engine with respect to said Website reviews, emotions of said multiple users;

instantaneously generating, by said computer processor based on results of said analyzing, said comparing, and said determining said emotions, a real time weighted score for each segment of each said plurality of segments for each said group;

generating, by a clustering analyzer of said feedback analyzer, a composite score for each said group based on each said score for each segment of each said plurality of segments for each said group;

generating, by said computer processor, an opinion based table comprising a second plurality of rows and a second plurality of columns intersecting said second plurality of rows to define social media based opinions associated with said Website reviews and said emotions with respect to each said composite score, wherein said opinion based table configures storage of said social media based opinions associated with said Website reviews and said emotions with respect to each said composite score;

transmitting, by said computer processor, each said weighted score through an application program interface hardware circuit;

storing within a structured repository, each said weighted score;

bundling, by said application program interface, results of an outcome rating into a bundle;

applying, by said application program interface, said bundle as feedback to a Website;

generating, by said feedback analyzer, feedback associated with a transaction experience determining how a consumer interpreted a process of purchasing a product or service;

generating, by said feedback analyzer, feedback associated with a price specifying a value for money exchanged;

generating, by said feedback analyzer, feedback associated with a determination as to whether a product or service is rated based on current trending;

generating, by said feedback analyzer, feedback associated with a service describing feedback with respect to a type of service received after a transaction has been completed;

generating, by said feedback analyzer, feedback associated with a determination with respect to a satisfaction level for a product and a service; and

generating, by said feedback analyzer, a software solution for ingesting said feedback from a database comprising said structured data and said unstructured data being analyzed by a psycholinguistic scoring process executing models leveraging a psycholinguistic library database for generating recommended steps for improvement for interpreting reviews associated with user satisfaction.

2. The method of claim 1, further comprising:

generating, by said computer processor based on each said composite score for each said group, an analysis report ranking said product or service.

3. The method of claim 1, wherein said categories comprise a transaction experience category, a price category, a fashion category, a service category, and a delivery category.

4. The method of claim 1, wherein said context change comprises a change of nouns used with respect to said keyword.

5. The method of claim 1, wherein said structured data comprises survey data, consumer reports data, and transaction data, and wherein said unstructured data comprises product/service review data, social media data, call-in data, call center data, and service record data.

6. The method of claim 1, wherein each said score for each segment of each said plurality of segments for each said group comprises a scoring range.

7. The method of claim 1, wherein each said score for each segment of each said plurality of segments for each said group is generated based on:

an analysis of word frequency in said consumer feedback data, an analysis of phrasing in said consumer feedback data, and sentiment analysis of said consumer feedback data.

8. The method of claim 1, further comprising:

applying, by said computer processor, a weighting factor to each said score for each segment of each said plurality of segments for each said group; and

generating, by said computer processor based on said applying, a weighted score for each said score for each segment of each said plurality of segments for each said group.

9. The method of claim 1, further comprising:

displaying, by said computer processor, each said weighted score.

10. The method of claim 1, further comprising:

providing at least one support service for at least one of creating, integrating, hosting, maintaining, and deploying computer-readable code in the computing system, said code being executed by the computer processor to implement said accessing, said receiving, said dividing, said analyzing, said generating said score, and said generating said composite score.

11. A computing system comprising a feedback analysis system comprising a computer processor coupled to a computer-readable memory unit, said memory unit comprising instructions that when executed by the computer processor implements a social media network stream feedback improvement method comprising:

accessing, by said computer processor, a keyword and word mapping database comprising keywords and phrases indicating positive and negative sentiments with respect to products and services;

identifying a source social media data stream via a fixed URL structure and associated hashtags;

receiving, by said computer processor, said source social media data stream comprising consumer feedback data associated with a product or service, wherein said consumer feedback data comprises feedback data groups, each of the feedback data groups being associated with a data source, wherein said consumer feedback data comprises structured data and unstructured data, and wherein said consumer feedback data comprises voice to text converted data;

mapping, by said computer processor, said structured data and unstructured data into a consumer data feedback hub;

dividing, by said computer processor based on word analysis, each said group into a plurality of segments, wherein each segment of each said plurality of segments for each said group is associated with a category of a list of categories;

analyzing, by a feedback analyzer executing a psycholinguistic scoring engine with respect to said keyword and word mapping database, each segment of each said plurality of segments for each said group, wherein said analyzing comprises;

performing a text analysis process with respect to text within each segment of each said plurality of segments for each said group; and

generating via a machine learning process, by a machine learning circuit, a preprogrammed radius around a keyword within said text, wherein said preprogrammed radius is generated based on;

a fixed number of words associated with said keyword, punctuation associated with said keyword, a context change associated with said keyword, or spacing associated with said keyword;

comparing, by social media analytics circuits, multiple Website reviews from multiple users;

generating, by said computer processor, a quantitative scoring natural language based table comprising a first plurality of rows and a first plurality of columns intersecting said first plurality of rows to define social media based attributes associated with said Website reviews with respect to quantitative scores associated with said consumer feedback data, wherein said table configures storage of said social media based attributes associated with said Website reviews with respect to said quantitative scores;

determining, by said computer processor executing said psycholinguistic scoring engine with respect to said Website reviews, emotions of said multiple users;

instantaneously generating, by said computer processor based on results of said analyzing, said comparing, and said determining said emotions, a real time weighted score for each segment of each said plurality of segments for each said group;

generating, by a clustering analyzer of said feedback analyzer, a composite score for each said group based on each said score for each segment of each said plurality of segments for each said group;

generating, by said computer processor, an opinion based table comprising a second plurality of rows and a second plurality of columns intersecting said second plurality of rows to define social media based opinions associated with said Website reviews and said emotions with respect to each said composite score, wherein said opinion based table configures storage of said social media based opinions associated with said Website reviews and said emotions with respect to each said composite score;

transmitting, by said computer processor, each said weighted score through an application program interface hardware circuit;

storing within a structured repository, each said weighted score;

bundling, by said application program interface, results of an outcome rating into a bundle;

applying, by said application program interface, said bundle as feedback to a Website;

generating, by said feedback analyzer, feedback associated with a transaction experience determining how a consumer interpreted a process of purchasing a product or service;

generating, by said feedback analyzer, feedback associated with a price specifying a value for money exchanged;

generating, by said feedback analyzer, feedback associated with a determination as to whether a product or service is rated based on current trending;

generating, by said feedback analyzer, feedback associated with a service describing feedback with respect to a type of service received after a transaction has been completed;

generating, by said feedback analyzer, feedback associated with a determination with respect to a satisfaction level for a product and a service; and

generating, by said feedback analyzer, a software solution for ingesting said feedback from a database comprising said structured data and said unstructured data being analyzed by a psycholinguistic scoring process executing models leveraging a psycholinguistic library database for generating recommended steps for improvement for interpreting reviews associated with user satisfaction.

12. The computing system of claim 11, wherein said method further comprises:

generating, by said computer processor based on each said composite score for each said group, an analysis report ranking said product or service.

13. The computing system of claim 11, wherein said categories comprise a transaction experience category, a price category, a fashion category, a service category, and a delivery category.

14. The computing system of claim 11, wherein said context change comprises a change of nouns used with respect to said keyword.

15. A computer program product for analysis, the computer program product comprising:

one or more computer-readable, non-transitory tangible storage devices;

program instructions, stored on at least one of the one or more storage devices, to access a keyword and word mapping database comprising keywords and phrases indicating positive and negative sentiments with respect to products and services;

program instructions, stored on at least one of the one or more storage devices to identify a source social media data stream via a fixed URL structure and associated hashtags;

program instructions, stored on at least one of the one or more storage devices, to receive said source social media data stream comprising consumer feedback data associated with a product or service, wherein said consumer feedback data comprises feedback data groups, each of the feedback data groups being associated with a data source, wherein said consumer feedback data comprises structured data and unstructured data, and wherein said consumer feedback data comprises voice to text converted data;

program instructions, stored on at least one of the one or more storage devices, to map said structured data and unstructured data into a-consumer data feedback hub;

program instructions, stored on at least one of the one or more storage devices, to divide based on word analysis, each said group into a plurality of segments, wherein each segment of each said plurality of segments for each said group is associated with a category of a list of categories;

program instructions, stored on at least one of the one or more storage devices, to analyze by a feedback analyzer, executing a psycholinguistic scoring engine with respect to said keyword and word mapping database, each segment of each said plurality of segments for each said group by;

performing a text analysis process with respect to text within each segment of each said plurality of segments for each said group; and

generating via a machine learning process, by a machine learning circuit, a preprogrammed radius around a keyword within said text, wherein said preprogrammed radius is generated based on;

a fixed number of words associated with said keyword, punctuation associated with said keyword, a context change associated with said keyword, or spacing associated with said keyword;

program instructions, stored on at least one of the one or more storage devices, to compare by social media analytics circuits, multiple Website reviews from multiple users;

program instructions, stored on at least one of the one or more storage devices, to generate a quantitative scoring natural language based table comprising a first plurality of rows and a first plurality of columns intersecting said first plurality of rows to define social media based attributes associated with said Website reviews with respect to quantitative scores associated with said consumer feedback data, wherein said table configures storage of said social media based attributes associated with said Website reviews with respect to said quantitative scores;

program instructions, stored on at least one of the one or more storage devices, to determine by said computer processor executing said psycholinguistic scoring engine with respect to said Website reviews, emotions of said multiple users;

program instructions, stored on at least one of the one or more storage devices, to instantaneously generate, based on results of the analysis, the comparison, and determining the emotions, a weighted score for each segment of each said plurality of segments for each said group;

program instructions, stored on at least one of the one or more storage devices, to generate, by a clustering analyzer of said feedback analyzer, a composite score for each said group based on each said score for each segment of each said plurality of segments for each said group;

program instructions, stored on at least one of the one or more storage devices, to generate an opinion based table comprising a second plurality of rows and a second plurality of columns intersecting said second plurality of rows to define social media based opinions associated with said Website reviews and said emotions with respect to each said composite score, wherein said opinion based table configures storage of said social media based opinions associated with said Website reviews and said emotions with respect to each said composite score;

program instructions, stored on at least one of the one or more storage devices, to transmit each said weighted score through an application program interface;

program instructions, stored on at least one of the one or more storage devices, to store within a structured repository, each said weighted score;

program instructions, stored on at least one of the one or more storage devices to bundle by said application program interface, results of an outcome rating into a bundle;

program instructions, stored on at least one of the one or more storage devices to apply, by said application program interface, said bundle as feedback to a Website;

program instructions, stored on at least one of the one or more storage devices to generate, by said feedback analyzer, feedback associated with a transaction experience determining how a consumer interpreted a process of purchasing a product or service;

program instructions, stored on at least one of the one or more storage devices to generate by said feedback analyzer, feedback associated with a price specifying a value for money exchanged;

program instructions, stored on at least one of the one or more storage devices to generate, by said feedback analyzer, feedback associated with a determination as to whether a product or service is rated based on current trending;

program instructions, stored on at least one of the one or more storage devices to generate, by said feedback analyzer, feedback associated with a service describing feedback with respect to a type of service received after a transaction has been completed;

program instructions, stored on at least one of the one or more storage devices to generate, by said feedback analyzer, feedback associated with a determination with respect to a satisfaction level for a product and a service; and

program instructions, stored on at least one of the one or more storage devices to generate, by said feedback analyzer, a software solution for ingesting said feedback from a database comprising said structured data and said unstructured data being analyzed by a psycholinguistic scoring process executing models leveraging a psycholinguistic library database for generating recommended steps for improvement for interpreting reviews associated with user satisfaction.