SYSTEMS AND METHOD OF DERIVING A SENTIMENT RELATING TO A BRAND

• US 20090119157A1
• Filed: 10/17/2008
• Published: 05/07/2009
• Est. Priority Date: 11/02/2007
• Status: Abandoned Application

• Current Assignee: KPMG LLP (KPMG International Cooperative)
• Original Assignee: Wise Window, Inc.
• Inventors: Dulepet, Rajiv
• Application Number:
  US12/253,567
• Publication Number:
  US 20090119157A1
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Methods for deriving a brand sentiment are presented. Phrases and ratings associated with the brand are stored in a database. The phrases are analyzed and compared to each other and to the ratings to derive a statistical significance of a phrase usage relative to other phrases. A sentiment score is derive from the statistical significance.

Claims

1. A method of deriving a sentiment relating to a brand, the method comprising:

gathering a set of phrases and a set of ratings associated with a first brand entity and with a second brand entity by crawling web-based documents;

storing occurrences the set of phrases and the set of rating in a database;

identifying a first correlated usage between a first phrase from the set of phrases to a first rating from the set of ratings with respect to the first brand entity;

identifying a second correlated usage between the first phrase to a second rating from the set of ratings with respect to the second brand entity;

deriving a relative statistical significance between the first and the second correlated usage;

deriving a sentiment score associated with the first entity as a function of the relative statistical significance; and

presenting the sentiment score to a researcher via a user interface.

2. The method of claim 1, wherein the step of storing occurrences includes indexing the occurrences hierarchically based on the first and the second entity.

3. The method of claim 1, further comprising establishing a many-to-many relationship between the first and the second entity and phrases.

4. The method of claim 1, further comprising establishing a many-to-many relationship between the phrases and the ratings.

5. The method of claim 1, further comprising cleansing data entries in the database as a function of a threshold associated with the relative statistical significance.

6. The method of claim 1, wherein the first entity is selected from the group consisting of a company, a person, a product, the brand, a place, and a service.

7. The method of claim 1, wherein the relative statistical significance is domain specific.

8. The method of claim 1, wherein the sentiment is domain specific.

9. The method of claim 1, wherein the step of deriving a sentiment score includes refining the sentiment score through interaction with the researcher.

10. The method of claim 1, wherein the step of deriving a sentiment score includes performing a regression analysis on the set of phrases and the set of ratings in the database.

11. The method of claim 1, wherein the first phrase is selected from the group consisting of an acronym, and an abbreviation.

12. The method of claim 1, wherein the first brand entity is represented by an image.

13. The method of claim 1, wherein the first brand entity is represented by a sound.