Method and system for programmatic analysis of consumer reviews

• US 9,317,566 B1
• Filed: 06/01/2015
• Issued: 04/19/2016
• Est. Priority Date: 06/27/2014
• Status: Active Grant

• Current Assignee: Groupon Incorporated
• Original Assignee: Groupon Incorporated
• Inventors: Karablieh, Feras, Levi, Dor, Koren, Amit, L'Huillier, Gaston, Larrain, Francisco Jose, Arroyo Garcia, Hernan Enrique, Li, Juzheng, Langdon, Daniel, Esterhazy, Jonathan, Vedanarayanan, Srinivasa Raghavan, Jeffery, Shawn, Bansal, Bhupesh
• Primary Examiner(s): Morrison, Jay
• Application Number:
  US14/727,852
• Time in Patent Office: 323 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2455 Query execution
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/0281 Customer communication at a business location, e.g. providing product or service information, consulting
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products

Abstract

Embodiments provide a computer-executable method, computer system and non-transitory computer-readable medium for programmatically analyzing a consumer review. The method includes programmatically accessing, via a network device, one or more consumer reviews for a commercial entity or a commercial object. The method also includes executing a consumer review processing engine to programmatically identify an attribute descriptor in the one or more consumer reviews, and executing the consumer review processing engine to programmatically generate a sentiment score associated with the one or more consumer reviews. The method further includes storing, on a non-transitory computer-readable storage device, the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

Claims

1. An apparatus for programmatically analyzing a consumer review, the apparatus comprising:

a network device configured to programmatically access one or more consumer reviews for a commercial entity or a commercial object;

a consumer review processing engine configured to

programmatically identify an attribute descriptor in the one or more consumer reviews, and

programmatically generate a sentiment score associated with the one or more consumer reviews, wherein programmatic generation of the sentiment score comprises;

using a natural language processing engine to programmatically parse the consumer review into a set of sentences;

using the natural language processing engine to programmatically parse each sentence in the set of sentences into a set of words;

for each word in the set of words in each sentence, programmatically generating a word sentiment score, wherein programmatic generation of a first word sentiment score for a first word in the consumer review comprises;

determining a word polarity score indicating whether a sentiment expressed in the first word is positive, negative, neutral or unknown, a word negation score indicating whether there is a negation in the context of the first word, and a word intensity score indicating an intensity or magnitude of the sentiment expressed in the first word; and

applying a machine learning algorithm to determine a relationship between the first word sentiment score and the word polarity score, the word negation score, and the word intensity score;

for each sentence in the set of sentences, programmatically generating a sentence sentiment score, the sentence sentiment score generated based on word sentiment scores associated with words in the sentence; and

programmatically generating the sentiment score by combining sentence sentiment scores associated with the set of sentences in the consumer review; and

a non-transitory computer-readable storage device configured to store the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

2. The apparatus of claim 1, wherein the network device is further configured to:

transmit one or more computer-executable instructions to a computing device, the one or more computer-executable instructions programmed to cause the computing device to visually display the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

3. The apparatus of claim 1, wherein the sentiment score is associated with the attribute descriptor.

4. The apparatus of claim 1, wherein the sentiment score is associated with the commercial entity or the commercial object but not specifically with the attribute descriptor.

5. The apparatus of claim 1, wherein programmatic generation of the first sentence sentiment score for the first sentence in the consumer review comprises:

programmatically identifying one or more emoticons in the first sentence; and

assigning one or more emoticon scores to the one or more emoticons based on sentiments expressed in the emoticons;

wherein the first sentence sentiment score is determined based on the one or more emoticon scores.

6. The apparatus of claim 1, wherein programmatic generation of the word polarity score comprises:

detecting a plurality of consecutive vowels in the first word; and

adjusting the word polarity score based on the plurality of consecutive vowels.

7. The apparatus of claim 1, wherein programmatic generation of the word intensity score comprises:

detecting a plurality of consecutive vowels in the first word; and

adjusting the word intensity score based on the plurality of consecutive vowels.

8. The apparatus of claim 1, wherein programmatic generation of the word intensity score comprises:

detecting whether the first word includes all capital letters; and

adjusting the word intensity score in response to determining that the first word includes all capital letters.

9. The apparatus of claim 1,

wherein the consumer review processing engine is further configured to:

programmatically analyze prior purchase data or profile data associated with a first consumer to generate a first attribute descriptor of interest to the first consumer;

programmatically match the first attribute descriptor to the attribute descriptor generated based on the one or more consumer reviews; and

wherein the network device is further configured to, based on the matching, transmit computer-executable instructions to a computing device associated with the first consumer to cause a visual display device associated with the computing device to display the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

10. The apparatus of claim 1, wherein the commercial entity is a merchant.

11. The apparatus of claim 1, wherein the commercial object is a product, service or promotion.

12. The apparatus of claim 1, wherein the attribute descriptor indicates a context associated with the commercial entity or the commercial object.

13. The apparatus of claim 1, wherein the attribute descriptor indicates a quality associated with the commercial entity or the commercial object.

14. The apparatus of claim 1, wherein the attribute descriptor indicates a category or sub-category associated with the commercial entity or the commercial object.

15. A computer-executed method for programmatically analyzing a consumer review, the method comprising:

programmatically accessing, via a network device, one or more consumer reviews for a commercial entity or a commercial object;

executing a consumer review processing engine to programmatically identify an attribute descriptor in the one or more consumer reviews;

executing the consumer review processing engine to programmatically generate a sentiment score associated with the one or more consumer reviews, wherein programmatically generating the sentiment score comprises;

using a natural language processing engine to programmatically parse the consumer review into a set of sentences;

using the natural language processing engine to programmatically parse each sentence in the set of sentences into a set of words;

for each word in the set of words in each sentence, programmatically generating a word sentiment score, wherein programmatic generation of a first word sentiment score for a first word in the consumer review comprises;

determining a word polarity score indicating whether a sentiment expressed in the first word is positive, negative, neutral or unknown, a word negation score indicating whether there is a negation in the context of the first word, and a word intensity score indicating an intensity or magnitude of the sentiment expressed in the first word; and

applying a machine learning algorithm to determine a relationship between the first word sentiment score and the word polarity score, the word negation score, and the word intensity score;

for each sentence in the set of sentences, programmatically generating a sentence sentiment score, the sentence sentiment score generated based on word sentiment scores associated with words in the sentence; and

programmatically generating the sentiment score by combining sentence sentiment scores associated with the set of sentences in the consumer review; and

storing, on a non-transitory computer-readable storage device, the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

16. The method of claim 15, wherein programmatic generation of the first word sentiment score for the first word in the consumer review comprises:

programmatically generating the word polarity score based on;

detecting a plurality of consecutive vowels in the first word; and

adjusting the word polarity score based on the plurality of consecutive vowels;

programmatically generating the word negation score; and

programmatically generating the word intensity score based on;

detecting a plurality of consecutive vowels in the first word or detecting whether the first word includes all capital letters; and

adjusting the word intensity score based on the plurality of consecutive vowels or in response to determining that the first word includes all capital letters.

17. A non-transitory computer-readable medium for programmatically analyzing a consumer review, the non-transitory computer-readable medium having encoded thereon one or more computer-executable instructions that, when executed, cause a computing device to:

access one or more consumer reviews for a commercial entity or a commercial object;

identify an attribute descriptor in the one or more consumer reviews;

generate a sentiment score associated with the one or more consumer reviews, wherein programmatic generation of the sentiment score comprises;

using a natural language processing engine to programmatically parse the consumer review into a set of sentences;

using the natural language processing engine to programmatically parse each sentence in the set of sentences into a set of words;

for each word in the set of words in each sentence, programmatically generating a word sentiment score, wherein programmatic generation of a first word sentiment score for a first word in the consumer review comprises;

determining a word polarity score indicating whether a sentiment expressed in the first word is positive, negative, neutral or unknown, a word negation score indicating whether there is a negation in the context of the first word, and a word intensity score indicating an intensity or magnitude of the sentiment expressed in the first word; and

applying a machine learning algorithm to determine a relationship between the first word sentiment score and the word polarity score, the word negation score, and the word intensity score;

for each sentence in the set of sentences, programmatically generating a sentence sentiment score, the sentence sentiment score generated based on word sentiment scores associated with words in the sentence; and

programmatically generating the sentiment score by combining sentence sentiment scores associated with the set of sentences in the consumer review; and

store the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.