Product recommendation with product review analysis

Patent metadata

Field	Value
Publication number	US-9652799-B2
Application number	US-201414576051-A
Country	US
Kind code	B2
Filing date	Dec 18, 2014
Priority date	Dec 18, 2014
Publication date	May 16, 2017
Grant date	May 16, 2017

Abstract

Techniques for product recommendation involve receiving, from a user, a text input expressing a question asking for a product recommendation. In response to the question, an answer that identifies a product for recommendation to the user may be generated. A plurality of product reviews comprising natural language text evaluations of the product may be analyzed, involving identifying at least one passage of text in at least one of the product reviews as providing supporting evidence for the product in answer to the question. The answer and the at least one passage in the at least one product review identified as providing supporting evidence for the answer may be presented to the user in response to the text input.

First claim

What is claimed is:

1. A method comprising:
receiving, from a user, a text input expressing a question asking for recommendation of a product having a first requested characteristic and a second requested characteristic;
identifying a product having the first requested characteristic, at least in part by locating in an ontology a node corresponding to the first requested characteristic, and traversing a relationship in the ontology from the node corresponding to the first requested characteristic to a node corresponding to the product;
analyzing, using a natural language analysis component implemented via at least one processor, a plurality of product reviews comprising natural language text evaluations of the product identified from the ontology as having the first requested characteristic, including analyzing the natural language in at least one passage of text in at least one product review of the plurality of product reviews to determine whether the natural language in the at least one passage of text has a meaning indicating that the product also has the second requested characteristic;
in response to determining via the natural language analysis that the natural language in the at least one passage of text has a meaning indicating that the product also has the second requested characteristic, identifying the at least one passage of text in the at least one product review as providing supporting evidence for the product in answer to the question;
generating, in response to the question, an answer that identifies the product having the first and second requested characteristics for recommendation to the user; and
presenting to the user, in response to the text input, the answer and the at least one passage in the at least one product review identified as providing supporting evidence for the answer.

2. The method of claim 1 , wherein the at least one product review comprises an evaluation of the product made by a purchaser of the product.

3. The method of claim 1 , wherein the at least one product review comprises an evaluation of the product made based on use of the product by an author of the at least one product review.

4. The method of claim 1 , further comprising retrieving the at least one passage in the at least one product review from at least one source selected from the group consisting of: user product reviews, expert product reviews, and social media.

5. The method of claim 1 , wherein the answer comprises information about the product, selected from the group consisting of: a name of the product, a brand of the product, a model identifier of the product, an image of the product, a price for the product, and identification of a merchant offering the product for sale.

6. The method of claim 1 , wherein analyzing the plurality of product reviews further comprises:
scoring passages in at least some of the plurality of product reviews based at least in part on strength of the passages' supporting evidence for the product in answer to the question; and
selecting the at least one passage in the at least one product review for presentation to the user based at least in part on the scoring.

7. At least one computer-readable storage medium storing computer-executable instructions that, when executed, perform a method comprising:
receiving, from a user, a text input expressing a question asking for recommendation of a product having a requested characteristic;
generating, in response to the question, an answer that identifies a product having the requested characteristic for recommendation to the user;
analyzing, using natural language analysis, a plurality of product reviews comprising natural language text evaluations of the product, including analyzing the natural language in at least one passage of text in at least one product review of the plurality of product reviews to determine whether the natural language in the at least one passage of text has a meaning indicating that the product has the requested characteristic, at least in part by converting the question to one or more hypotheses and determining whether the natural language in the at least one passage of text has a meaning that entails at least one of the one or more hypotheses;
in response to determining via the natural language analysis that the natural language in the at least one passage of text has a meaning indicating that the product has the requested characteristic, identifying the at least one passage of text in the at least one product review as providing supporting evidence for the product in answer to the question; and
presenting to the user, in response to the text input, the answer and the at least one passage in the at least one product review identified as providing supporting evidence for the answer.

8. The at least one computer-readable storage medium of claim 7 , wherein the at least one product review comprises an evaluation of the product made by a purchaser of the product.

9. The at least one computer-readable storage medium of claim 7 , wherein the at least one product review comprises an evaluation of the product made based on use of the product by an author of the at least one product review.

10. The at least one computer-readable storage medium of claim 7 , wherein the method further comprises retrieving the at least one passage in the at least one product review from at least one source selected from the group consisting of: user product reviews, expert product reviews, and social media.

11. The at least one computer-readable storage medium of claim 7 , wherein the answer comprises information about the product, selected from the group consisting of: a name of the product, a brand of the product, a model identifier of the product, an image of the product, a price for the product, and identification of a merchant offering the product for sale.

12. The at least one computer-readable storage medium of claim 7 , wherein analyzing the plurality of product reviews further comprises:
scoring passages in at least some of the plurality of product reviews based at least in part on strength of the passages' supporting evidence for the product in answer to the question; and
selecting the at least one passage in the at least one product review for presentation to the user based at least in part on the scoring.

13. Apparatus comprising:
at least one processor; and
at least one storage medium storing processor-executable instructions that, when executed by the at least one processor, perform a method comprising:
receiving, from a user, a text input expressing a question asking for recommendation of a product having a requested characteristic;
generating, in response to the question, an answer that identifies a product having the requested characteristic for recommendation to the user;
analyzing, using natural language analysis, a plurality of product reviews comprising natural language text evaluations of the product, including analyzing the natural language in at least one passage of text in at least one product review of the plurality of product reviews to determine whether the natural language in the at least one passage of text has a meaning indicating that the product has the requested characteristic, at least in part by converting the question to one or more hypotheses and determining whether the natural language in the at least one passage of text has a meaning that entails at least one of the one or more hypotheses;
in response to determining via the natural language analysis that the natural language in the at least one passage of text has a meaning indicating that the product has the requested characteristic, identifying the at least one passage of text in

Assignees

• Nuance Communications Inc

Inventors

• Boston Marisa Ferrara
• Crouch Richard Stamford
• Goulet Sylvain
• Ozcan Ali Erdem
• Pieraut Francis
• Stubley Peter

Classifications

• G06Q30/0282

  Rating or review of business operators or products · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title

• G06F40/10

  Text processing (natural language analysis G06F40/20; semantic analysis G06F40/30; processing or translation of natural language G06F40/40) · CPC title

• G06F40/289

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• G06F17/2775

  Physics · mapped topic