Extracting quotes from customer reviews of collections of items

Patent metadata

Field	Value
Publication number	US-9965470-B1
Application number	US-201113097233-A
Country	US
Kind code	B1
Filing date	Apr 29, 2011
Priority date	Apr 29, 2011
Publication date	May 8, 2018
Grant date	May 8, 2018

Abstract

Technologies are described herein for extracting quotes from customer reviews for a collection of items in order to provide salient comments for a particular item and/or upsell the collection. Individual sentences contained the customer reviews regarding the collection of items are parsed into a collection of sentences. A list of topics is generated from the collection of sentences, and each sentence in is assigned to one or more of the topics. A number of the topics from the list of topics are identified as related to the particular item, and quotes are extracted from the sentences assigned to the identified topics to be displayed as customer reviews of the particular item. Similarly, a number of topics in the list of topics are identified as related to a customer satisfaction with the collection of items, and quotes are extracted from the sentences assigned to the identified topics to be displayed in conjunction with an offer to upsell the collection of items.

First claim

What is claimed is:

1. A non-transitory computer-readable storage medium having computer-executable instructions stored thereon that, when executed by a computer, cause the computer to:
receive an identifier of an audio track;
retrieve customer reviews regarding the audio track and an album containing the audio track based at least in part on the identifier of the audio track;
determine a topic associated with a plurality of sentences of the customer reviews, the topic related to customer satisfaction with the album;
determine sentiment indicators for the plurality of sentences, wherein an individual sentiment indicator of the sentiment indicators corresponds to an individual sentence of the plurality of sentences;
determine a majority sentiment based at least in part on the sentiment indicators determined for the sentences assigned to the topic;
extract a quote from the plurality of sentences based at least in part on the majority sentiment; and
cause the quote to display on a remote customer computer in conjunction with a first offer to sell the audio track and a second offer to sell the album, wherein the second offer is associated with a user interface element to enable the remote customer computer to initiate a purchase of the album.

2. The computer-readable storage medium of claim 1 , wherein extract a quote from the plurality of sentences based at least in part on the majority sentiment further comprises:
parse a first plurality of sentences from the customer reviews;
generate a list of topics for the plurality of sentences;
assign a second plurality of sentences of the first plurality of sentences to one or more topics in the list of topics;
identify one or more topics in the list of topics related to the overall customer satisfaction with the album; and
extract the quote from the second plurality of sentences related to the overall customer satisfaction with the album.

3. The computer-readable storage medium of claim 1 , wherein determine a topic associated with a plurality of sentences of the customer reviews further comprises determine the topic associated with the plurality of sentences of the customer reviews utilizing latent Dirichlet allocation.

4. The computer-readable storage medium of claim 1 , wherein extract a quote from the plurality of sentences based at least in part of the majority sentiment further comprises extract the quote from the plurality of sentences based at least in part on the majority sentiment of the plurality of sentences being positive.

5. The computer-readable storage medium of claim 1 , wherein the quote from the plurality of sentences based at least in part on the majority sentiment comprises an individual sentence from the sentences assigned to the one or more topics related to the overall customer satisfaction with the album.

6. A computer-implemented method of displaying a collection of items containing an individual item to a customer, the method comprising:
receiving an identifier of the individual item, by way of one or more computer systems;
retrieving, based at least in part on the identifier of the individual item, customer reviews regarding the individual item and the collection of items containing the individual item, by way of one or more computer systems;
parsing a first plurality of sentences from the customer reviews, by way of one or more computer systems;
generating a list of topics for the first plurality of sentences, by way of one or more computer systems;
assigning, by way of one or more computer systems, a second plurality of sentences of the first plurality of sentences to a particular topic of the list of topics, the particular topic related to an overall customer satisfaction with the collection of items;
determining, by way of one or more computer systems, sentiment indicators for the second plurality of sentences, wherein an individual sentiment indicator of the sentiment indicators corresponds to an individual sentence of the second plurality of sentences;
determining, by way of one or more computer systems, a majority sentiment for the particular topic related to the overall customer satisfaction from the sentiment indicators determined for the sentences assigned to the particular topic;
extracting, by way of one or more computer systems, one or more quotes from the second plurality of sentences assigned to the particular topic based at least in part on an attribute of the majority sentiment; and
causing the one or more quotes to display via an application of a remote customer computer, by way of one or more computer systems, along with a user interface element that enables the remote customer computer to display information about the collection of items.

7. The computer-implemented method of claim 6 , wherein extracting, by way of one or more computer systems, one or more quotes from the second plurality of sentences assigned to the particular topic based at least in part on an attribute of the majority sentiment, further comprises:
extracting the one or more quotes from the second plurality of sentences assigned to the particular topic based at least in part on the attribute being positive as indicated by a majority sentiment.

8. The computer-implemented method of claim 6 , further comprising:
determining a readability level for the second plurality of sentences; and
extracting the one or more quotes from the second plurality of sentences further comprises extracting the one or more quotes from the second plurality of sentences having a readability level greater than or equal to a minimum readability level.

9. The computer-implemented method of claim 6 , wherein generating a list of topics for the first plurality of sentences further comprises discovering the list of topics for the first plurality of sentences based at least in part on a latent Dirichlet allocation and assigning the second plurality of sentences to the particular topic further comprises assigning the sentences to the topics utilizing latent Dirichlet allocation.

10. The computer-implemented method of claim 6 , wherein retrieving, based at least in part on the identifier of the individual item, customer reviews regarding the individual item and the collection of items containing the individual item further comprises using item relationship data and the identifier to determine one or more collections of items containing the individual item.

11. The computer-implemented method of claim 10 , wherein the individual item comprises an audio track, the collection of items comprises an album, and the item relationship data comprises one or more of track to album mappings or title authority sets.

12. The computer-implemented method of claim 6 , further comprising:
storing the one or more quotes in extracted quote data; and
generating a relationship between the extracted quote data and the collection of items.

13. The computer-implemented method of claim 6 , wherein each of the one more quotes comprises an individual sentence from the sentences assigned to the particular topic related to the overall customer satisfaction with the collection of items.

14. An apparatus for selling a collection of items containing an individual item to a customer, the apparatus comprising:
at least one processor; and
a computer-readable storage medium having computer-executable instructions stored thereon which, when executed on the at least one processor, cause the apparatus to:
receive an identifier of an audio track;
retrieve customer reviews regarding the audio track and an album containing the audio track based at least in part on the identifier of the audio track;
determine a topic associat

Assignees

• Fox Eric B
• Dicker Russell A
• Dillard Logan L
• Amazon Tech Inc

Inventors

• Fox Eric B
• Dicker Russell A
• Dillard Logan L

Classifications

• G06F9/00

  Arrangements for program control, e.g. control units (program control for peripheral devices G06F13/10) · CPC title

• G06F17/30 Primary

  Physics · mapped topic

• G06F16/345 Primary

  Summarisation for human users · CPC title

• G06F16/00 Primary

  Information retrieval; Database structures therefor; File system structures therefor · CPC title