Dynamic visual representation of phrases

• US 8,676,795 B1
• Filed: 08/04/2011
• Issued: 03/18/2014
• Est. Priority Date: 08/04/2011
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Durgin, Cyrus J., Brezinski, Dominique I., Buneci, Emilia S., O'Reilly, Martin M., LaRue, Lane R., Kirzhner, Benjamin S., Stathakopoulos, George N.
• Primary Examiner(s): MOSER, BRUCE M
• Application Number:
  US13/198,410
• Time in Patent Office: 957 Days
• Field of Search: None
• US Class Current: 707/730
• CPC Class Codes: G06F 16/34 Browsing; Visualisation therefor
• CPC Class Codes: G06F 40/216 using statistical methods

Abstract

A plurality of phrases may be extracted from documents associated with one or more document sources. The plurality of phrases may be filtered and processed to determine a frequency in which the plurality of phrases appear in the documents and/or a number of the document sources in which each phrase appears. A weight may be assigned to each of the phrases and, based at least in part on the assigned weight, a visual representation of the plurality of phrases may be presented. The visual representation may be dynamically updated based at least in part on an updated frequency or an updated total number of document sources associated with any one of the plurality of phrases.

Claims

1. A system comprising:

one or more processors;

memory, accessible by the one or more processors; and

a phrase analysis service stored in the memory and executed on the one or more processors to;

extract a plurality of phrases from documents associated with one or more document sources;

filter the plurality of phrases to exclude one or more of the plurality of phrases that are non-representative of a topic of documents from which the one or more phrases were extracted;

analyze the filtered plurality of phrases to determine a frequency in which individual phrases of the plurality of phrases appear in the documents and a number of the one or more document sources in which the individual phrases appear;

assign, based at least in part on the analyzing, a weight to the individual phrases;

provide a visual representation of the plurality of phrases based at least in part on assigned weights; and

dynamically modify the visual representation of the plurality of phrases as additional phrases become available such that the visual representation represents a current frequency in which the individual phrases appear in the documents and the document sources.

2. The system as recited in claim 1, wherein the phrase analysis service further provides a phrase table that includes, for the individual phrases, the assigned weight, the frequency associated with the documents, and the number of document sources that the individual phrases appear.

3. The system as recited in claim 1, wherein the visual representation is dynamically modified by changing a representation of at least one of the plurality of phrases.

4. The system as recited in claim 3, wherein the representation of the at least one phrase is changed by varying a size, a color, a shape, or a font of the representation of the at least one phrase.

5. The system as recited in claim 1, wherein the document sources include human created content or social media.

6. The system as recited in claim 1, wherein the phrase analysis service:

periodically monitors the one or more document sources to determine an existence of one or more new documents; and

in response to determining that the one or more new documents are identified, extracts additional phrases from the one or more new documents.

7. The system as recited in claim 6, wherein the additional phrases include additional instances of the plurality of phrases or new phrases not included in the plurality of phrases.

8. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

extracting a plurality of phrases from documents associated with one or more document sources;

filtering the plurality of phrases to;

identify a portion of the plurality of phrases that do not relate to content or a topic of documents in which the portion of the plurality of phrases were extracted; and

remove the portion of the plurality of phrases;

analyzing the filtered plurality of phrases to determine a frequency in which the filtered plurality of phrases appear in the documents or the one or more document sources;

assigning a weight to the analyzed plurality of phrases; and

outputting, based at least in part on the assigned weight, the plurality of phrases as a visual representation.

9. The one or more non-transitory computer-readable media as recited in claim 8, wherein the filtering further comprises at least one of:

determining whether at least one word of a particular one of the plurality of phrases appears in a dictionary having a predetermined number of words;

determining a number of words included in the particular phrase;

determining whether at least one word of the particular phrase exceeds a predetermined number of characters; and

determining whether at least one word of the particular phrase includes a stop word.

10. The one or more non-transitory computer-readable media as recited in claim 9, wherein the filtering further comprises removing a particular phrase from the plurality of phrases when:

the at least one word of the particular phrase does not appear in the dictionary;

the number of words of the particular phrase exceeds a word count;

the at least one word of the particular phrase exceeds the predetermined number of characters;

or

the at least one word of the particular phrase includes a stop word.

11. The one or more non-transitory computer-readable media as recited in claim 8, wherein the analyzing further comprises determining whether the individual phrases are statistically improbable phrase with respect to at least one of the documents, a phrase of the plurality of phrases being a statistically improbable phrase in response to determining that the phrase appears in a first document of the documents at a higher frequency than in multiple other documents of the documents.

12. The one or more non-transitory computer-readable media as recited in claim 8, wherein the visual representation of the plurality of phrases indicates which of the plurality of documents are appearing most frequently in the document sources and which of the plurality of phrases appear in a greatest number of the document sources.

13. The one or more non-transitory computer-readable media as recited in claim 8, wherein the operations further comprise:

extracting additional phrases from the documents associated with the one or more document sources; and

dynamically modifying an appearance of the plurality of phrases included in the visual representation as the additional phrases are assigned a weight.

14. The one or more non-transitory computer-readable media as recited in claim 8, wherein:

the assigned weights are degraded over time;

more recently extracted phrases are weighted more heavily than existing phrases;

or

phrases that have appeared more frequently in a recent predetermined period of time are emphasized in the visual representation.

15. The one or more non-transitory computer-readable media as recited in claim 8, wherein the plurality of phrases are maintained such that the visual representation may be accessible for any particular point in time.

16. A method, comprising:

under control of one or more computer systems configured with executable instructions,

receiving a visual representation of a plurality of phrases extracted from documents associated with multiple document sources, individual phrases of the plurality phrases being individually identified in the visual representation and at least one of the plurality of phrases including two or more words; and

displaying a varying representation of the individual phrases to indicate a frequency in which the individual phrases appear in the documents and to indicate a number of the multiple document sources in which the individual phrases appear.

17. The method as recited in claim 16, further comprising modifying the varying representation of at least one of the plurality of phrases based at least in part on an updated frequency of the at least one phrase in the documents or an updated number of the multiple document sources in which the at least one phrase appears.

18. The method as recited in claim 17, wherein modifying the varying representation of at least one of the plurality of phrases includes adjusting a size, a color, a font, or an orientation of the varying representation.

19. The method as recited in claim 16, further comprising presenting, either independently or in conjunction with the visual representation, a phrase table that provides data associated with the plurality of phrases.

20. The method as recited in claim 16, wherein the varying representation of the individual phrases is based at least in part on a weight assigned to the individual phrases.

21. A method, comprising:

under control of one or more server systems configured with executable instructions,

providing a visual representation of a plurality of phrases extracted from documents associated with multiple document sources, at least one of the plurality of phrases including two or more words;

extracting additional instances of phrases included in the plurality of phrases or new phrases not included in the plurality of phrases; and

dynamically modifying, based at least in part on the extracted additional instances or the new phrases, the visual representation to indicate a number of the multiple document sources in which individual phrases of the plurality of phrases appear.

22. The method as recited in claim 21, wherein the visual representation of the plurality of phrases is based at least in part on a frequency in which the individual phrases appear in the documents or a number of the multiple document sources in which the individual phrases appear.

23. The method as recited in claim 21, further comprising:

assigning weights to the plurality of phrases or the new phrases based at least in part on a frequency in which the plurality of phrases or the new phrases occur in the documents and the number of the multiple document sources in which the plurality of phrases or the new phrases appear; and

dynamically modifying representations of the plurality of phrases or the new phrases within the visual representation based at least in part on the assigned weights.

24. The method as recited in claim 21, further comprising dynamically modifying representations of the plurality of phrases within the visual representation by changing characteristics of the representations.

25. The method as recited in claim 24, wherein dynamically modifying the representations includes modifying a size, a color, a font, or an orientation of the representations.