Presenting documents to a user based on topics and collective opinions expressed in the documents

Patent metadata

Field	Value
Publication number	US-9436663-B2
Application number	US-201213717151-A
Country	US
Kind code	B2
Filing date	Dec 17, 2012
Priority date	Dec 17, 2012
Publication date	Sep 6, 2016
Grant date	Sep 6, 2016

Abstract

A content platform for presenting documents to a user based on topics and collective opinions expressed in the documents is disclosed. The content platform mines a corpus of documents to identify a set of topics and analyzes each document in the corpus of documents to determine a set of opinions associated with the set of topics. The corpus of documents is presented to the user based on the set of topics and the set of opinions. Each document in the corpus of documents is visually modified to focus the user's attention on the set of opinions associated with the set of topics.

First claim

What is claimed is:

1. A computer implemented method for presenting documents to a user based on topics and collective opinions expressed in the documents, comprising:
mining a corpus of documents to identify a set of topics;
analyzing, by a processor, each document in the corpus of documents to determine a set of opinions associated with the set of topics;
classifying each opinion in the set of opinions as a representative opinion or as an out-of-line opinion;
ranking the representative opinions and the out-of-line opinions into tiers;
presenting the corpus of documents based on the set of topics and the set of opinions; and
visually modifying a document in the corpus of documents according to the tiers of the set of opinions, including rendering text in the document associated with an opinion according to the opinion's tier.

2. The computer implemented method of claim 1 , wherein mining a corpus of documents to identify a set of topics comprises extracting a body copy of each document in the corpus of documents.

3. The computer implemented method of claim 1 , wherein mining a corpus of documents to identify a set of topics comprises ranking each word in the corpus of documents by importance across the set of topics and the corpus of documents.

4. The computer implemented method of claim 1 , wherein analyzing, by a processor, each document in the corpus of documents to determine a set of opinions associated with the set of topics comprises statistically parsing each sentence in each document to determine a set of modifiers associated with a subject in each sentence.

5. The computer implemented method of claim 4 , further comprising calculating a sentiment score for each modifier.

6. The computer-implemented method of claim 5 , wherein the set of opinions associated with the set of topics comprises a set of modifiers associated with a set of subjects.

7. The computer implemented method of claim 6 , further comprising determining a distribution of opinions across the corpus of documents.

8. A content platform system for presenting documents to a user based on topics and collective opinions expressed in the documents, comprising:
a processor; and
a non-transitory computer readable medium storing instructions that when executed by the processor cause the processor to:
mine a corpus of documents to determine a set of opinions associated with a set of topics;
classify each opinion in the set of opinions as a representative opinion or as an out-of-line opinion;
rank the representative opinions and the out-of-line opinions into tiers;
present the corpus of documents based on the set of topics and the set of opinions; and
visually modify a document in the corpus of documents according to the tiers of the set of opinions, including rendering text in the document associated with an opinion according to the opinion's tier.

9. The content platform system of claim 8 , wherein the instructions are to cause the processor to parse each sentence in each document to determine a set of modifiers associated with a subject in each sentence, the set of modifiers associated with a subject representing a plurality of opinions about the subject.

10. The content platform system of claim 8 , wherein the instructions are to cause the processor to display documents that have opinions on a clicked topic from the set of topics.

11. The content platform system of claim 8 , wherein the instructions are to cause the processor to sort documents in the corpus of documents according to the representative opinions and the out-of-line opinions.

12. The content platform system of claim 8 , wherein, to visually modify a document according to the tiers of the set of opinions, the instructions are to cause the processor to modify the document with call-outs or called-out text corresponding to representative and out-of-line opinions.

13. A non-transitory computer readable medium comprising instructions executable by a processor to:
identify a set of topics and a set of opinions associated with the set of topics in a corpus of documents;
classify each opinion in the set of opinions as a representative opinion or as an out-of-line opinion;
rank the representative opinions and the out-of-line opinions into tiers;
present the corpus of documents based on the set of topics and the set of opinions; and
visually modify the presentation of a document in the corpus of documents according to the tiers of the set of opinions, including rendering text in the document associated with an opinion according to the opinion's tier.

14. The non-transitory computer readable medium of claim 13 , wherein to present the corpus of documents based on the set of topics and the set of opinions comprises to list documents in the corpus of documents by the set of topics and sort documents in each topic in the set of topics based on the set of opinions.

15. The non-transitory computer readable medium of claim 13 , wherein to visually modify the presentation of a document in the corpus of documents, the instructions are to cause the processor to display a plurality of sparklines illustrating distributions of opinions in the set of opinions and to call out text in the document as a cursor hovers over one of the plurality of sparklines.

Assignees

• Hewlett Packard Development Co Lp
• Hewlett Packard Entpr Dev Lp

Inventors

• Hailpern Joshua
• Huberman Bernardo

Classifications

• G06F16/35

  Clustering; Classification · CPC title

• G06F40/258 Primary

  Heading extraction; Automatic titling; Numbering · CPC title

• G06F40/166 Primary

  Editing, e.g. inserting or deleting · CPC title

• G06F17/24 Primary

  Physics · mapped topic

• G06F17/2745

  Physics · mapped topic