System and method for using banded topic relevance and time for article prioritization

• US 8,452,781 B2
• Filed: 01/27/2009
• Issued: 05/28/2013
• Est. Priority Date: 01/27/2009
• Status: Expired due to Fees

• Current Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Original Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Inventors: Stefik, Mark J., Mittal, Sanjay, Good, Lance E.
• Primary Examiner(s): SYED, FARHAN M
• Application Number:
  US12/360,823
• Publication Number:
  US 20100191741A1
• Time in Patent Office: 1,582 Days
• US Class Current: 707/749
• CPC Class Codes: G06F 16/3328 using graphical result space presentation or visualisation
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing

Abstract

A system and method for using banded topic relevance and time for article prioritization is provided. Articles of digital information and at least one social index are maintained. The social index includes topics that each relate to one or more of the articles. Fine-grained topic models matched to the digital information for each topic are retrieved. The articles are succinctly classified under the topics using the fine-grained topic models. Each of the articles is relevancy scored within the topic under which the article was classified. The articles are arranged into discrete bands by relevance score. The articles are temporally sorted within the discrete bands. The articles are presented within the discrete bands.

Claims

1. A computer-implemented method for using banded topic relevance and time for article prioritization, comprising:

maintaining articles of digital information and at least one social index comprising topics that each relate to one or more of the articles;

retrieving fine-grained topic models matched to the digital information for each topic;

succinctly classifying the articles under the topics by testing the articles against the fine-grained topic models;

for each topic in the at least one social index, maintaining a coarse-grained topic model comprising;

specifying characteristic words extracted from the articles identified by the fine-grained topic models of each topic; and

assigning scores to the characteristic words;

for each of the articles within the topic under which the article was classified, accumulating the scores of the characteristic words into a relevancy score for that article;

selecting only the articles within the topic having relevancy scores satisfying a minimum threshold relevancy;

for each topic in the at least one social index, arranging the selected articles into discrete bands with each discrete band representing comparable levels of relevancy, comprising;

setting a range of relevancy scores for each discrete band; and

classifying each of the articles within the topic into the discrete band within whose range of relevancy scores the article'"'"'s relevancy score falls;

temporally sorting the articles within each of the discrete bands; and

presenting the temporally-sorted articles within the discrete bands.

2. A method according to claim 1, further comprising:

accepting a time period, which defines a temporal cut-off; and

excluding the articles within the discrete bands falling outside the time period accepted.

3. A method according to claim 2, wherein the time periods comprise one or more of 24 hours, two days, one week, one month, and unlimited.

4. A method according to claim 1, further comprising:

defining a normalized scale and dividing the normalized scale over the discrete bands;

normalizing the relevancy scores over the normalized scale; and

grouping the normalized relevancy scores into the discrete bands.

5. A method according to claim 4, wherein the discrete bands, as normalized, comprise, in order of decreasing relevancy, one or more groups of the articles labeled as centrally-relevant, relevant, somewhat relevant, and not relevant.

6. A method according to claim 1, wherein each article includes each of content and one of a publication date and indication of recency.

7. A method according to claim 1, further comprising:

maintaining a plurality of the social indexes;

evaluating each of the social indexes based on the topics presented within each social index; and

presenting the articles within the discrete bands within the context of each of the social indexes.

8. A computer-implemented system for using banded topic relevance and time for article prioritization, comprising:

an electronic database, comprising;

articles of digital information and at least one social index comprising topics that each relate to one or more of the articles maintained for social indexing; and

fine-grained topic models matched to the digital information for each topic;

a processor and memory within which code for execution by the processor is stored, further comprising;

a fine-grained classifier module succinctly classifying the articles by testing the articles against the topics using the fine-grained topic models;

a coarse-grained classifier module configured to maintain, for each topic in the at least one social index, a coarse-grained topic model, comprising;

an extraction module configured to specify characteristic words extracted from the articles identified by the fine-grained topic models of each topic; and

a scoring module configured to assign scores to the characteristic words;

a relevancy scoring module configured to accumulate, for each of the articles within the topic under which the article was classified, the scores of the characteristic words into a relevancy score for that article;

a selection module configured to select only the articles having relevancy scores satisfying a minimum threshold relevancy;

a band module configured to arrange, for each topic in the at least one social index, the selected articles into discrete bands with each discrete band representing comparable levels of relevancy, comprising;

a range of relevancy scores for each discrete band; and

a discrete band classifier configured to classify each of the articles within the topic into the discrete band within whose range of relevancy scores the article'"'"'s relevancy score falls; and

a temporal module temporally sorting the articles within each of the discrete bands; and

a user interface visually presenting the temporally-sorted articles within the discrete bands.

9. A system according to claim 8, wherein the processor and memory further comprise:

a time control comprised in the user interface accepting a time period, which defines a temporal cut-off,

wherein the articles within the discrete bands falling outside the time period accepted are excluded.

10. A system according to claim 8, wherein the time periods comprise one or more of 24 hours, two days, one week, one month, and unlimited.

11. A system according to claim 8, wherein the processor and memory further comprise:

an electronically-stored normalized scale, wherein the normalized scale is divided over the discrete bands; and

a normalizing module normalizing the relevancy scores over the normalized scale, and grouping the normalized relevancy scores into the discrete bands.

12. A system according to claim 11, wherein the discrete bands, as normalized, comprise, in order of decreasing relevancy, one or more of groups the articles labeled as centrally-relevant, relevant, somewhat relevant, and not relevant.

13. A system according to claim 8, wherein each article includes each of content and one of a publication date and indication of recency.

14. A system according to claim 8, wherein the database further comprises:

a plurality of the social indexes; and

the processor and memory further comprise;

a multi-index evaluation module evaluating each of the social indexes based on the topics presented within each social index, wherein the articles within the discrete bands are presented within in the user interface the context of each of the social indexes.