Selectively delivering an article

• US 8,954,440 B1
• Filed: 04/09/2010
• Issued: 02/10/2015
• Est. Priority Date: 04/09/2010
• Status: Active Grant

• Current Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Original Assignee: Wal-Mart Stores Texas LLC (WalMart Inc.)
• Inventors: Gattani, Abhishek, Rajaraman, Anand
• Primary Examiner(s): Perveen, Rehana
• Assistant Examiner(s): Hershley, Mark E
• Application Number:
  US12/757,915
• Time in Patent Office: 1,768 Days
• Field of Search: 707/732, 707/738, 707/740
• US Class Current: 707/738
• CPC Class Codes: G06F 16/93 Document management systems

Abstract

Selectively delivering an article is disclosed. A user preference is received. A document is also received. An entity pair comprising a concept included in a concept taxonomy and a textual representation included in the document is determined. The document is categorized based at least in part on the concept. A user associated with the user preference is selectively notified of the document.

Claims

1. A system for selectively delivering an article, comprising:

a communications interface configured to;

receive a user preference; and

receive a document;

a processor configured to;

identify a plurality of entity pairs each comprising a concept included in a concept taxonomy and a textual representation relating to the concept included in the document, each concept having a corresponding category vector including a plurality of adjacent nodes of the concept in a taxonomy;

compute a document vector (dv) for the document as a sum of the category vectors for the plurality of entity pairs;

select a subset of concepts from the plurality of entity pairs according to a comparison of the document vector and the category vectors of the concepts of the plurality of entity pairs by;

calculating a similarity score ds for each concept according to ds=dv·

cv, where dv is the document vector and cv is a concept vector;

selecting the subset of concepts as concepts having similarity scores ds with respect to the document vector hi her than a threshold;

categorize the document based at least in part on the selected subset of concepts; and

determine that the selected subset of concepts corresponds to the user preference;

in response to determining that the selected subset of concepts corresponds to the user preference, notify a user associated with the user preference of the document; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1 wherein the user preference includes a preference for a top level node in the concept taxonomy.

3. The system of claim 1 wherein the user preference includes a preference for a leaf node concept in the concept taxonomy.

4. The system of claim 1 wherein the user preference is not included in the taxonomy and the processor is configured to determine at least one concept in the taxonomy associated with the preference.

5. The system of claim 1 wherein the document is a news article.

6. The system of claim 1 wherein the document is a forum post.

7. The system of claim 1 wherein the document is a status update.

8. The system of claim 1 wherein the processor is further configured to include the document in a group with at least one other document based at least in part on the processor performing a clustering action.

9. The system of claim 1 wherein the processor is further configured to analyze the document and automatically generate a summarization of the document according to the analysis.

10. The system of claim 1, further comprising pruning concepts according to ambiguity resolution further comprises comparing similarity scores ds for ambiguous entity pairs of the plurality of entity pairs.

11. A method for selectively delivering an article, comprising:

receiving a user preference;

receiving a document;

identifying a plurality of textual representations included in the document;

determining a plurality of entity pairs each comprising a concept included in a concept taxonomy and a textual representation of the plurality of textual representations each concept having a corresponding category vector including nodes adjacent to the concept in a taxonomy;

computing a document vector (dv) for the document as a sum of the category vectors for the plurality of entity pairs

selecting a subset of concepts from the plurality of entity pairs according to a comparison of a document vector to the category vectors of the plurality of concepts, the document vector being a combination of the category vectors for the plurality of concepts by;

calculating a similarity score ds for each concept according to ds=dv·

cv, where dv is the document vector and cv is a concept vector;

selecting the subset of concepts as concepts having similarity scores ds espect to the document vector higher than a threshold;

categorizing the document with the subset of concepts; and

;

determining that the selected subset of concepts corresponds to the user preference;

in response to determining that the selected subset of concepts corresponds to the user preference, notifying a user associated with the user preference of the document.

12. The method of claim 11, further comprising pruning concepts according to ambiguity resolution by comparing similarity scores for ambiguous entity pairs.

13. A computer program product for selectively delivering an article, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a user preference;

receiving a document;

identifying a plurality of textual representations included in the document;

determining a plurality of entity pairs each comprising a concept included in a concept taxonomy and a textual representation of the plurality of textual representations, each concept having a corresponding category vector including nodes adjacent to the concept in a taxonomy;

computing a document vector (dv) for the document as a sum of the category vectors for the plurality of entity pairs

selecting a subset of concepts from the plurality of concepts of the entity pairs according to a comparison of a document vector to the category vectors of the plurality of concepts by;

calculating a similarity score ds for each concept according to ds=dv·

cv, where dv is the document vector and cv is a concept vector;

selecting the subset of concepts as concepts having similarity scores ds with respect to the document vector higher than a threshold,

categorizing the document based at least in part on the subset of concepts;

determine that the selected subset of concepts corresponds to the user preference; and

in response once to determining that selected subset of concepts corresponds to the user preference, notifying a user associated with the user preference of the document.

14. The computer program product of claim 13, further comprising pruning concepts according to ambiguity resolution by comparing similarity scores corresponding to ambiguous entity pairs.