Method and apparatus for assembling a set of documents related to a triggering item

Patent metadata

Field	Value
Publication number	US-9904681-B2
Application number	US-201113182245-A
Country	US
Kind code	B2
Filing date	Jul 13, 2011
Priority date	Jan 12, 2009
Publication date	Feb 27, 2018
Grant date	Feb 27, 2018

Abstract

The present invention relates to a method and apparatus for assembling a set of documents related to a triggering item. One embodiment of a method for assembling a set of electronic documents related to an electronic triggering item detected by a computing device being operated by a user includes automatically extracting by the computing device a set of features from the triggering item, without receiving a request by the user to assemble the set of electronic documents, and assembling as the set of electronic documents a plurality of documents that is relevant to the set of features, wherein the plurality of documents is retrieved from a plurality of different types of electronic sources.

First claim

What is claimed is:

1. A computer-implemented method for assembling a set of electronic documents related to an electronic triggering item, the method comprising:
using a machine-learning based classifier, classifying electronic documents in a set of harvested documents as being associated with at least one activity type, the at least one activity type describing an event involving human activity in the physical world;
when the triggering item is received, identifying an activity type based on the triggering item;
using the machine learning-based classifier, selecting a template that corresponds to the activity type that is based on the triggering item, wherein the selected template is selected from a set of templates that are associated with different activity types and the templates in the set of templates include lists of different types of documents that are associated with the different activity types;
using the selected template, ranking the documents in the set of harvested documents based on how closely the documents in the set of harvested documents match the activity type that is based on the triggering item, wherein the ranking comprises increasing the ranking of a document when an activity type of the document matches the activity type associated with the template and decreasing the ranking of a document when an activity type of the document does not match the activity type associated with the template;
using the ranking, outputting the set of electronic documents relevant to the triggering item.

2. The method of claim 1 , wherein the triggering item comprises an electronic reference to at least one user activity, and the set of features relates to the at least one user activity.

3. The method of claim 2 , wherein the triggering item comprises an electronic mail message.

4. The method of claim 2 , wherein the triggering item comprises an electronic file.

5. The method of claim 2 , wherein the triggering item comprises an entry in a calendar application.

6. The method of claim 2 , wherein the at least one user activity comprises a trip.

7. The method of claim 2 , wherein the at least one user activity comprises a meeting.

8. The method of claim 7 , wherein the meeting comprises an employment interview.

9. The method of claim 2 , wherein the at least one user activity comprises a proposal activity.

10. The method of claim 2 , wherein the at least one user activity comprises a review of a technical paper.

11. The method of claim 1 , comprising extracting a set of features from the triggering item, wherein the set of features comprises at least one of: a keyword, a person, a date, a location, or an acronym.

12. The method of claim 1 , comprising extracting a set of features from the triggering item, wherein the set of features defines an activity type associated with the triggering item.

13. The method of claim 1 , comprising:
extracting a set of features from the triggering item
building a query in accordance with the set of features;
running a search using the query over a plurality of available documents from the plurality of different types of electronic sources.

14. The method of claim 13 , wherein the query comprises a weight associated with at least one feature in the set of features.

15. The method of claim 13 , wherein the running produces a list of documents ranked according to relevance to the set of features and document frequency.

16. The method of claim 13 , further comprising:
automatically harvesting the plurality of available documents using a classifier.

17. The method of claim 1 , wherein the plurality of documents is retrieved from a plurality of different types of electronic sources.

18. The method of claim 17 , wherein the plurality of different types of electronic sources includes a local storage device of the computing device.

19. The method of claim 1 , further comprising: re-ranking the set of documents.

20. The method of claim 1 , further comprising: outputting the plurality of documents to an input/output device.

21. The method of claim 20 , further comprising:
receiving feedback relating to at least one document in the set of harvested documents.

22. The method of claim 21 , further comprising:
adjusting the ranking in accordance with the feedback.

23. The method of claim 22 , wherein the adjusting comprises updating a global dictionary used to guide the ranking.

24. The method of claim 22 , wherein the adjusting comprises associating the set of harvested documents with a signature of the triggering item.

25. The method of claim 22 , wherein the adjusting comprises adjusting an algorithm or model used to assign a score to the set of harvested documents.

26. The method of claim 1 , wherein the plurality of documents includes at least one personal document.

27. The method of claim 1 , wherein the triggering item is a single item detected by the computing device.

28. The method of claim 1 , comprising, when information relevant to the triggering item is not included in the set of electronic documents, generating an alert and outputting the alert with the outputting of the set of electronic documents.

29. A non-transitory computer readable medium containing an executable program for assembling a set of electronic documents related to an electronic triggering item, where the program performs steps comprising:
using a machine-learning based classifier, classifying electronic documents in a set of harvested documents as being associated with at least one activity type, the at least one activity type describing an event involving human activity in the physical world;
when the triggering item is received, identifying an activity type based on the triggering item;
using the machine learning-based classifier, selecting a template that corresponds to the activity type that is based on the triggering item, wherein the selected template is selected from a set of templates that are associated with different activity types and the templates in the set of templates include lists of different types of documents that are associated with the different activity types;
using the selected template, ranking the documents in the set of harvested documents based on how closely the documents in the set of harvested documents match the activity type that is based on the triggering item, wherein the ranking comprises increasing the ranking of a document when an activity type of the document matches the activity type associated with the template and decreasing the ranking of a document when an activity type of the document does not match the activity type associated with the template;
using the ranking, outputting the set of electronic documents relevant to the triggering item.

30. A system for assembling a set of electronic documents related to an electronic triggering item, the system comprising at least one processor for:
classifying electronic documents in a set of harvested documents as being associated with at least one activity type, the at least one activity type describing an event involving human activity in the physical world;
when the triggering item is received, identifying an activity type based on the triggering item;
using the machine learning-based classifier, selecting a template that corresponds to the activity type that is based on the triggering item, wherein the select

Assignees

• Nitz Kenneth
• Dunkley David
• Donneau Golencer Thierry
• Cheyer Adam
• Pound Leslie
• Hardt Stephen L
• Stanford Res Inst Int

Inventors

• Nitz Kenneth
• Dunkley David
• Donneau Golencer Thierry
• Cheyer Adam
• Pound Leslie
• Hardt Stephen L

Classifications

• G06F16/35

  Clustering; Classification · CPC title

• G06F16/338

  Presentation of query results · CPC title

• G06F16/335

  Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536) · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06Q10/10 Primary

  Office automation; Time management · CPC title