Program-listing appendix

• US 6,353,822 B1
• Filed: 08/22/1996
• Issued: 03/05/2002
• Est. Priority Date: 08/22/1996
• Status: Expired due to Fees

• Current Assignee: Massachusetts Institute of Technology, Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Lieberman, Henry
• Primary Examiner(s): Alam, Hosain T.
• Assistant Examiner(s): Colbert, Ella
• Application Number:
  US08/701,242
• Time in Patent Office: 2,021 Days
• Field of Search: 395/601, 395/602, 395/603, 395/604, 395/605, 395/606, 395/607, 395/608, 395/609, 395/610, 395/611, 395/616, 395/704, 395/759, 395/680, 395/828, 707/1, 707/2, 707/3, 707/4, 707/5, 707/10, 707/100, 707/200, 707/704, 707/759, 707/500, 707/513, 707/501.1
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

Methods and apparatus for assisting a user in retrieving documents or other data items of interest operates in tandem with a conventional document-retrieval facility, such as a web browser, by tracking the choices made by the user in retrieving and viewing data items. The invention identifies additional items likely to be of interest to the user. Preferably, the invention operates automously, without interruption of the user'"'"'s activities, delivering (in real-time or upon request) a set of current recommendations. The recommendations take the form of (or include) links to the recommended items, and the user is free to execute any of these links to examine the contents of a recommendation.

Claims

1. Apparatus for identifying electronically encoded data items of interest to a user, the data items being stored on at least one computer and being identified by an item identifier, at least some of the data items being linked to at least one other data item by specifying the identifier of the at least one other data item, the apparatus comprising:

a. a computer memory;

b. interactive retrieval means comprising;

i. means responsive to a user-provided identifier for causing retrieval to the computer memory and display of the data item specified by the identifier; and

ii. means responsive to a user'"'"'s selection of an identifier appearing in an already-retrieved data item for causing retrieval to the computer memory of the linked data item specified by the identifier appearing in the already-retrieved data item;

c. observational means, responsive to the interactive retrieval means, for identifying user preference criteria through analysis of a user'"'"'s operation of the interactive retrieval means to retrieve data items but without interruption of said operation;

d. automated retrieval means for causing automatic retrieval to the computer memory of a plurality of new data items, each new data item being linked by a sequence of at least one identifier to a user-selected data item or a data item linked thereto;

e. means for screening the items retrieved by the automated retrieval means in accordance with the preference criteria to produce a search set; and

f. means for reporting the search set to a user.

2. The apparatus of claim 1 wherein the data items are stored on a computer network, the network comprising a plurality of network computers having data items accessible to the apparatus via a communication linkage and specified in accordance with a network protocol, each item identifier comprising a protocol identifier.

3. The apparatus of claim 2 wherein the computer network is the Internet.

4. The apparatus of claim 1 wherein the data items are web pages and data items are linked to other data items by means of hyperlinks.

5. The apparatus of claim 4 wherein the data items are web pages.

6. The apparatus of claim 1 wherein the means for identifying user preference criteria comprises means for recording user selections of data items.

7. The apparatus of claim 6 wherein the data items have contents and the means for identifying user preference criteria further comprises means for recording the contents of the user-selected data items and the linked data items.

8. The apparatus of claim 7 wherein the screening means compares the contents of the user-selected data items and the linked data items with the contents of items retrieved by the automated retrieval means.

9. The apparatus of claim 8 wherein the comparison is based on the product of term frequency and inverse document frequency.

10. The apparatus of claim 1 further comprising display means for viewing a retrieved data item, the automated retrieval means causing the retrieval of new data items without display thereof on the display means.

11. The apparatus of claim 10 wherein the automated retrieval means is operative during viewing of a displayed data item to cause automatic retrieval of new data items linked by a sequence of protocol identifiers to the displayed data item, the automatic retrieval ceasing when the data item is no longer displayed.

12. The apparatus of claim 1 further comprising:

a. means for accessing an Internet search engine;

b. means for formulating a search query from the user preference criteria; and

c. means for submitting the search query to the search engine, thereby causing the search engine to identify data items matching the search query,

the automated retrieval means being configured to retrieve data items identified by the search engine for screening by the screening means.

13. The apparatus of claim 1 wherein:

a. the items include textual material;

b. the preference criteria include (i) item-level criteria specifying a relevance level for an item and (ii) content-level criteria specifying a relevance level for textual material;

c. the item-level criteria govern retrieval of new data items by the automated retrieval means; and

d. the content-level criteria govern screening by the screening means of items retrieved by the automated retrieval means.

14. The apparatus of claim 1 wherein the new data items are related to the user-selected data items and the linked data items at a plurality of hierarchical levels, each level comprising new data items linked to a user-selected data item by an identical number of sequential links, the automated retrieval means performing a breadth-first search by causing automatic retrieval to the computer memory of items at one hierarchical level before to causing retrieval of items at a lower hierarchical level.

15. The apparatus of claim 14 wherein the breadth-first search is a best-first search, the automated retrieval means being configured to apply the preference criteria to the items at each hierarchical level to rank said items in an order, the automatic retrieval to the computer memory of items at each hierarchical level occurring in the ranked order.

16. The apparatus of claim 1 wherein automated retrieval proceeds at a user-adjustable rate.

17. A method of identifying electronically encoded data items of interest to a user, the data items being stored on at least one computer and being identified by an item identifier, at least some of the data items being linked to at least one other data item by specifying the identifier of the at least one other data item, the method comprising the steps of:

a. responding to a plurality of user-provided identifiers by causing retrieval and display of the data items specified by the user-provided identifiers;

b. responding to a user'"'"'s selection, in an already-retrieved data item, of at least one identifier appearing therein and specifying a linked data item by causing retrieval to the computer memory of the at least one linked data item;

c. identifying user preference criteria through observation of the user'"'"'s retrieval of data items and linked data items but without interruption of said retrieval;

d. causing automatic retrieval to the computer memory of a plurality of new data items, each new data item being linked by a sequence of at least one identifier to a user-selected data item or a data item linked thereto;

e. screening the items retrieved by the automated retrieval means in accordance with the preference criteria to produce a search set; and

f. reporting the search set to a user.

18. The method of claim 17 wherein the data items are stored on a computer network, the network comprising a plurality of network computers having data items accessible to the apparatus via a communication linkage and specified in accordance with a network protocol, each item identifier comprising a protocol identifier.

19. The method of claim 18 wherein the computer network is the Internet.

20. The method of claim 19 wherein the protocol identifiers specify the World Wide Web.

21. The method of claim 17 wherein the data items are web pages and data items are linked to other data items by means of hyperlinks.

22. The method of claim 17 wherein the step of identifying user preference criteria comprises analyzing user selections of data items.

23. The method of claim 22 wherein the data items have contents and the step of identifying user preference criteria further comprises analyzing the contents of user-selected data items and linked data items.

24. The method of claim 23 wherein the screening step comprises comparing the contents of user-selected data items and linked data items with the contents of the automatically retrieved new data items.

25. The method of claim 24 wherein the comparison is based on the product of term frequency and inverse document frequency.

26. The method of claim 17 further comprising the steps of:

a. formulating a search query from the user preference criteria;

b. accessing an Internet search engine; and

c. submitting the search query to the search engine, thereby causing the search engine to identify data items matching the search query;

d. retrieving data items identified by the search engine; and

e. screening the data items identified by the search engine in accordance with the preference criteria.

27. The method of claim 17 wherein:

a. the items include textual material;

b. the preference criteria include (i) item-level criteria specifying a relevance level for an item and (ii) content-level criteria specifying a relevance level for textual material;

c. the item-level criteria govern retrieval of new data items; and

d. the content-level criteria govern screening of the automatically retrieved items.

28. The method of claim 17 wherein the new data items are related to the user-selected data items and the linked data items at a plurality of hierarchical levels, each level comprising new data items linked to a user-selected item by an identical number of sequential links, and further comprising the step of performing a breadth-first search by causing automatic retrieval of items at one hierarchical level before to causing retrieval of items at a lower hierarchical level.

29. The method of claim 28 wherein the breadth-first search is a best-first search, the preference criteria being applied to the items at each hierarchical level to rank said items in an order, the automatic retrieval of items at each hierarchical level occurring in the ranked order.

30. The method of claim 17 wherein automated retrieval proceeds at a user-adjustable rate.