|
|
|
| United States Patent | 5983214 |
| Link to this page | http://www.wikipatents.com/5983214.html |
| Inventor(s) | Lang; Andrew K. (Pittsburgh, PA), Kosak; Donald M. (Pittsburgh, PA) |
| Abstract | An information entity rating system includes a content subsystem having a
structured data sub-subsystem and an unstructured data sub-subsystem. The
content subsystem receives content-based profile data for an information
entity and separately processes structured and unstructured data to
combine content-based profile data for an individual system user with the
content-based profile data for the information entity to determine
computed rating functions indicating structured and unstructured
content-based value of the information entity to the user. A collaboration
subsystem receives collaborative input data for the information entity and
for processes the collaborative input data to determine at least one
computed collaborative rating function indicating a collaboration-based
value of the information entity to the user. A correlation subsystem
receives data from the content subsystem and from the collaboration
subsystem to determine exceptions to the computed rating functions on the
basis of comparisons of data included in the content-based and
collaboration data and to generate an exception data value function
indicating an opposing value to at least one of the content-based and
collaboration values. An output system combines the structured
content-based. unstructured content-based, and collaboration-based value
functions, and the exception data value function in generating an output
rating predictor of the informon for consideration by the user. |
|
|
 |
Title Information  |
|
|
|
|
|
Drawing from US Patent 5983214 |
|
|
System and method employing individual user content-based data and user
collaborative feedback data to evaluate the content of an information
entity in a large information communication network |
|
|
|
|
|
| Publication Date |
November 9, 1999 |
|
|
|
|
|
| Filing Date |
November 5, 1998 |
|
|
|
|
|
|
|
|
|
|
|
| Parent Case |
This application is a continuation of application Ser. No. 08/627,436,
filed Apr. 4, 1996, now U.S. Pat. No. 5,867,799. |
|
|
|
|
|
|
|
|
|
|
 |
|
 |
Title Information |
|
|
|
References |
|
|
| *references marked with an asterisk below are user-added references |
 |
U.S. References |
|
|
| Add a new US reference: |
| | Reference | Relevancy | Comments | Reference | Relevancy | Comments | 5867799
Lang
707/1
Feb,1999 |  Your vote accepted
[0 after 0 votes] | | 5649186
Ferguson
707/10
Jul,1997 | Your vote accepted
[0 after 0 votes] | | 5608447
Farry
725/116
Mar,1997 | Your vote accepted
[0 after 0 votes] | | 5563999
Yaksich
715/507
Oct,1996 | Your vote accepted
[0 after 0 votes] | | 5563998
Yaksich
715/507
Oct,1996 | Your vote accepted
[0 after 0 votes] | | 5544049
Henderson
704/7
Aug,1996 | Your vote accepted
[0 after 0 votes] | | 5537586
Amram
707/3
Jul,1996 | Your vote accepted
[0 after 0 votes] | | 5471610
Kawaguchi
707/6
Nov,1995 | Your vote accepted
[0 after 0 votes] | | 5249262
Baule
706/56
Sep,1993 | Your vote accepted
[0 after 0 votes] | | 5117349
Tirfing
707/3
May,1992 | Your vote accepted
[0 after 0 votes] | | 5019961
Addesso
700/87
May,1991 | Your vote accepted
[0 after 0 votes] | | | | | |
|
 |
|
|
|
U.S. References |
|
|
|
Foreign References |
|
|
|
|
|
|
|
|
Foreign References |
|
|
|
Other References |
|
|
| Add a new Other reference: |
| Post related web sites and other references in this section |
| | Reference | Relevancy | Comments | Knowles, Software Agent Technoly Delivers Customized Information for BBN's PIN, Dialogue, pp. 62-63, May 1995.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Resnick et al, Open Archeticture for Collaboration Filtering of Netnews, IDS, pp. 1-12, Mar. 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Goldberg et al, Using Collaborative Filtering to Weave an Information Tapestry, IDS, pp. 61-70, Dec. 1992.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Sheth, Learning Approach to Personalized Information Filtering, IDS, pp. 1-74, Feb. 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Susan Dumais, et al. Using Latent Semantic Analysis to Improve Access to Textual Information. In Proceedings of CHI-88 Conference on Human Factors in Computing Systems, New York: ACM, 1998.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | David Evans et al. A Summary of the CLARIT Project. Technical Report, Laboratory for Computational Linguistics, Carnegie Melon University, Sep. 1991.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | G. Fisher and C. Stevens. Information Access in Complex, Poorly Structured Information Spaces. In Proceedings of CHI-91 Conference on Human Factors in Computing Systems, ACM, 1991.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Simon Haykin, Adaptive, Filter Theory. Prentice-Hall, englewood Clffs, NJ, pp. 100-380, 1986.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Simon Haykin. Neural Networks: A Comprehensive Foundation. Macmillian College Publishing Co., New York, pp. 18-589, 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Yezdi Lashkari et al. Collaborative Interface Agents. In Conference of the American Association for Artificial Intelligence. Seattle, WA, Aug. 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Paul Resnick, et al. Grouplens: An Open Architecture for Collaborative Filtering of Netnews. In Proceeding of ACM 1994 Conference on Computer Supported Cooperative Work. pp. 175-186, 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Anil Rewari, et al. Al Research and Applications In Digital's Service Organization. Al Magazine: pp. 68-69, 1992.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | J. Rissanen. Modeling by Shortest Data Description, Automatica, 14:465-471, 1978.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Gerard Salton. developments in Automatic Text Retrieval Science, 253:974-980, Aug. 1991.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | C. E. Shannon. A Mathematical Theory of Communication. Bell Sys. Tech. Journal, 27:379-423, 1948.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Beerud Sheth. A Learning Approach to Personalized Information Filtering, Master's Thesis, Massachusetts Institute of Technology, Feb. 1994.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | F. Mosteller, et al. Applied Bayesian and Classical Inference: The Case of the Federalist Papers. Springer-Verlag, New York, pp. 65-66, 1984.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | T. W. Yan et al. Index Structures for Selective Dissemination of Information. Technical Report STAN-CS-92-1454, Stanford University, 1992.
.
Nov,2006 | Your vote accepted
[0 after 0 votes] | | Yiming Yang. An Example-Based Mapping Mehod for Text Categorization and Retrieval. ACM Transactions on Information Systems. Vol. 12, 3, pp. 252-277, Jul. 1994..
Nov,2006 | Your vote accepted
[0 after 0 votes] | | |
|
|
|
|
|
|
Other References |
|
|
|
|
|
|
|
References |
|
|
|
|
|
|
| Market Size |
|
Estimate the gross annual revenues of the relevant market
sector:
|
| | |
| |
|
|
| Market Share |
|
Estimate the percentage of the relevant market sector this invention will capture:
|
| | |
| |
|
|
| Reasonable Royalty |
|
What percentage of gross sales should the inventor or assignee be paid?
|
| | |
| |
|
|
|
Public's "Guesstimation" of Royalty Value
|
| Market Size | N/A | [No votes] | | x | Market Share | N/A | [No votes] | | x | Reasonable Royalty | N/A | [No votes] |
| | N/A | |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Market Review |
|
|
|
Technical Review |
|
|
|
Claims |
|
|
What is claimed is:
1. An information entity rating system comprising:
a content subsystem for receiving content-based profile data for an information entity and for combining content-based profile data for an individual system user with the content-based profile data for the information entity to determine at least
one computed rating function indicating a content-based value of the information entity to the user;
a collaboration subsystem for receiving collaborative input data for the information entity and for processing the collaborative input data to determine at least one computed collaborative rating function indicating a collaboration-based value of
the information entity to the user; and
an output subsystem for combining the content-based and collaboration-based value functions to generate an output rating predictor of the informon for consideration by the user.
2. The system of claim 1 wherein:
the content subsystem includes a structured feature sub-subsystem for receiving structured data of the content-based profile data for the information entity and for combining structured data of the content-based profile data with the structured
data of the information entity to determine a computed structured-data rating function indicating a structured content value of the information entity to the user;
the content subsystem further includes an unstructured feature sub-subsystem for receiving unstructured data of the content-based profile data for the information entity and for combining unstructured data of the content-based profile data with
the unstructured data of the information entity to determine a computed unstructured-data rating function indicating an unstructured content value of the information entity to the user; and
the output system combines the structured content-based, unstructured content-based, and collaboration-based value functions in generating the output rating predictor.
3. The system of claim 1 wherein:
a correlation subsystem receives data from the content subsystem and from the collaboration subsystem to determine exceptions to the computed rating functions on the basis of comparisons of data included in the content-based and collaboration
data and to generate an exception data value function indicating an opposing value to at least one of the content-based and collaboration values; and
the output system further combines the exception data value function in generating the output rating predictor.
4. The system of claim 2 wherein the correlation subsystem receives structured and unstructured data from the structured feature and unstructured feature sub-subsystems and determines exceptions using data including the structured and
unstructured data.
5. The system of claim 3 wherein the content subsystem includes a structured feature sub-subsystem for receiving structured data of the content-based profile data for the information entity and for combining structured data of the content-based
profile data with the structured data of the information entity to determine a computed structured-data rating function indicating a structured content value of the information entity to the user;
the content subsystem further includes an unstructured feature sub-subsystem for receiving unstructured data of the content-based profile data for the information entity and for combining unstructured data of the content-based profile data with
the unstructured data of the information entity to determine a computed unstructured-data rating function indicating an unstructured content value of the information entity to the user,
the correlation subsystem further receives structured and unstructured data from the structured feature and unstructured feature sub-subsystems and determines exceptions using data including the structured and unstructured data; and
the output system combines the structured content-based, unstructured content-based collaboration-based, and exception data value functions in generating the output rating predictor.
6. The system of claim 1 wherein the content subsystem and the collaboration subsystem employ respective learning functions in computing value functions.
7. The system of claim 1 wherein:
the content subsystem determines at least one independent rating predictor and at least one uncertainty predictor from which the content value function is determined; and
the collaboration subsystem determines at least one independent rating predictor and at least one uncertainty predictor from which the collaboration value function is determined.
8. The system of claim 7 wherein each value function is determined by dividing the associated independent rating predictor by the associated uncertainty predictor.
9. The system of claim 3 wherein:
the content subsystem determines at least one independent rating predictor and at least one uncertainty predictor from which the content value function is determined;
the collaboration subsystem determines at least one independent rating predictor and at least one uncertainty predictor from which the collaboration value function is determined; and
the correlation subsystem determines at east one independent rating predictor and at least one uncertainty predictor from which the exception data value function is determined.
10. The system of claim 1 wherein the output system employs a certainty weighting function in combining the content and collaboration value functions.
11. An information processing system including the information entity rating system of claim 1 wherein:
a multi-level filter structure is provided with a content-based filter containing content-based profile data which includes content-based data applicable to the individual user and compares the filter content-based profile data with profile data
representing information in a network sourced informon; and
the content-based filter determines whether the informon profile data sufficiently matches the user profile data of the content-based filter, and, if so, routing the informon to the information entity rating system to obtain a rating of the
informon for the individual user.
12. The system of claim 11 wherein:
the content subsystem includes a structured feature sub-subsystem for receiving structured data of the content profile data for the information entity and for combining structured data of the content-based profile data with the structured data of
the information entity to determine a computed structured-data rating function indicating a structured content value of the information entity to the user;
the content subsystem further includes an unstructured feature sub-subsystem for receiving unstructured data of the content profile data for the information entity and for combining unstructured data of the content-based profile data with the
unstructured data of the information entity to determine a computed unstructured-data rating function indicating an unstructured content value of the information entity to the user; and
the output system combines the structured content-based, unstructured content-based, and collaboration-based value functions in generating the output rating predictor.
13. The system of claim 11 wherein:
a correlation subsystem receives data from the content subsystem and from the collaboration subsystem to determine exceptions to the computed rating functions on the basis of comparisons of data included in the content-based and collaboration
data and to generate an exception data value function indicating an opposing value to at least one of the content-based and collaboration values; and
the output system further combines the exception data value function in generating the output rating predictor.
14. An information entity rating system comprising:
means for receiving content profile data for an information entity and for combining content-based profile data for an individual system user with the content profile data for the information entity to determine at least one computed rating
function indicating a content-based value of the information entity to the user;
means for receiving collaborative input data for the information entity and for processing the collaborative input data to determine at least one computed collaborative rating function indicating a collaboration-based value of the information
entity to the user; and
means for combining the content-based and collaboration-based value functions to generate an output rating predictor of the informon for consideration by the user.
15. The system of claim 14 wherein:
the content data receiving means includes means for receiving structured data of the content-based profile data for the information entity and for combining structured data of the content-based profile data with the structured data of the
information entity to determine a computed structured-data rating function indicating a structured content value of the information entity to the user;
the content data receiving means further includes means for receiving unstructured data of the content-based profile data for the information entity and for combining unstructured data of the content-based profile data with the unstructured data
of the information entity to determine a computed unstructured-data rating function indicating an unstructured content value of the information entity to the user; and
the combining means combines the structured content-based, unstructured content-based, and collaboration-based value functions in generating the output rating predictor.
16. The system of claim 14 wherein:
means are provided for receiving data from the content and collaborative data receiving means to determine exceptions to the computed rating functions on the basis of comparisons of data included in the content-based and collaborative data and to
generate an exception data value function indicating an opposing value to at least one of the content-based and collaboration values; and
the combining means further combines the exception data value function in generating the output rating predictor.
17. A method for operating an information entity rating system, the method steps comprising:
receiving content profile data for an information entity and combining content-based profile data for an individual system user with the content profile data for the information entity to determine at least one computed rating function indicating
a content-based value of the information entity to the user;
receiving collaborative input data for the information entity and processing the collaborative input data to determine at least one computed collaborative rating function indicating a collaboration-based value of the information entity to the
user; and
combining the content-based and collaboration-based value functions to generate an output rating predictor of the informon for consideration by the user.
18. The method of claim 17 wherein:
the content profile data receiving step includes receiving structured data of the content-based profile data for the information entity and combining structured data of the content-based profile data with the structured data of the information
entity to determine a computed structured-data rating function indicating a structured content value of the information entity to the user;
the content profile data receiving step further includes receiving unstructured data of the content-based profile data for the information entity and combining unstructured data of the content-based profile data with the unstructured data of the
information entity to determine a computed unstructured-data rating function indicating an unstructured content value of the information entity to the user; and
the combining step combining the structured content-based unstructured content-based, and collaboration-based value functions in generating the output rating predictor.
19. The method of claim 17 wherein the method steps further include:
receiving correlated portions of the content profile data and the collaborative input data to determine exceptions to the computed rating functions on the basis of comparisons of the correlated data and to generate an exception data value
function indicating an opposing value to at least one of the content-based and collaboration values, and
the combining step further combines the exception data value function in generating the output rating predictor.
20. The method of claim 17 wherein:
the content profile data receiving step includes determining at least one independent rating predictor and at least one uncertainty predictor from which the content value functionned; and
the collaborative input data receiving step includes determining at least one independent rating predictor and at least one uncertainty predictor from which the collaboration value function is determined.
21. A method for operating an information processing system including the information entity rating method claim 17 wherein the method steps further include:
operating a multi-level filter structure having a content-based filter containing content-based profile data which includes content-based data applicable to the individual user;
comparing the filter content-based profile data with profile data representing information in a network sourced informon; and
determining whether the informon profile data sufficiently matches the user profile data of the content-based filter, and, if so, routing the informon to the information entity rating system to obtain a rating of the informon for the individual
user. |
|
|
|
|
|
|
Claims |
|
|
|
Description |
|
|
BACKGROUND OF THE INVENTION
The present invention relates to information processing systems for large or massive information networks, such as the internet, and more particuarly to such information systems in which an information filter structure uses collaborative feedback
data in determining the value of a document or other information entity (informon) to a user.
In the operation of the internet, a countless number of informons are available for downloading from any of at least thousands of sites for consideration by a user at the user's location. A user typicaly connects to a portal or other web site
having a search capability, and thereafter enters a particular query, i.e., a request for informons relevant to a topic, a field of interest, etc. Thereafter, the search site typically employs a "spider" scanning system and a content-based filter in a
search engine to search the internet for informons which match the query. This process is basically a pre-search process in which matching informons are found, at the time of initiating the search for the user's query. by comparing informons in an
"informon data base" to the user's query.
The return list of matching informons can be very extensive according to the subject of the query and the breadth of the query. More specific queries typically result in shorter return lists. In some cases, the search site may also be
structured to find web sites which probably have stored informons matching the entered query.
Collaborative data can be made available to assist in informon rating when a user actually downloads an informon, considers and evaluates it, and returns data to the search site as a representation of the value of the considered informon to the
user.
In the patent application parent to this divisional application, i.e., Ser. No. 08/627,436, now U.S. Pat. No. 5,867,799, filed by the present inventors on Apr. 4, 1996, and hereby incorporated by reference, an advanced
collaborative/content-based information filter system is employed to provide superior fitering in the process of finding and rating informons which match a user's query. The information filter structure in this system integrates content-based filtering
and collaborative filtering to determine relevancy of informons received from various sites in the internet or other network. In operation, an individual user enters a query and a corresponding "wire" is established, i.e., the query is profiled in
storage on a content basis and adaptively updated over time, and informons obtained from the network are compared to the profile for relevancy and ranking. A continuously operating "spider" scans the network to find informons which are received and
processed for relevancy to the individual user's wire and for relevancy to wires established by numerous other users.
The integrated filter system compares received informons to the individual user's query profile data, combined with collaborative data, and ranks, in order of value, informons found to be relevant. The system maintains the ranked informons in a
stored list from which the individual user can select any listed informon for consideration.
As the system continues to operate the individual user's wire, the stored relevant informon list typically changes due to factors including a return of new and more relevant informons, adjustments in the user's query, feedback evaluations by the
user for considered informons, and updatings in collaborative feedback data. Received informons are similarly processed for other users' wires established in the information filter system. Thus, the integrated information filter system compares network
informons to multiple user's queries to find matching informons for various users' wires over the course of time, whereas conventional search engines initiate a search in response to an individual user's query and use content-based filtering to compare
the query to accessed network informons to find matching informons during a limited search time period.
The present invention is directed to an informon rating system in which content-based filter profile data and collaborative feedback filter data are integrated and compared to data representative of an informon being rated to determine the
relevancy and value of the informon to an individual user. This system is embodied in the multi-level, integrated collaborative/content-based filter disclosed in the parent application, and it receives informon data, which is passed downwardly through
the filter structure, and collaborative feedback data which is sent from a collaborative feedback data processsing system called a mindpool system. Another copending patent application, entitled MULTI-LEVEL MINDPOOL SYSTEM ESPECIALLY ADAPTED TO PROVIDE
COLLABORATIVE FILTER DATA FOR A LARGE-SCALE INFORMATION FILTERING SYSTEM, Serial Number (Atty. docket # LYC2), filed by the current inventors concurrenty herewith, provides further discosure and explanation of the mindpool system.
SUMMARY OF THE INVENTION
An information entity rating system comprises a content subsystem for receiving content-based profile data for an information entity and for combining content-based profile data for an individual system user with the content-based profile data
for the information entity to determine at least one computed rating function indicating a content-based value of the information entity to the user. A collaboration subsystem receives collaborative input data for the information entity and processes
the collaborative input data to determine at least one computed collaborative rating function indicating a collaboration-based value of the information entity to the user. An output subsystem combines the content-based and collaboration-based value
functions to generate an output rating predictor of the informon for consideration by the user.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is an diagrammatic representation of an embodiment of an information filtering apparatus according to the present invention.
FIG. 2 is an diagrammatic representation of another embodiment of an information filtering apparatus according to the present invention.
FIG. 3 is a flow diagram for an embodiment of an information filtering method according to the present invention.
FIG. 4 is a flow diagram for another embodiment of an information filtering method according to the present invention.
FIG. 5 is a flow diagram for yet another embodiment of an information filtering method according to the present invention.
FIG. 6 is an illustration of a three-component-input model and pro | | |
