Knowledge discovery agent system and method

• US 8,352,388 B2
• Filed: 09/05/2011
• Issued: 01/08/2013
• Est. Priority Date: 05/22/2002
• Status: Expired due to Term

• Current Assignee: Digital Reasoning Systems, Inc (Smarsh Incorporated)
• Original Assignee: Digital Reasoning Systems, Inc (Smarsh Incorporated)
• Inventors: Estes, Timothy W.
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): BROWN JR, NATHAN H
• Application Number:
  US13/225,546
• Publication Number:
  US 20120059776A1
• Time in Patent Office: 491 Days
• Field of Search: 706/52
• US Class Current: 706/12
• CPC Class Codes: G06F 16/2425 Iterative querying; Query formulation based on the results of a preceding query
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: H04N 21/251 Learning process for intelligent management, e.g. learning user preferences for recommending movies details of learning user preferences for the retrieval of video data in a video database G06F16/739; computer systems using learning methods G06N3/08
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61
• CPC Class Codes: H04N 21/4662 characterized by learning algorithms
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/4826 using recommendation lists, e.g. of programs or channels sorted out according to their score

Abstract

A system and method for processing information in unstructured or structured form, comprising a computer running in a distributed network with one or more data agents. Associations of natural language artifacts may be learned from natural language artifacts in unstructured data sources, and semantic and syntactic relationships may be learned in structured data sources, using grouping based on a criteria of shared features that are dynamically determined without the use of a priori classifications, by employing conditional probability constraints.

Claims

1. A personal search agent system, comprising:

a user agent comprising a processor or microprocessor, wherein the processor or microprocessor is programmed to;

receive a query statement from a user;

format the query statement;

forward the query statement to a server-based search manager agent;

receive the response to the query statement; and

display the response to the user;

a search manager agent comprising a processor or microprocessor, wherein the processor or microprocessor is programmed to;

receive a query statement from the user agent;

assign search tasks to one or more search engine task agents;

collect search results;

validate search results;

order search results into a response; and

forward the response to the user agent;

a user miner agent comprising a processor or microprocessor, wherein the processor or microprocessor is programmed to;

observe the user'"'"'s interaction with the user agent; and

learn the behavior and preferences of the user; and

a data mining agent comprising a processor or microprocessor, wherein the processor or microprocessor is programmed to;

analyze the responses to the user'"'"'s queries; and

build a conceptual network of the knowledge contained in the response.

2. The system of claim 1, wherein the data mining agent further learns associations of natural language artifacts in unstructured data sources, wherein said artifacts include at least one of words, phrases, subjects, predicates, modifiers, and other syntactic forms.

3. The system of claim 2, further wherein learned associations resulting from said learning associations of natural language artifacts are formed using grouping of one natural language artifact in an interaction window with another at least one natural language artifact in said interaction window, based on a criteria of shared features of one or more sets from said grouping constituting measurements from said data sources.

4. The system of claim 3, further wherein said criteria of shared features are dynamically determined without the use of a priori classifications and using conditional probability constraints between sets of learned associations.

5. The system of claim 4, further wherein said grouping creates a network of conditional probabilities between all encountered natural language artifacts or a subset thereof, determined by consideration of conditional interaction probabilities based on a history of measurements from said data sources.

6. The system of claim 2, further wherein learned associations are represented in a specific format compatible for operations for mapping between a plurality of data structures and languages comprising one or more of arrays, vector spaces, first order predicate logic, Conceptual Graphs, SQL, and typed programming languages.

7. The system of claim 6, further wherein said typed program languages comprise one or more of Java, C++, and other conventional programming languages.

8. The system of claim 2, further wherein hierarchies of association are constructed across a state space of term usage compatible for interpolation of mapping functions between sets of terms.

9. The system of claim 8, further wherein said mapping functions include one or more of fuzzy-type, weighted-type, or other types of mapping functions.

10. The system of claim 2, further where a structure of mapping functions is generated, said structure comprising a formal semantic structure of one or more of programming languages, modal logics, frame systems, and ontologies of objects and relationships.