Guided natural language interface system and method

• US 7,027,975 B1
• Filed: 08/08/2000
• Issued: 04/11/2006
• Est. Priority Date: 08/08/2000
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Object Services and Consulting, Inc.
• Inventors: Thompson, Craig, Pazandak, Paul N.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Storm, Donald L.
• Application Number:
  US09/634,108
• Time in Patent Office: 2,072 Days
• Field of Search: 704/9, 704/257, 704/270.1, 707/3
• US Class Current: 704/9
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A method for a guided natural language interface includes inputting to a thin client a query, communicating to an interface intermediary, communicating to an interface descriptor data source, generating an interface descriptor, communicating the interface descriptor to the interface intermediary, communicating the interface descriptor to a parser farm, requesting an appropriate parser corresponding to the interface descriptor, assigning an appropriate parser, parsing, communicating a translation from the step of parsing, to the interface intermediary, and communicating the translation to the thin client. The thin client can be geographically remote from any or all of the steps other than the step of inputting, such that the method is performed over a disparate enterprise, such as a network, for example, the Internet.

Claims

1. A method of a natural language interface, comprising the steps of:

inputting to a thin client a query component for communicating via the natural language interface;

requesting an appropriate parser for the query component for obtaining the natural language interface;

assigning an appropriate parser;

parsing the query component;

predicting a next query component based on prior query components; and

returning a parsed translation.

2. The method of claim 1, wherein the steps of inputting, parsing, and predicting are repeated until a well formed natural language grammatical constituent is completed.

3. A method of a natural language interface, comprising the steps of:

inputting to a thin client a query component for obtaining the natural language interface;

communicating to an interface intermediary;

communicating to an interface descriptor data source for enabling the natural language interface;

generating an interface descriptor;

communicating the interface descriptor to the interface intermediary;

communicating the interface descriptor to a parser farm for enabling the thin client to communicate via the natural language interface;

requesting an appropriate parser corresponding to the interface descriptor for obtaining the natural language interface;

assigning an appropriate parser;

parsing;

communicating a translation from the step of parsing, to the interface intermediary; and

communicating the translation to the thin client.

4. The method of claim 3, further comprising repeating all of the steps of inputting, communicating to an interface intermediary, communicating to an interface descriptor data source, generating, communicating the interface descriptor to the interface intermediary, communicating the interface descriptor to a parser farm, parsing, communicating a translation from the step of parsing, and communicating the translation to the thin client, all until a full sentence is translated.

5. The method of claim 4, further comprising the steps of:

delivering the translation to a command intermediary; and

processing a target application using the translation.

6. The method of claim 4, wherein the steps of delivering and processing each follow the step of repeating.

7. The method of claim 3, wherein the step of inputting is performed geographically remote from the steps of assigning and parsing.

8. The method of claim 3, wherein the step of inputting and the step of assigning maintain correspondence through communications over a network.

9. The method of claim 3, further comprising the step of predicting at least one of a next allowable query component based on prior steps of inputting and parsing;

listing the at least one of the next allowable query component; and

selecting the at least one of the next allowable query component from the list for a next step of inputting.

10. The method of claim 9, further comprising the step of repeating, until a full well formed sentence is translated, the steps of:

predicting at least one of a next allowable query component based on prior steps of inputting and parsing, listing the next allowable query component, and selecting the next allowable query component from the list for a next step of inputting, inputting, communicating to an interface intermediary, communicating to an interface descriptor data source, generating, communicating the interface descriptor to the interface intermediary, communicating the interface descriptor to a parser farm, parsing, communicating a translation from the step of parsing, and communicating the translation to the thin client.

11. The method of claim 3, wherein the steps of parsing and predicting occur at a first geographic location;

and the steps of listing and selecting occur at a second geographic location.

12. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

an interface intermediary communicatively connected to the thin client;

an interface descriptor data source for enabling the natural language interface, communicatively connected to the interface intermediary;

a command intermediary communicatively connected to the interface intermediary;

a parser farm for enabling the thin client to communicate via the natural language interface, communicatively connected to the interface intermediary; and

a parser for obtaining the natural language interface, communicatively connected to the parser farm and the interface intermediary;

wherein the thin client includes a user interface for providing a display selected from the group consisting of;

a translation status indicator for at least one of the elements of the system;

a cascaded menu display with subitems, a sequence of links;

at least one of a next allowed query constituent;

a guide for input of at least one of the query constituent;

a result of a speech recognizer;

a drop down menu;

an outline menu;

a filtered list of the at least one of the next allowed query constituent;

a grouped list of the at least one of the next allowed query constituent showing subcategories;

an active list of the at least one of the next allowed query constituent based on previous partial query constituent input; and

an explicit call to the system from the target application;

a remote element from the thin client, wherein the thin client receives from the remote element the at least one of the next allowed query constituent and the remote element is the speech recognizer.

13. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

an interface intermediary communicatively connected to the thin client;

an interface descriptor data source for enabling the natural language interface, communicatively connected to the interface intermediary;

a command intermediary communicatively connected to the interface intermediary;

a parser farm for enabling the thin client to communicate via the natural language interface, communicatively connected to the interface intermediary; and

a parser for obtaining the natural language interface, communicatively connected to the parser farm and the interface intermediary;

wherein the thin client includes a user interface for providing a display selected from the group consisting of;

a translation status indicator for at least one of the elements of the system;

a cascaded menu display with subitems;

a sequence of links;

at least one of a next allowed query constituent;

a guide for input of at least one of the query constituent;

a result of a speech recognizer;

a drop down menu;

an outline menu;

a filtered list of the at least one of the next allowed query constituent;

a grouped list of the at least one of the next allowed query constituent showing subcategories;

an active list of the at least one of the next allowed query constituent based on previous partial query constituent input; and

an explicit call to the system from the target application;

a remote element from the thin client, wherein the thin client receives from the remote element the at least one of the next allowed query constituent and the remote element is the speech recognizer;

wherein the thin client and the interface intermediary are each included in a client device communicatively connected to the network;

wherein the client device downloads the thin client and the interface intermediary over the network.

14. A method, comprising the steps of:

receiving an interface descriptor for specifying a type of a parser and initialization of the parser;

initializing the parser;

determining at least one of a next allowed query constituent;

communicating the at least one of the next allowed query constituents;

receiving a communicated selection corresponding to the step of communicating;

inputting the selection to the parser; and

repeating the steps of determining, communicating, receiving and inputting.

15. The method of claim 14, further comprising the steps of:

translating the at least one of the next allowed query constituent;

executing the result from the step of translating; and

returning the next result from the step of executing to the step of determining.

16. The method of claim 14, wherein the parser is included in at least one parser farm.

17. A method of operation of at least one parser and a parser farm, comprising the steps of:

receiving a request by the parser farm for a select one of the at least one parser;

assigning the select one of the at least one parser by the parser farm;

initializing the select one of the at least one parser by the parser farm, via an interface descriptor input to the select one of the at least one parser; and

communicating by the select one of the at least one parser of an identifier of the select one of the at least one parser, to the parser farm.

18. A system for a natural language interface comprising:

a thin client for communicating via the natural language interface;

a parser farm for enabling the client device to communicate via the natural language interface;

a parser for obtaining the natural language interface;

a first communications network for communicatively connecting the thin client and the parser farm;

a second communications network for communicatively connecting the parser to another element selected from the group consisting of the parser farm and the thin client;

wherein the parser is selected from the group consisting of;

NLMenu grammar parser, flat grammer parser, tree grammar parser, and directory structure parser.

19. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

a parser farm for enabling the client device to communicate via the natural language interface;

a parser for obtaining the natural language interface;

a first communications network for communicatively connecting the thin client and the parser farm;

a second communications network for communicatively connecting the parser to another element selected from the group consisting of the parser farm and the thin client;

wherein an operational characteristic of the parser is selected from the group consisting of;

a stateless mode, a stateful mode, single-threaded, multi-threaded, and semi-stateful mode.

20. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

a parser farm for enabling the client device to communicate via the natural language interface;

a parser for obtaining the natural language interface;

a first communications network for communicatively connecting the thin client and the parser farm;

a second communications network for communicatively connecting the parser to another element selected from the group consisting of the parser farm and the thin client;

wherein the thin client includes the parser.

21. The system of claim 20, wherein the parser is downloaded over the communication network by the thin client.

22. A system for a natural language interface comprising:

a thin client for communicating via the natural language interface;

an interface intermediary;

a first communications network connecting the thin client and the interface intermediary;

an interface descriptor data source for enabling the natural language interface;

a second communications network connecting the interface descriptor data source and the interface intermediary;

a command intermediary;

a third communications network connecting the command intermediary and the interface intermediary;

a parser farm for enabling the client device to communicate via the natural language interface;

a fourth communications network connecting the parser farm and the interface intermediary;

a parser for obtaining the natural language interface;

a fifth communications network communicatively connecting the parser to the parser farm; and

a sixth communications network for communicatively connecting the parser and the interface intermediary;

wherein the interface descriptor data source communicates an interface descriptor for defining an interface and providing a parameter of the system, selected from the group consisting of;

property, grammar, lexicon, portable specification, expert, and translation.

23. The system of claim 22, wherein the system includes more than one natural language interface, wherein the more than one natural language interface are composed by merging more than one portable specification to create a combined portable specification.

24. The system of claim 22, wherein constraints in a grammar are employed to turn-on or turn-off at least one collection of grammar rules for reasons selected from the group consisting of:

supporting training on subsets of the grammar;

supporting controlling the complexity of the grammar;

reflecting security access control privileges;

reflecting roles of users; and

reflecting personalization and user profiles.

25. A method for using a portable specification by a natural language interface system, comprising the steps of:

extracting a schema information from a data source;

translating the schema information to an encoded format of structural information as the portable specification; and

directly obtaining a natural language interface corresponding to the portable specification;

wherein the step of generating includes accessing an ODBC database management systems (DBMS) to obtain an associativity for use by a remote device to enable access to an associated database of the system.

26. A system for a natural language interface comprising:

a thin client for communicating via the natural language interface;

an interface intermediary;

a first communications network connecting the thin client and the interface intermediary;

an interface descriptor data source for enabling the natural language interface

a second communications network connecting the interface descriptor data source and the interface intermediary

a command intermediary;

a third communications network connecting the command intermediary and the interface intermediary;

a parser farm for enabling the client device to communicate via the natural language interface;

a fourth communications network connecting the parser farm and the interface intermediary;

a parser for obtaining the natural language interface;

a fifth communications network communicatively connecting the parser to the parser farm; and

a sixth communications network for communicatively connecting the parser and the interface intermediary;

wherein the interface descriptor data source communicates an interface descriptor for defining an interface and providing a parameter of the system selected from the group consisting of;

property, grammar, lexicon, portable specification, expert, and translation;

wherein the parameter of the system is selected from the group consisting of;

at least one type of grammar, at least one grammar component, and at least one type of grammar that is an at least one grammar component.

27. The system of claim 26, further comprising a combiner, wherein the combiner combines more than one of the parameters.

28. The system of claim 26, wherein a first of the at least one grammar includes a first non-terminal and the first non-terminal referencing a second non-terminal of a second of the at least one grammar.

29. The system of claim 28, wherein a characteristic of the referencing is selected from the group consisting of:

Include meaning to read rules from the second of the at least one grammar into the first of the at least one grammar during initialization of the parser; and

External meaning that during derivation of a sentence, if an external referencing symbol of the first of the at least one grammar is encountered, then a new parser is dynamically assigned and initialized with the second of the at least one grammar.

30. The system of claim 29, wherein the characteristic of the referencing is selected from the group consisting of:

Import and Export;

wherein the Import and Export are each defined for more than one database management system to enable any combined interface system to support all more than one database management system in Join operations in which one argument relation of a join is exported from a first one of the database management system and imported to a second one of the database management system at which the join is executed to provide as a result a cross-database management system join.

31. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

an interface intermediary;

a first communications network connecting the thin client and the interface intermediary;

an interface descriptor data source for enabling the natural language interface;

a second communications network connecting the interface descriptor data source and the interface intermediary;

a command intermediary;

a third communications network connecting the command intermediary and the interface intermediary;

a parser farm for enabling the client device to communicate via the natural language interface;

a fourth communications network connecting the parser farm and the interface intermediary;

a parser for obtaining the natural language interface;

a fifth communications network communicatively connecting the parser to the parser farm; and

a sixth communications network for communicatively connecting the parser and the interface intermediary;

wherein the interface descriptor data source communicates an interface descriptor for defining an interface and providing a parameter of the system, selected from the group consisting of;

property, grammar, lexicon, portable specification, expert, and translation;

comprising more than one parser, wherein at least a select one of the more than one parser is assigned and initialized via the interface descriptor.

32. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

an interface intermediary;

a first communications network connecting the thin client and the interface intermediary;

an interface descriptor data source for enabling the natural language interface;

a second communications network connecting the interface descriptor data source and the interface intermediary;

a command intermediary;

a third communications network connecting the command intermediary and the interface intermediary;

a parser farm for enabling the client device to communicate via the natural language interface;

a fourth communications network connecting the parser farm and the interface intermediary;

a parser for obtaining the natural language interface;

a fifth communications network communicatively connecting the parser to the parser farm; and

a sixth communications network for communicatively connecting the parser and the interface intermediary;

wherein the interface descriptor data source communicates an interface descriptor for defining an interface and providing a parameter of the system, selected from the group consisting of;

property, grammar, lexicon, portable specification, expert, and translation;

comprising more than one parser, wherein at least a select one of the more than one parser is assigned and initialized via the interface descriptor;

a merger, communicatively connected to the interface descriptor data source, the thin client and each parser assigned and initialized via the interface descriptor;

wherein each parser assigned and initialized via the interface descriptor communicates to the merger a selection of next allowed query components and the merger merges the selection from each parser and communicates a resulting list of the merged selection from each parser to the thin client.

33. A system for a natural language interface, comprising:

a thin client for communicating via the natural language interface;

an interface intermediary;

a first communications network connecting the thin client and the interface intermediary;

an interface descriptor data source for enabling the natural language interface;

a second communications network connecting the interface descriptor data source and the interface intermediary;

a command intermediary;

a third communications network connecting the command intermediary and the interface intermediary;

a parser farm for enabling the client device to communicate via the natural language interface;

a fourth communications network connecting the parser farm and the interface intermediary;

a parser for obtaining the natural language interface;

a fifth communications network communicatively connecting the parser to the parser farm; and

a sixth communications network for communicatively connecting the parser and the interface intermediary;

wherein the interface descriptor data source communicates an interface descriptor for defining an interface and providing a parameter of the system, selected from the group consisting of;

property grammar, lexicon, portable specification, expert, and translation;

comprising more than one parser, wherein at least a select one of the more than one parser is assigned and initialized via the interface descriptor;

a merger, communicatively connected to the interface descriptor data source, the thin client and each parser assigned and initialized via the interface descriptor;

wherein each parser assigned and initialized via the interface descriptor communicates to the merger a selection of next allowed query components and the merger merges the selection from each parser and communicates a resulting list of the merged selection from each parser to the thin client;

a list of active parsers;

wherein each parser assigned and initialized via the interface descriptor is eliminated from the list of active parsers if the parser can not parse a next allowed query component selected from the merged selection.