System and method to answer a question

• US 5,884,302 A
• Filed: 01/17/1997
• Issued: 03/16/1999
• Est. Priority Date: 12/02/1996
• Status: Expired due to Term

• Current Assignee: Hanger Solutions, LLC (IP Investments Group LLC), ProfessorQ, Inc.
• Original Assignee: IpLearn LLC
• Inventors: Ho, Chi Fai
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Ho, Ruay Lian
• Application Number:
  US08/786,109
• Time in Patent Office: 788 Days
• Field of Search: 707/1-10, 707/100-104, 707/200-206, 704/1, 704/8, 704/9, 704/260, 704/2, 704/3
• US Class Current: 1/1
• CPC Class Codes: G09B 5/00 Electrically-operated educational appliances working with questions and answers G09B7/00; simulators G09B9/00; advertising or displaying in general G09F
• CPC Class Codes: G09B 7/00 Electrically-operated teaching apparatus or devices working with questions and answers mechanically operated G09B3/00; computing arrangements G06F
• CPC Class Codes: G09B 7/04 characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question, supplying a further explanation
• CPC Class Codes: Y10S 706/927 Education or instruction
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A database-processing method and system for answering a natural-language question by a computer system with a database. In one embodiment, the system analyzes the grammatical structure of the question. In the analysis, the system scans the question to extract each word in the question; then based on a pre-defined context-free grammatical structure, the question is parsed into its grammatical components using one or more grammatically rules and the database. After parsing the question, the system transforms the components into one or more instructions, using one or more semantic rules and the database. The instructions, when executed, at least access data from the database for generating the answer to the natural-language question. In another embodiment, the system compares the natural-language question with questions stored in the database; each question in the database has its corresponding answer. If there is a match, the system retrieves the answer to the corresponding matched question and presents the answer to the student.

Claims

1. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure;

a word in the question is not in the database; and

the method further comprises the steps of replacing the word with a word in the database with a minimal number of different characters from the word not in the database.

2. A database-processing method as recited in claim 1 wherein after the step of transforming, the method further comprising the step of executing, by the system, the one or more instructions to at least access data from the database for generating the answer to the natural-language question.

3. A database-processing method as recited in claim 1, wherein the pre-defined grammatical structure includes an interrogative pronoun, followed by an auxiliary verb, which is followed by a noun-phrase.

5. A database-processing method as recited in claim 1, wherein

one grammatical component is a noun;

the database includes one or more topic-related tables, where each table identifies a plurality of data related to a topic; and

the step of transforming includes the steps of;

transforming the noun into one or more instructions to select one topic-related table from the database; and

transforming the grammatical components in the question other than the component that is the noun into one or more instructions for at least identifying data in the selected table.

6. A database-processing method as recited in claim 1, wherein

one grammatical component is a noun;

and

the step of transforming includes the steps of;

identifying the algorithm corresponding to the noun; and

applying the algorithm to data selected by grammatical components in the question other than the component that is the noun.

9. A database-processing method as recited in claim 1, wherein

the database includes one or more topic-related tables;

each table has a plurality of data;

one grammatical component is an adjective;

at least one of the grammatical components in the question other than the component that is the adjective has selected one or more tables; and

the step of transforming includes the step of transforming the adjective into one or more instructions to at least identify data in the one or more selected tables.

16. A database-processing method as recited in claim 1, wherein the analyzing step further comprises the step of suggesting the user to re-enter the question based on the pre-defined structure, if the grammatical components of the question is different from the pre-defined structure.

4. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules;

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

the pre-defined grammatical structure includes an interrogative pronoun, followed by an auxiliary verb, which is followed by a noun-phrase;

the noun-phrase includes a group-of-nouns followed by a prepositional-noun-phrase;

the group-of-nouns includes a modify-article followed by an adjective followed by one or more nouns;

the prepositional-noun-phrase includes a preposition followed by a noun-phrase;

the question further includes a verb-phrase following the noun-phrase; and

the verb-phrase includes a non-auxiliary verb followed by a prepositional-noun-phrase.

7. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure;

one grammatical component is a non-auxiliary verb;

the database includes an event table containing a plurality of non-auxiliary verbs, with each verb including one or more attributes;

the step of transforming includes the steps of;

transforming said non-auxiliary verb into one or more instructions to select, from the event table, one or more non-auxiliary verbs having similar meaning with said non-auxiliary verb in the question; and

transforming the grammatical components in the question other than the component that is said non-auxiliary verb into one or more instructions for at least identifying data in the selected attributes.

8. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure;

one grammatical component is an interrogative pronoun;

the database includes one or more topic-related tables;

each table includes one or more attributes;

the grammatical components in the question other than the component that is the interrogative pronoun have selected one or more tables;

the step of transforming includes the step of transforming the pronoun into one or more instructions for at least identifying one or more attributes in the one or more selected tables.

10. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

the database includes one or more tables;

each table includes one or more attributes;

each attribute identifies a plurality of data;

one grammatical component is an adjective;

the adjective is mapped to an algorithm;

at least one of the grammatical components in the question other than the component that is the adjective has selected one or more tables; and

the step of transforming includes the steps of identifying one or more attributes in the one or more selected tables for the algorithm to operate on the data in those attributes.

11. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

the database includes one or more tables;

each table includes a plurality of data;

one grammatical component is a preposition;

the preposition is mapped to an algorithm; and

at least one of the grammatical components in the question other than the component that is the preposition has selected one or more tables for the algorithm to operate on the data in those tables.

12. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

one grammatical component is a preposition; and

the step of transforming includes the step of determining by the preposition a grammatical component being processed before another grammatical component.

13. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

an extracted word in the question has more than one grammatical meaning; and

the method further comprises the step of producing an answer for each meaning if there is one.

14. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

a word in the question is not in the database; and

the method further comprises the step of removing the word from the question to generate an answer for the remaining words in the question.

15. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

the pre-defined grammatical structure includes a non-essential grammatical component; and

if the grammatical components of the question do not include the non-essential grammatical component, the method further comprises the step of ignoring the non-essential grammatical component in generating the answer.

17. A database-processing method of answering a question by a computer system with a database that includes many questions, each with its corresponding answer, the method comprising the steps of:

comparing the question with questions in the database; and

if there is a match, retrieving the answer corresponding to the matched question in the databases;

wherein;

said question is a natural-language question;

each question includes more than one grammatical component;

each question includes essential and non-essential grammatical components;

before the step of comparing, the method further comprises the step of extracting the grammatical components from the natural-language question;

the step of comparing includes the steps of;

comparing the grammatical components of the natural-language question with the grammatical components of the questions in the database; and

comparing the essential components of the natural-language question with the essential components of the questions in the database.

18. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules;

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

if the question cannot be parsed into its grammatical components based on the pre-defined structure, then

comparing the natural-language question with questions in the database; and

if there is a match, retrieving the answer corresponding to the matched question in the database.

19. A database-processing method of answering a natural-language question by a computer system with a database, the method comprising the steps of:

analyzing, by the system, the grammatical structure of the natural-language question for parsing the question into its grammatical components based on a pre-defined context-free grammatical structure, using the database and one or more grammatical rules; and

transforming, by the system, at least one component into one or more instructions, using one or more semantic rules and the database, for generating the answer to the natural-language question;

wherein

the step of analyzing includes the steps of;

scanning the question to extract each word in the question; and

parsing the question by applying the grammatical rules and the database on each of the extracted word to identify its grammatical meaning and to match the question with the pre-defined grammatical structure; and

an extracted word in the question has more than one meaning; and

the method further comprises the step of producing an answer for each meaning if there is one.

20. A database-processing method as recited in claim 19, wherein after the step of transforming, the method further comprising the step of executing, by the system, the one or more instructions to at least access data from the database for generating the answer to the natural-language question.

21. A database-processing method as recited in claim 19, wherein

one grammatical component is a noun;

the database includes one or more topic-related tables, where each table identifies a plurality of data related to a topic; and

the step of transforming includes the steps of;

transforming the noun into one or more instructions to select one topic-related table from the database; and

transforming the grammatical components in the question other than the component that is the noun into one or more instructions for at least identifying data in the selected table.

22. A database-processing method as recited in claim 19, wherein

one grammatical component is a noun;

and

the step of transforming includes the steps of;

identifying the algorithm corresponding to the noun; and

applying the algorithm to data selected by grammatical components in the question other than the component that is the noun.

23. A database-processing method as recited in claim 19, wherein

the database includes one or more topic-related tables;

each table has a plurality of data;

one grammatical component is an adjective;

at least one of the grammatical components in the question other than the component that is the adjective has selected one or more tables; and

the step of transforming includes the step of transforming the adjective into one or more instructions to at least identify data in the one or more selected tables.

24. A database-processing method as recited in claim 19, wherein the pre-defined grammatical structure includes an interrogative pronoun, followed by an auxiliary verb, which is followed by a noun-phrase.

25. A database-processing method as recited in claim 19, wherein the analyzing step further comprises the step of suggesting the user to re-enter the question based on the pre-defined structure, if the grammatical components of the question is different from the pre-defined structure.