IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0940564
(2004-09-14)
|
등록번호 |
US-7257574
(2007-08-14)
|
발명자
/ 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
15 인용 특허 :
5 |
초록
▼
A method performed in connection with an arrangement of nodes representable as a graph, and an inverted index containing a correlation among keywords and nodes such that at least some nodes containing a given keyword are indexed to that given keyword, involves receiving a word, searching the inverte
A method performed in connection with an arrangement of nodes representable as a graph, and an inverted index containing a correlation among keywords and nodes such that at least some nodes containing a given keyword are indexed to that given keyword, involves receiving a word, searching the inverted index to determine whether the word is a keyword and, if the word is a keyword, jumping to a node identified in the inverted index as correlated to that keyword, otherwise, learning a meaning for the word based upon reaching a result node and applying at least one specified rule such that a new input containing the word can be received thereafter and the word from the new input will be treated according to the learned meaning.
대표청구항
▼
What is claimed is: 1. A method performed in connection with an arrangement of interconnected nodes representable as a graph, and an inverted index containing a correlation among keywords and nodes such that at least some nodes containing a given keyword are indexed to that given keyword, to facili
What is claimed is: 1. A method performed in connection with an arrangement of interconnected nodes representable as a graph, and an inverted index containing a correlation among keywords and nodes such that at least some nodes containing a given keyword are indexed to that given keyword, to facilitate navigation among the nodes towards a user goal, the method comprising: receiving an input from a user comprising a word; searching the inverted index to determine whether the word is a keyword and if the word is a keyword, jumping to a node identified in the inverted index as correlated to that keyword and providing a response to the user, otherwise if the word is neither a keyword nor, if a thesaurus is available for searching, a synonym of a keyword, learning a meaning for the word based upon reaching a result node and applying at least one specified rule such that a new input containing the word can be received thereafter and the word from the new input will be treated according to the meaning that was learned and thereby result in a jump to at least one node that was correlated to the meaning and provision of an appropriate response to the new input. 2. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting a least common ancestor of the result node and at least one other node as the meaning. 3. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting at least one sibling of the result node. 4. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting at least one child of the result node. 5. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting at least one descendant of the result node. 6. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting at least one ancestor of the result node. 7. The method of claim 1 wherein the applying at least one specified rule comprises at least: selecting a random node. 8. The method of one of claims 1 through 7 wherein the applying at least one specified rule comprises: selecting from among two or more rules. 9. The method of claim 8 wherein the selecting comprises: applying a selection rule to determine which of the two or more rules is a prevailing rule. 10. The method of claim 8 wherein the selecting comprises: contextually determining that all but one of the two or more rules do not apply. 11. The method of claim 8 wherein the selecting comprises: a determination made based upon one or more of: a) a rank relationship among the two or more rules, b) a relative weighting among the two or more rules, c) a frequency of use relationship among the two or more rules, or d) a hierarchical relationship among the two or more rules. 12. The method of claim 8 wherein the selecting comprises a function of the result node. 13. The method of claim 8 wherein the selecting comprises a function of at least one node other than the result node. 14. The method of claim 8 wherein the selecting comprises a function of the result node and at least one other node. 15. The method of claim 1, wherein the learning further comprises: correlating the word to the result node via at least one particular keyword. 16. The method of claim 1, wherein when the thesaurus is available and the word does not have any synonym in the thesaurus, the learning further comprises: correlating the word to the result node via at least one synonym. 17. The method of claim 1 wherein, following the learning the meaning for the word, the method further comprises: jumping to the at least one node that was correlated to the meaning based upon a receipt of the word; and responding to the new input based upon a response rule. 18. The method of claim 1, further comprising: creating the inverted index prior to the receiving the input; and establishing a ranking of at least some of the correlations among the keywords and nodes. 19. The method of claim 18, wherein the appropriate response is provided based upon the ranking. 20. A method performed in connection with an arrangement of interconnected nodes representable as a graph, and an inverted index containing a correlation among keywords and nodes such that at least some nodes containing a given keyword are indexed to that given keyword, to facilitate navigation among the nodes towards a user goal, the method comprising: receiving an input from a user containing a word; searching the inverted index to determine whether the word is a keyword and if the word is a keyword, jumping to a node identified in the inverted index as correlated to that keyword and providing a response to the user, otherwise if the word is neither a keyword nor, if a thesaurus is available for searching, a synonym of a keyword, learning a meaning for the word based upon reaching a result node and applying at least one specified rule such that a new input containing the word can be received thereafter and the word from the new input will be treated according to the meaning that was learned and thereby result in a jump to at least one node that was correlated to the meaning and provision of an appropriate response, and wherein the applying at least one specified rule comprises at least one of: a) selecting a least common ancestor of the result node and at least one other node as the meaning, b) selecting at least one sibling of the result node, c) selecting at least one child of the result node, d) selecting at least one descendant of the result node, e) selecting at least one ancestor of the result node, or f) selecting a random node. 21. The method of claim 20 wherein two or more of a) through f) comprise potentially applicable rules, the method further comprising one of: applying a selection rule to determine which of the two or more rules is a prevailing rule, contextually determining that only one of a) through f) will apply. 22. The method of claim 20 wherein only one of a) through f) applies and wherein the selecting comprises one of: a) a determination made based upon one or more of: i) a rank relationship among the two or more rules, ii) a relative weighting among the two or more rules, iii) a frequency of use relationship among the two or more rules, or iv) a hierarchical relationship among the two or more rules; b) a function of the result node; c) a function of at least one node other than the result node; d) a function of the result node and at least one other node.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.