Search engine with fill-the-blanks capability
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06F-007/00
G06F-017/30
출원번호
UP-0114971
(2005-04-25)
등록번호
US-7693829
(2010-05-20)
발명자
/ 주소
Alshawi, Hiyan
출원인 / 주소
Google Inc.
대리인 / 주소
Morgan, Lewis & Bockius LLP
인용정보
피인용 횟수 :
13인용 특허 :
2
초록▼
A method of searching for information is described. A sequence of terms, including one or more term segments and one or more identifiers corresponding to one or more missing terms, is received. The sequence of terms is converted into a corresponding search pattern, including a set of one or more que
A method of searching for information is described. A sequence of terms, including one or more term segments and one or more identifiers corresponding to one or more missing terms, is received. The sequence of terms is converted into a corresponding search pattern, including a set of one or more query expressions and one or more ordering constraints. The search pattern is compared with a plurality of documents to identify a set of documents. Match scores for one or more matches between the search pattern and documents in the set of documents are determined. Content in the set of documents corresponding to the one or more missing terms in the search pattern are identified and a ranked set of information items containing the identified content is provided in accordance with the match scores.
대표청구항▼
What is claimed is: 1. A computer-implemented method of searching for information, comprising: at a server system: receiving from a client system a fill-the-blank query comprising one or more term segments and one or more missing term identifiers signifying missing information sought by a user; con
What is claimed is: 1. A computer-implemented method of searching for information, comprising: at a server system: receiving from a client system a fill-the-blank query comprising one or more term segments and one or more missing term identifiers signifying missing information sought by a user; converting the fill-the-blank query into a corresponding search pattern, wherein the search pattern includes: one or more missing content identifiers corresponding to the one or more missing term identifiers, and a set of one or more query expressions corresponding to each term segment; identifying a set of documents matching the search pattern; identifying content in the set of documents corresponding to the search pattern, the identified content including one or more potential answers corresponding to the one or more missing term identifiers; and responding to the query by providing to the client system at least one of the one or more potential answers. 2. The method of claim 1, including determining match scores for one or more matches between the search pattern and documents in the set of documents; wherein responding to the query includes providing a ranked set of information items containing the identified content in accordance with the match scores. 3. The method of claim 2, wherein providing at least one of the one or more potential answers includes providing a ranked list of documents containing the identified content in accordance with the match scores. 4. The method of claim 2, further comprising determining document scores for the set of documents, wherein a respective document score is based on a highest match score and a total number of matches between the search pattern and the respective document in the set of documents. 5. The method of claim 4, wherein the respective document score is further based on a quality of document metric. 6. The method of claim 4, wherein the respective document score is a weighted summation of the highest match score, the total number of matches between the search pattern and the respective document, and a quality of document metric. 7. The method of claim 4, further comprising determining, for the identified content in the respective document, a relative frequency of the identified content in the set of documents; and determining final scores for the set of documents, wherein a respective final score for a respective document in the set of documents is based on the respective document score for the respective document and the relative frequency for the identified content in the respective document. 8. The method of claim 7, wherein the respective final score for a respective document in the set of documents is a weighted product of the relative frequency for the identified content in the respective document and the respective document score for the respective document. 9. The method of claim 2, wherein a respective match score for a respective document in the set of documents is based on a uniqueness metric for the identified content in the respective document and a uniqueness metric for padding between the one or more query expression in the search pattern. 10. The method of claim 9, wherein the respective match score is further based on weights of one or more synonyms used in converting one or more term segments in the sequence of terms into one or more query expressions. 11. The method of claim 10, wherein the respective match score is a weighted summation of the uniqueness metric for the identified content in the respective document, the uniqueness metric for padding between the one or more query expression in the search pattern and the weights of the one or more synonyms. 12. The method of claim 1, further comprising receiving one or more context terms that identify a category for the one or more potential answers and using the context terms for further processing selected from the group consisting of determining rankings of the potential answers, including the one or more context of terms in the search pattern, and filtering the identified set of documents. 13. The method of claim 1, wherein the converting one or more of the term segments into respective query expressions includes mapping one or more term segments into one or more paraphrases. 14. The method of claim 1, further comprising specifying one or more ordering constraints wherein the ordering constraints specify one or more re-orderings of the term segments in the sequence of terms. 15. The method of claim 1, wherein the identifying content includes limiting a length of the content in accordance with a uniqueness metric of candidate terms for the content. 16. The method of claim 15, wherein the uniqueness metric is an inverse of a number of documents in the set of documents in which a respective candidate term appears. 17. The method of claim 1, wherein a length of the identified content is less than a pre-determined value. 18. The method of claim 1, further comprising providing for display snippets of text containing the one or more potential answers, wherein the one or more potential answers are highlighted. 19. The method of claim 1, wherein the fill-the-blank query includes a respective missing term identifier located between two respective term segments. 20. A computer-program product, comprising: a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising: instructions for receiving a fill-the-blank query comprising one or more term segments and one or more missing term identifiers signifying missing information sought by a user; instructions for converting the fill-the-blank query into a corresponding search pattern, wherein the search pattern includes: one or more missing content identifiers corresponding to the one or more missing term identifiers, and a set of one or more query expressions corresponding to each term segment; instructions for identifying a set of documents matching the search pattern; instructions for identifying content in the set of documents corresponding to the search pattern, the identified content including one or more potential answers corresponding to the one or more missing term identifiers; and instructions for responding to the query by providing at least one of the one or more potential answers. 21. The computer program product of claim 20, including instructions for determining match scores for one or more matches between the search pattern and documents in the set of documents; wherein instructions for responding to the query include instructions for providing a ranked set of information items containing the identified content in accordance with the match scores. 22. The computer program product of claim 21, wherein providing at least one of the one or more potential answers includes providing a ranked list of documents containing the identified content in accordance with the match scores. 23. The computer program product of claim 20, further comprising instructions for receiving one or more context terms that identify a category for the potential answers and instructions for using the context terms for further processing selected from the group consisting of determining rankings of the potential answers, including the one or more context of terms in the search pattern, and filtering the identified set of documents. 24. The computer program product of claim 20, further comprising instructions for providing for display snippets of text containing the one or more potential answers wherein the one or more potential answers are highlighted. 25. The computer program product of claim 20, wherein the fill-the-blank query includes a respective missing term identifier is located between two respective term segments. 26. A computer system, comprising: memory; one or more processors; and one or more programs stored in the memory and executed by the one or more processors, the one or more programs including: instructions for receiving from a client system a fill-the-blank query comprising one or more term segments and one or more missing term identifiers signifying missing information sought by a user; instructions for converting the fill-the-blank query into a corresponding search pattern, wherein the search pattern includes: one or more missing content identifiers corresponding to the one or more missing term identifiers, and a set of one or more query expressions corresponding to each term segment; instructions for identifying a set of documents matching the search pattern; instructions for identifying content in the set of documents corresponding to the search pattern, the identified content including one or more potential answers corresponding to the one or more missing term identifiers; and instructions for responding to the query by providing to the client system at least one of the one or more potential answers. 27. The system of claim 26, including instructions for determining match scores for one or more matches between the search pattern and documents in the set of documents; wherein instructions for responding to the query include instructions for providing a ranked set of information items containing the identified content in accordance with the match scores. 28. The system of claim 27, wherein providing at least one of the one or more potential answers includes providing a ranked list of documents containing the identified content in accordance with the match scores. 29. The system of claim 26, further comprising instructions for receiving one or more context terms that identify a category for the one or more potential answers and instructions for using the context terms for further processing selected from the group consisting of determining rankings of the potential answers, including the one or more context of terms in the search pattern, and filtering the identified set of documents. 30. The system of claim 26, further comprising instructions for providing for display snippets of text containing the one or more potential answers wherein the one or more potential answers are highlighted. 31. The computer system of claim 26, wherein the fill-the-blank query includes a respective missing term identifier is located between two respective term segments.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (2)
Woods William A., Method and apparatus for generating query responses in a computer-based document retrieval system.
Kreiner Barrett Morris ; Litwin Edward Paul, Method and system for processing a memory map to provide listing information representing data within a database.
Roshen, Waseem A., Prioritizing search for non-exact matching service description in service oriented architecture (SOA) service registry system with advanced search capability.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.