CN102289482A - Unstructured data query method - Google Patents
Unstructured data query method Download PDFInfo
- Publication number
- CN102289482A CN102289482A CN2011102202447A CN201110220244A CN102289482A CN 102289482 A CN102289482 A CN 102289482A CN 2011102202447 A CN2011102202447 A CN 2011102202447A CN 201110220244 A CN201110220244 A CN 201110220244A CN 102289482 A CN102289482 A CN 102289482A
- Authority
- CN
- China
- Prior art keywords
- statement
- query
- aql
- data
- inquiry
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an unstructured data query method which comprises the following steps of: (1) improving the conventional extensive makeup language (XML) query method and extending an XQuery language by using an unstructured advanced query language (AQL); (2) simplifying a function irrelevant with the unstructured data query; (3) defining an execution plan; (4) resolving an AQL sentence and identifying word fields of an example file; (5) separating description word fields of different attributes in the sentence; and (6) forming a query sentence according to a result in the step (5) and generating the execution plan. Relevance search and example query for the basic attribute, the semantic attribute and the bottom characteristic are realized according to four faceted constraints of data to be queried, advanced query is realized according to a characteristic constraint during data clustering, and relevance search for different types of data is realized by constraints of relevant points during cross-type search, so that a query requirement of the unstructured data can be met completely.
Description
Technical field
The present invention relates to a kind of at the unstructured data storehouse, non-structural data enquiry method based on tetrahedron model and XQuery language, this method is used unstructured data intelligence inquire language AQL (Advanced Query Language, AQL, advanced inquiry language).
Background technology
Relational database can not satisfy the application demand of internet because of its inherent limitations, arises at the historic moment in the unstructured data storehouse.The unstructured data storehouse is a theoretical foundation with the tetrahedron model, and the tetrahedron model comes complete data of description with base attribute, semantic attribute, low-level image feature and four facets of source document.Wherein base attribute and semantic attribute can be stored with the XML database, and existing XML query language is XQuery, but XQuery can only be used for the information of Query XML form, can not realize the content-based inquiries of multimedia messages such as video, audio frequency.
The data class of storing in the unstructured data storehouse is many and in large scale, query manipulation demand for these data also is various, have three kinds specifically: basic query, correlation inquiry and intelligence inquire, existing querying method can't be finished novel query manipulation.
Summary of the invention
Technology of the present invention is dealt with problems: overcome the deficiencies in the prior art, a kind of method that realizes non-structural data enquiry is provided, realize multiaspect inquiry and Query By Example according to constraint to four facets of the data that will inquire about, the constraint of feature selecting realizes intelligence inquire during according to data clusters, by the constraint of relating dot being realized many bodies inquiries, can satisfy the demand of non-structural data enquiry operation fully to striding data class when retrieval.
Technical solution of the present invention: a kind of non-structural data enquiry method, this method use unstructured data intelligence inquire language AQL, it is characterized in that step is as follows:
(1) the unstructured data intelligence inquire language AQL of Shi Yonging expands the XQuery language, increases base attribute, semantic attribute, low-level image feature and cluster mode key sentence, to support multiaspect inquiry, Query By Example, intelligence inquire and the inquiry of many bodies;
(2) simplify and the irrelevant function of non-structural data enquiry, to improve analyzing efficiency;
(3) the definition executive plan is used to represent the implementation of non-structural data enquiry;
(4) resolving of AQL statement is: pre-service, and identification instance document field if statement comprises then log file of instance document, otherwise forwards step (5) to;
(5) according to the AQL grammer description field of different attribute in the statement is separated;
(6) different field of separating according to step (5) forms corresponding query statement and generates executive plan.
According to a further aspect of the invention, wherein step (1) further comprises step:
(1.a) AQL has added the statement FilepathClause that is used to specify the instance document path based on the FLWOR statement of XQuery language thereafter, is the instance document key word with filepath;
(1.b) the where statement of AQL expanding XQuery increases base attribute key word ba, semantic attribute key word sf, and low-level image feature matching way key word 1f key word is to realize the constraint to the different faces of data;
(1.c) AQL increases the intelligence inquire statement, is key word with classify, the feature selecting when being used to specify cluster.
According to a further aspect of the invention, wherein step (2) further comprises:
(2.a) reduce the nested of FLWOR statement, the nested of regulation FLWOR statement can only carry out in for statement in AQL;
(2.b) rreturn value of qualification return statement, the rreturn value that regulation is nested in the FLWOR statement of internal layer can only be the raw data id of data.
According to a further aspect of the invention, wherein step (3) further comprises:
(3.a) executive plan is deposited with the XML file layout;
(3.b) executive plan is divided by many body inquiries;
(3.c) monomer inquiry inside comprises instance document, query context, base attribute query statement, semantic attribute query statement, low-level image feature matching way and intelligence inquire statement.
According to a further aspect of the invention, wherein step (6) further comprises:
(6.a) respective field of separation base attribute, semantic feature, low-level image feature inquiry;
(6.b) inquiry of base attribute, semantic feature forms the form of XQuery statement, and the inquiry of low-level image feature forms the query statement that meets the definition of AQL language grammar;
(6.c) query statement that forms according to step (6.b) generates executive plan.
The present invention's advantage compared with prior art is: the present invention considers the query demand in unstructured data storehouse, has realized content-based inquiry when keeping the XML query function, also can finish novel query manipulation simultaneously.
Description of drawings
Fig. 1 is a functional diagram of the present invention;
Fig. 2 is a process of analysis of the present invention;
Fig. 3 carries out flow process for many body inquiries of algorithm of the present invention;
Fig. 4 carries out flow process for the multiaspect inquiry of algorithm of the present invention;
Fig. 5 is a functional test results of the present invention;
Fig. 6 is a The performance test results of the present invention.
Embodiment
Below with reference to accompanying drawing, embodiments of the invention are described in detail.
At first to the present invention at functional requirement describe.
According to the requirement of data query function in the unstructured data storehouse, the function of this querying method (as shown in Figure 1) mainly contains:
1. basic query: utilize the base attribute and the semantic feature of data, search specific data by text message;
2. bottom inquiry: the input sample data, as a sub-picture, one section voice or one section video record, by the highest data set of low-level image feature matched and searched similarity;
3. multiaspect inquiry: comprehensive utilization base attribute, semantic feature and low-level image feature are realized correlation inquiry, accurately rapidly locating;
4. many body inquiries: utilize semantic feature, realize the correlation inquiry of multi-threaded numerous types of data;
5. intelligence inquire: realize intelligence inquire services such as cluster, classification, multidimensional data analysis to Query Result;
6. comprehensive inquiry: above-mentioned functions can nestedly be carried out, and can carry out on the basis of multiaspect inquiry as intelligence inquire.
The employed intelligence inquire language of non-structural data enquiry method AQL is expansion and simplification on the XQuery language basis, below this is elaborated:
1. expand the bottom query function: the inquiry about base attribute and semantic feature among the AQL can realize with the XQuery language, but the low-level image feature of tetrahedron model is not with the XML format description, so XQuery does not possess the function of low-level image feature retrieval, AQL inquires about low-level image feature and expands;
2. expansion intelligent query function: intelligence inquire comprises data is carried out multidimensional analysis, cluster and sort operation that XQuery does not possess corresponding function, and this is expanded;
3. expand the multiaspect query function: the multiaspect inquiry is meant that three faces of base attribute, semantic feature, low-level image feature to certain data carry out correlation inquiry.In the unstructured data storehouse, four face features of each data are stored respectively, and the multiaspect inquiry just can obtain final Query Result after will being merged by the result of different characteristic inquiry.XQuery do not support bottom inquiry and realizes needing multilayer nest, for complete realization multiaspect query function and the AQL language is become concisely, the multiaspect query function expanded;
4. expand many bodies query function: a plurality of body correlation inquiries are meant the correlation inquiry of different types of data, as image and text correlation inquiry, need inquire about a plurality of appointment collected works.XQuery can't finish the retrieval to a plurality of collected works in a statement, therefore expand many bodies query function;
5. reduce the nested of statement: in XQuery, can be nested flexibly between each statement.Too much is nested unnecessary to the AQL after the expansion, and influences the succinct of language, and this is simplified, and the nested of regulation AQL language can only carry out in for statement;
6. limit the rreturn value of return statement: simplify the rreturn value type of the statement that is nested in internal layer, regulation can only be returned raw data id, is convenient to the operation associated of a plurality of files.
Provide the grammar definition of AQL language below:
1. the outermost layer FLWOR statement syntax:
FLWORExpr::=ForClause?LetClause?WhereClause?OrderByClause?ReturnClause(FilePathClause)?
The outermost structure of AQL language is based on the FLWOR sentence structure of XQuery language, but added the statement that is used to specify the input file path behind the FLWOR statement;
2. the nested FLWOR statement syntax of internal layer:
FLWORExprInner::=ForClause(LetClause)?WhereClause?OrderByClause?ReturnClauseInner(FilePathClause)?
The FLWOR statement that is nested in internal layer is compared with outer field FLWOR statement, and difference mainly is return statement difference, and the FLWOR statement of internal layer can only return the raw data id of the data of looking into, and can not specify the concrete data item of returning;
3.for the statement syntax:
ForClause::=″for″″$″VarName″in″TypePath(“(”FLWORExprChanged”)”)?
The for statement is used to specify the scope of the involved data type of a multiaspect inquiry, and can realize many body inquiries by nested FLWOR statement;
4.let the statement syntax:
LetClause::=″let″″$″VarName″:=″LocalPart?(″,$″VarName″:=″LocalPart)*
The let statement is used to define the used path variable of a multiaspect inquiry, uses variable can improve the search efficiency of XML database;
5.where the statement syntax:
WhereClause::=″where″(“BA{”condition”}”
(“SF{”condition”}”)?(“LF{”lfcondition”}”)?)|(“SF{”condition”}”(“LF{”lfcondition”}”)?)|(”LF{”lfcondition”}”}
The where statement is used to define the restrictive condition that multiaspect is inquired about each facet.According to the facet kind, restrictive condition is divided into three kinds: base attribute condition, semantic feature condition, low-level image feature condition;
6.order the by statement syntax:
OrderByClause::=(“ba””{”((″order″″by″)|(″stable″″order″″by″))TypePath?OrderSpecList”}”)?(“sf””{”((″order″″by″)|(″stable″″order″″by″))TypePath?OrderSpecList”}”)?
Order by statement is used to define putting in order of rreturn value.Wherein, arrangement mode has by the base attribute arrangement with by two kinds of semantic feature arrangements;
7.return the statement syntax:
ReturnClause::=″$″VarName?LocalPart(IntelligenceOperation(“,”IntelligenceOperation)*)?
The return statement is used for specifying the rreturn value content, and in addition, intelligent query function is also expanded in the syntax of return statement.
The following describes the implementation of non-structural data enquiry method:
As shown in Figure 2, the idiographic flow of non-structural data enquiry method implementation is: the user sends after the AQL query requests, and system carries out pre-service earlier, and identification instance document field is if the request statement comprises then log file of instance document.The corresponding description field of separating base attribute, semantic feature, low-level image feature by keyword recognition.Form the query statement of XQuery form afterwards according to the description field of base attribute, semantic feature, form the query statement that meets the definition of AQL language grammar, generate executive plan according to the query statement that forms then according to the description field of low-level image feature.
Executive plan is the intermediate product that the AQL language is resolved.The AQL resolver can resolve to executive plan with the AQL language earlier, then execution is resolved in executive plan, obtains Query Result.The introducing of executive plan has strengthened the extensibility of AQL language, and operations such as query optimization can be carried out in the parsing of executive plan is carried out.
It is as follows that the XML formal definition is pressed in executive plan:
1. skin is the sign of each body in the inquiry of many bodies, is used for the correlation inquiry of several individualities that the inquiry of AQL statement is related to make a distinction;
2. internal layer is the sign of the inner multiaspect inquiry of each body, comprises base attribute, semantic feature, low-level image feature and four kinds of statements of intelligence inquire.Wherein, base attribute and semantic feature statement are the XQuery statement that can directly carry out, and the data type and the low-level image feature querying condition of inquiry described in the low-level image feature statement, separates with branch.The intelligence inquire condition described in the intelligence inquire statement.
3. the return results of last this inquiry of record of executive plan is represented with the XPath path expression.
Below AQL language implementation is described:
Form and level according to executive plan are divided, and the execution of AQL language also is divided into two-layer:
(3.1) the outer many bodies query execution plan (as shown in Figure 3) of carrying out begins to carry out the multiaspect inquiry from the innermost layer body, carries out the multiaspect inquiry according to the outside one by one one deck body of association between each individuality.After all body inquiries finished, generated query is id as a result, takes out corresponding rreturn value.
(3.2) internal layer is carried out multiaspect query execution plan (as shown in Figure 4), resolves the individual inner multiaspect inquiry of fill order.Divide according to tetrahedral facet, actuator is divided into base attribute executive plan actuator, semantic feature executive plan actuator, three parts of low-level image feature executive plan actuator, the operating result combiner merges the execution result of three actuators, generates the multiaspect Query Result of single body.
Below, execution result correctness of the present invention and execution performance are illustrated by to functional test of the present invention and performance test.Test data set is divided into image and text two big classes.Image is divided into animal again, plant, and eight subclasses such as activity, text is divided into finance again, military, ten subclasses such as education.The image data amount size is 100000, and text data amount size is 50000.
In order to finish test, designed following inquiry situation as test case:
Use-case 1:
The basic query use-case: query contents is that inquiry author is the text of ying in the animal subclassification under the image type, and returns the key word of qualified image.
Used AQL query statement is:
for?$i?in/image/animal?where?ba(./author=″ying″}return?$i/keywords
Use-case 2:
Bottom inquiry use-case: query contents is to inquire about the image the most similar to input picture by low-level image feature matching algorithm CEDD under the plant subclassification under the image type, and returns the raw data id of qualified image.
Used AQL query statement is:
for?$i?in/image/plant?where?lf(imagetype?match{CEDD}}return?$i/rdid?filepath:={F:/MyEclipse/workspace/ExistConnect/bat(1).jpg}
Use-case 3:
Multiaspect inquiry use-case: query contents be in the activity subclassification under the image type in the inquiry base attribute author be ying, Format is jpg in the semantic feature, the low-level image feature matching algorithm is CEDD, and returns the raw data id. of image
Used AQL query statement is:
for?$i?in/image/activity?where?ba{./author=“ying”}sf{./Format=″jpg″}lf{imagetype?match{CEDD}}return?$i/rdid?filepath:={F:/MyEclipse/workspace/ExistConnect/bat(1).jpg}
Use-case 4:
Many bodies inquiry use-cases: query contents be at first in the activity subclassification of internal layer under image in the inquiry base attribute author be ying, format is jpg in the semantic feature, and the low-level image feature matching algorithm is the image of CEDD, and has specified the path of the image file of input.Skin is inquired about under the education of text subclassification then, the mode of operation of two types of nested execution of file of text and image for inquiry in text with the result's that inquires about among the image keywords input as the low-level image feature inquiry, therefore, the outer query condition has: author is longhuiping in the base attribute, type is pdf in the semantic feature, and the text that contains the keywords item content of internal layer image Query Result returns the raw data id of qualified text.
Used AQL query statement is:
for?$i?in/text/education{for?$j?in/image/activity?where?ba{./author=″ying″}sf{./Format″jpg″}lf{imagetype?match{CEDD}}return?$j/rdid?filepath:={F:/MyEclipse/workspace/ExistConnect/0100050000000144400.jpg}}whereba{./author=″longhuiping″}sf{./type=″pdf″}return?$i/rdid
Use-case 5:
The intelligence inquire use-case: query contents be in the plant subclassification under image in the inquiry base attribute author be the text of ying, and according to the keywords item cluster of text, rreturn value is the raw data id of result data.
Used AQL query statement is:
for?$i?in/image/plant?where?ba{./author=“ying”}return?$i/rdid?classify?on/keywords
According to above 5 test cases, the function and the performance of system are tested.Function test method is for comparing querying condition is imported result who obtains and the result that the equivalent query statement of describing with the AQL syntax obtains respectively by visual query facility by clauses and subclauses.Performance test methods is the parsing execution time and the analysis of the different use-cases of record.Functional test, The performance test results such as Fig. 5 are shown in 6.Through test, on function, with consistent, illustrated that AQL resolves the correctness of actuator by the result of non-structural data enquiry method inquiry by visual query facility; On performance, the parsing execution time of test case is Millisecond.
What may be obvious that for the person of ordinary skill of the art draws other advantages and modification.Therefore, the present invention with wider aspect is not limited to shown and described specifying and exemplary embodiment here.Therefore, under situation about not breaking away from, can make various modifications to it by the spirit and scope of claim and the defined general inventive concept of equivalents thereof subsequently.
Claims (5)
1. non-structural data enquiry method, this method use unstructured data intelligence inquire language AQL, it is characterized in that step is as follows:
(1) existing XML querying method is improved, employed unstructured data intelligence inquire language AQL expands the XQuery language, characteristics at unstructured data, increase base attribute, semantic attribute, low-level image feature and cluster mode key sentence, to support multiaspect inquiry, Query By Example, intelligence inquire and the inquiry of many bodies;
(2) simplify and the irrelevant function of non-structural data enquiry, to improve analyzing efficiency;
(3) the definition executive plan is used to represent the implementation of non-structural data enquiry;
(4) resolving of AQL statement is: pre-service, and identification instance document field if statement comprises then log file of instance document, otherwise forwards step (5) to;
(5) according to the AQL grammer description field of different attribute in the statement is separated;
(6) different field of separating according to step (5) forms corresponding query statement and generates executive plan.
2. non-structural data enquiry method according to claim 1 is characterized in that: described step (1) further comprises:
(1.a) AQL has added the statement FilepathClause that is used to specify the instance document path based on the FLWOR statement of XQuery language thereafter, is the instance document key word with filepath;
(1.b) the where statement of AQL expanding XQuery increases base attribute key word ba, semantic attribute key word sf, and low-level image feature matching way key word 1f key word is to realize the constraint to the different faces of data;
(1.c) AQL increases the intelligence inquire statement, is key word with classify, the feature selecting when being used to specify cluster.
3. non-structural data enquiry method according to claim 1 is characterized in that: described step (2) further comprises:
(2.a) reduce the nested of FLWOR statement, the nested of regulation FLWOR statement can only carry out in for statement in AQL;
(2.b) rreturn value of qualification return statement, the rreturn value that regulation is nested in the FLWOR statement of internal layer can only be the raw data id of data.
4. non-structural data enquiry method according to claim 1 is characterized in that: described step (3) further comprises:
(3.a) executive plan is deposited with the XML file layout;
(3.b) executive plan is divided by many body inquiries;
(3.c) monomer inquiry inside comprises instance document, query context, base attribute query statement, semantic attribute query statement, low-level image feature matching way and intelligence inquire statement.
5. non-structural data enquiry method according to claim 2 is characterized in that: described step (6) further comprises:
(6.a) respective field of separation base attribute, semantic feature, low-level image feature inquiry;
(6.b) inquiry of base attribute, semantic feature forms the form of XQuery statement, and the inquiry of low-level image feature forms the query statement that meets the definition of AQL language grammar;
(6.c) query statement that forms according to step (6.b) generates executive plan.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102202447A CN102289482A (en) | 2011-08-02 | 2011-08-02 | Unstructured data query method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102202447A CN102289482A (en) | 2011-08-02 | 2011-08-02 | Unstructured data query method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102289482A true CN102289482A (en) | 2011-12-21 |
Family
ID=45335909
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011102202447A Pending CN102289482A (en) | 2011-08-02 | 2011-08-02 | Unstructured data query method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102289482A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107402920A (en) * | 2016-05-18 | 2017-11-28 | 北京京东尚科信息技术有限公司 | The method and apparatus for determining relation database table connection complexity factor |
CN107451214A (en) * | 2016-09-21 | 2017-12-08 | 广州特道信息科技有限公司 | A kind of non-primary key querying method and distributed NewSQL Database Systems |
CN109947947A (en) * | 2019-03-29 | 2019-06-28 | 北京泰迪熊移动科技有限公司 | A kind of file classification method, device and computer readable storage medium |
US10339133B2 (en) | 2013-11-11 | 2019-07-02 | International Business Machines Corporation | Amorphous data preparation for efficient query formulation |
CN111897911A (en) * | 2020-06-11 | 2020-11-06 | 中国科学院计算机网络信息中心 | Unstructured data query method and system based on secondary attribute graph |
WO2022083520A1 (en) * | 2020-10-19 | 2022-04-28 | 华为技术有限公司 | Data query method and apparatus, device, and storage medium |
CN115168408A (en) * | 2022-08-16 | 2022-10-11 | 北京永洪商智科技有限公司 | Query optimization method, device, equipment and storage medium based on reinforcement learning |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101477568A (en) * | 2009-02-12 | 2009-07-08 | 清华大学 | Integrated retrieval method for structured data and non-structured data |
CN101510206A (en) * | 2009-02-19 | 2009-08-19 | 北京思普科科技开发有限公司 | Non-structural data enquiry method |
-
2011
- 2011-08-02 CN CN2011102202447A patent/CN102289482A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101477568A (en) * | 2009-02-12 | 2009-07-08 | 清华大学 | Integrated retrieval method for structured data and non-structured data |
CN101510206A (en) * | 2009-02-19 | 2009-08-19 | 北京思普科科技开发有限公司 | Non-structural data enquiry method |
Non-Patent Citations (3)
Title |
---|
《计算机应用与软件》 20080531 张哲 XML模式到本体的提升--概念搜索算法 147-149页 1-5 第25卷, 第5期 * |
张哲: "XML模式到本体的提升——概念搜索算法", 《计算机应用与软件》 * |
李未等: "一种非结构化数据库的四面体数据模型", 《中国科学》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10339133B2 (en) | 2013-11-11 | 2019-07-02 | International Business Machines Corporation | Amorphous data preparation for efficient query formulation |
CN107402920A (en) * | 2016-05-18 | 2017-11-28 | 北京京东尚科信息技术有限公司 | The method and apparatus for determining relation database table connection complexity factor |
CN107402920B (en) * | 2016-05-18 | 2020-02-07 | 北京京东尚科信息技术有限公司 | Method and device for determining correlation complexity of relational database table |
CN107451214A (en) * | 2016-09-21 | 2017-12-08 | 广州特道信息科技有限公司 | A kind of non-primary key querying method and distributed NewSQL Database Systems |
CN107451214B (en) * | 2016-09-21 | 2020-05-19 | 云润大数据服务有限公司 | Non-primary key query method and distributed NewSQL database system |
CN109947947A (en) * | 2019-03-29 | 2019-06-28 | 北京泰迪熊移动科技有限公司 | A kind of file classification method, device and computer readable storage medium |
CN111897911A (en) * | 2020-06-11 | 2020-11-06 | 中国科学院计算机网络信息中心 | Unstructured data query method and system based on secondary attribute graph |
WO2022083520A1 (en) * | 2020-10-19 | 2022-04-28 | 华为技术有限公司 | Data query method and apparatus, device, and storage medium |
CN115168408A (en) * | 2022-08-16 | 2022-10-11 | 北京永洪商智科技有限公司 | Query optimization method, device, equipment and storage medium based on reinforcement learning |
CN115168408B (en) * | 2022-08-16 | 2024-05-28 | 北京永洪商智科技有限公司 | Query optimization method, device, equipment and storage medium based on reinforcement learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111680173B (en) | CMR model for unified searching cross-media information | |
KR101646754B1 (en) | Apparatus and Method of Mobile Semantic Search | |
CN102289482A (en) | Unstructured data query method | |
CN1845104B (en) | System and method for intelligent retrieval and processing of information | |
CN102722542B (en) | A kind of resource description framework graphic mode matching method | |
CN111177591B (en) | Knowledge graph-based Web data optimization method for visual requirements | |
US20150074112A1 (en) | Multimedia Question Answering System and Method | |
US10942977B2 (en) | Systems and methods for targeting, reviewing, and presenting online social network data by evidence context | |
CN102087669A (en) | Intelligent search engine system based on semantic association | |
CN105930362B (en) | Search for target identification method, device and terminal | |
CN114218400A (en) | Semantic-based data lake query system and method | |
CN111026710A (en) | Data set retrieval method and system | |
CN112597285B (en) | Man-machine interaction method and system based on knowledge graph | |
JP2019040598A5 (en) | ||
US9063957B2 (en) | Query systems | |
CN114218472A (en) | Intelligent search system based on knowledge graph | |
CN111104437A (en) | Test data unified retrieval method and system based on object model | |
KR20100066919A (en) | Triple indexing and searching scheme for efficient information retrieval | |
KR20120047622A (en) | System and method for managing digital contents | |
CN117194647B (en) | Intelligent question-answering system, method and device for offline environment | |
CN111831787B (en) | Unstructured data information query method and system based on secondary attributes | |
US20070150458A1 (en) | System for extending data query using ontology, and method therefor | |
KR20100073793A (en) | Apparatus and method for knowledge structuralization using collective intelligence, and method for providing knowledge service using it | |
Rak et al. | Multilabel associative classification categorization of MEDLINE articles into MeSH keywords | |
Modoni et al. | A semantic framework for graph-based enterprise search |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20111221 |