Construct the method and device of party's portraitTechnical field
The present invention relates to data processing fields, in particular to a kind of method and device of building party portrait.
Background technique
The relevant information of many parties is usually contained in a administration of justice document, these relevant informations are to analysis partyFeature, building party's portrait have very big value.
The relevant technologies are mainly taken passages pair when analyzing the party in judicial document using manual type from judicial documentThe party's information answered is classified to these information, is normalized and statistical disposition.Wherein, message digest and classifying rules be allIt is to be determined by concrete application demand, the message digest and classifying rules of different application have different emphasis.
However, accuracy rate is poor, it is difficult in a short time since above-mentioned technical proposal is mainly the low efficiency by manually realizingThe analysis work of a large amount of judicial documents is completed, and manually obtains data there are standards that different, more or less, reuses and comparesDifficult defect.
Although additionally providing the effective building ordinary user portrait of one kind in the related technology (such as login/browsing electric business websiteUser portrait) technical solution, but the program is realized by the specified dimension of statistical framework data.AndJudicial document is as text data, and text data is unstructured data, thus existing building ordinary user portrait is automaticConstruction method cannot be directly used to building party's portrait.
In view of the above-mentioned problems, currently no effective solution has been proposed.
Summary of the invention
The embodiment of the invention provides a kind of method and devices of building party portrait, at least to solve in the related technologyThe technical issues of party's portrait can not be constructed automatically.
According to an aspect of an embodiment of the present invention, a kind of method of building party portrait is provided, comprising: from preparatoryTarget party is searched in the judicial domain ontology of building, wherein comprising Ontological concept and being used in above-mentioned judicial domain ontologyThe structured data of the attribute of Ontological concept is described, above-mentioned Ontological concept includes party；After finding above-mentioned target party,Use from the structured data that user's input is chosen or received in the above structure data of the attribute for describing Ontological conceptIn the structured data for the attribute for describing above-mentioned target party；According to selection or received work as thing for describing above-mentioned targetThe structured data of the attribute of people constructs party's portrait of above-mentioned target party.
Further, according to it is selection or it is received for describe above-mentioned target party attribute structured data,Construct above-mentioned target party party's portrait include: in the case where above-mentioned target party is individual party, according toThat chooses is used to describe some or all of the attribute of above-mentioned target party structured data, constructs working as above-mentioned individual partyThing people portrait；In the case where above-mentioned target party is group party, thing is worked as describing above-mentioned target according to selectionSome or all of attribute of people structured data constructs party's portrait of above-mentioned group party.
Further, above-mentioned judicial domain ontology is constructed by following steps: according to above-mentioned Ontological concept and for describingThe above-mentioned attribute of Ontological concept determines that the Feature Words of the syntax for parsing judicial document and the corresponding selectivity of Feature Words becomeAmount；According to the determining corresponding selective variable of Feature Words and Feature Words, the above-mentioned syntax are constructed；Using the above-mentioned syntax of building,Parsing needs the judicial document parsed, obtains judicial document parsing result；Above-mentioned judicial document parsing result is filled into above-mentionedIn judicial domain ontology.
Further, according to the corresponding selective variable of determining Feature Words and Feature Words, after constructing the above-mentioned syntax,The above method further include: obtain the style of writing feature of judicial document；According to the style of writing feature of the above-mentioned syntax and above-mentioned judicial document, structureGrammatical paragraph feature templates and grammatical paragraph position feature template are built, include corresponding template characteristic and the syntax in each templateSubset, wherein using the above-mentioned syntax of building, parsing needs the judicial document parsed, and obtaining judicial document parsing result includes:Use the above-mentioned grammatical paragraph feature templates of building or above-mentioned grammatical paragraph feature templates and above-mentioned grammatical paragraph position featureTemplate parses the judicial document of above-mentioned needs parsing paragraph by paragraph, obtains judicial document parsing result.
Further, it using above-mentioned grammatical paragraph feature templates and above-mentioned grammatical paragraph position feature template, parses paragraph by paragraphThe judicial document of above-mentioned needs parsing, obtaining judicial document parsing result includes: to mention from the judicial document that above-mentioned needs parseThe target paragraph taken；For the corresponding grammatical paragraph feature templates of above-mentioned target paragraph matching；If successful match, using matchingGrammatical paragraph feature templates, parse above-mentioned target paragraph, obtain corresponding parsing result, and jump to next target paragraphProcess of analysis；If it fails to match, for the corresponding grammatical paragraph position feature template of above-mentioned target paragraph matching, if matching atFunction parses above-mentioned target paragraph then using the grammatical paragraph position feature template matched, obtains corresponding parsing result, andJump to the process of analysis of next target paragraph.
Further, during parsing above-mentioned target paragraph, the above method further include: if corresponding parsing result isSky, then at least record the judicial document of above-mentioned needs parsing sequence number and above-mentioned target paragraph；Record result is filled intoIt states in judicial domain ontology.
Further, in the above-mentioned syntax using building, parsing needs the judicial document parsed, obtains judicial document parsingAs a result after, the above method further include: according to above-mentioned judicial document parsing result, construct the incidence relation between all parties；Count the Numeric Attributes of each party；After the completion of incidence relation between all parties constructs, each party is countedIncidence relation each dimension statistical value；By the numerical value of incidence relation, above-mentioned each party between above-mentioned all partiesThe statistical value of each dimension of type attribute and the incidence relation of above-mentioned each party is filled into above-mentioned judicial domain ontology.
According to another aspect of an embodiment of the present invention, a kind of device of building party portrait is additionally provided, comprising: searchUnit, for searching target party from the judicial domain ontology constructed in advance, wherein include in above-mentioned judicial domain ontologyThe structured data of Ontological concept and the attribute for describing Ontological concept, above-mentioned Ontological concept include party；Processing unit is usedIn after finding above-mentioned target party, chooses or connect from the above structure data of the attribute for describing Ontological conceptReceive the structured data of the attribute for describing above-mentioned target party in the structured data of user's input；First construction unit,For the structured data according to selection the or received attribute for being used to describe above-mentioned target party, constructs above-mentioned target and work asThe party of thing people draws a portrait.
Further, above-mentioned first construction unit includes: the first building module, for being individual in above-mentioned target partyIn the case where party, according to some or all of the attribute for describing above-mentioned target party of selection structured data, structureBuild party's portrait of above-mentioned individual party；Second building module, for being group party's in above-mentioned target partyIn the case of, according to some or all of the attribute for describing above-mentioned target party of selection structured data, construct above-mentioned groupThe party of body party draws a portrait.
Further, above-mentioned apparatus further include: determination unit is used for after obtaining judicial domain ontology, according to above-mentionedOntological concept and above-mentioned attribute for describing Ontological concept determine the Feature Words and feature of the syntax for parsing judicial documentThe corresponding selective variable of word；Second construction unit, for according to the corresponding selective variable of determining Feature Words and Feature Words,Construct the above-mentioned syntax；Resolution unit, for using the above-mentioned syntax of building, parsing needs the judicial document parsed, obtains the administration of justiceDocument parsing result；First fills unit, for above-mentioned judicial document parsing result to be filled into above-mentioned judicial domain ontology.
Further, above-mentioned apparatus further include: acquiring unit, for according to the determining corresponding choosing of Feature Words and Feature WordsSelecting property variable after constructing the above-mentioned syntax, obtains the style of writing feature of judicial document；Third construction unit, for according to above-mentioned textThe style of writing feature of method and above-mentioned judicial document, constructs grammatical paragraph feature templates and grammatical paragraph position feature template, Mei GemoAll comprising corresponding template characteristic and grammatical subset in plate, wherein above-mentioned resolution unit is also used to: using the above-mentioned syntax of buildingParagraph feature templates or above-mentioned grammatical paragraph feature templates and above-mentioned grammatical paragraph position feature template, parse above-mentioned paragraph by paragraphThe judicial document for needing to parse obtains judicial document parsing result.
Further, above-mentioned resolution unit includes: extraction module, for extracting from the judicial document that above-mentioned needs parseTarget paragraph；Matching module, for matching corresponding grammatical paragraph feature templates for above-mentioned target paragraph；First parsing mouldBlock is used for: if successful match, using the grammatical paragraph feature templates matched, being parsed above-mentioned target paragraph, is obtained correspondingParsing result, and jump to the process of analysis of next target paragraph；Second parsing module, is used for: being upper if it fails to matchIt states the corresponding grammatical paragraph position feature template of target paragraph matching and uses the grammatical paragraph position matched if successful matchFeature templates are set, above-mentioned target paragraph is parsed, obtain corresponding parsing result, and jump to the resolution flow of next target paragraphJourney.
Further, above-mentioned apparatus further include: recording unit is used for during parsing above-mentioned target paragraph, if rightThe parsing result answered is sky, then at least record the judicial document of above-mentioned needs parsing sequence number and above-mentioned target paragraph；SecondFills unit is filled into above-mentioned judicial domain ontology for that will record result.
Further, above-mentioned apparatus further include: the 4th construction unit, in the above-mentioned syntax using building, parsing to be neededThe judicial document to be parsed, after obtaining judicial document parsing result, according to above-mentioned judicial document parsing result, thing is respectively worked as in buildingIncidence relation between people；First statistic unit, for counting the Numeric Attributes of each party；Second statistic unit is usedAfter the completion of the incidence relation between all parties constructs, the statistics of each dimension of the incidence relation of each party is countedValue；Third fills unit, for by the Numeric Attributes of incidence relation, above-mentioned each party between above-mentioned all parties withAnd the statistical value of each dimension of the incidence relation of above-mentioned each party is filled into above-mentioned judicial domain ontology.
In embodiments of the present invention, by the way of based on judicial domain ontological construction party portrait, by from preparatoryTarget party is searched in the judicial domain ontology of building, wherein comprising Ontological concept and for describing in judicial domain ontologyThe structured data of the attribute of Ontological concept, Ontological concept include party；After finding target party, from for describing thisIt chooses or receives in the structured data of the attribute of body concept and be used to describe target party in the structured data of user's inputAttribute structured data；According to selection or received for describing the structured data of the attribute of target party, buildingThe party of target party draws a portrait, and has reached the dependency structure number of the attribute by the party in selection judicial domain ontologyThe purpose of the portrait of the party is constructed accordingly, to realize the technical effect of automatic building party portrait, and then is solvedThe technical issues of party's portrait in the related technology can not be constructed automatically.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hairBright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is a kind of flow chart of the method for optional building party portrait according to an embodiment of the present invention；
Fig. 2 is a kind of schematic diagram of the device of optional building party portrait according to an embodiment of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present inventionAttached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is onlyThe embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill peopleThe model that the present invention protects all should belong in member's every other embodiment obtained without making creative workIt encloses.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this wayData be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein orSequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that coverCover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited toStep or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, productOr other step or units that equipment is intrinsic.
Embodiment 1
According to embodiments of the present invention, a kind of embodiment of the method for building party portrait is provided, it should be noted thatThe step of process of attached drawing illustrates can execute in a computer system such as a set of computer executable instructions, also,It, in some cases, can be to be different from shown in sequence execution herein although logical order is shown in flow chartsThe step of out or describing.
Fig. 1 is a kind of flow chart of the method for optional building party portrait according to an embodiment of the present invention, such as Fig. 1 instituteShow, this method comprises the following steps:
Step S102 searches target party, wherein in judicial domain ontology from the judicial domain ontology constructed in advanceThe structured data of attribute comprising Ontological concept and for describing Ontological concept, Ontological concept include party；
Step S104 is selected from the structured data of the attribute for describing Ontological concept after finding target partyTake or receive the structured data of the attribute for describing target party in the structured data of user's input；
Step S106, according to selection or received for describing the structured data of the attribute of target party, buildingThe party of target party draws a portrait.
It should be noted that according to the professional standards such as China's laws and regulations and people's court's Information System configuration technical specification(abbreviation method mark) can construct the judicial domain ontology centered on judicial party.Wherein, judicial domain ontology includes ontologyThe structured data of concept and the attribute for describing Ontological concept.Herein, the Ontological concept of core is in addition to including in judicial documentExcept the party being related to, can also include case, applicable law, case by, accept the concepts such as law court, time, and judicial textThe other information being related in book can be used as the attribute of these concepts.When implementation, method mark and law related data can be usedThe type and value of specification Ontological concept and the attribute for describing Ontological concept, reach in judicial document main concept andExpress consistent purpose.For example, judicial domain ontology can be database, Ontological concept be can be " case ", describe ontologyThe structured data of the attribute of concept " case " can be the type of case, e.g., criminal case, civil case etc..It is led in the administration of justiceIn the ontology of domain, Ontological concept storage corresponding with the description structured data of attribute of Ontological concept.
Based on technical solution provided by the invention, when user needs to construct the party of certain party (i.e. target party)When portrait, target party first can be found out from all Ontological concepts of judicial domain ontology；Finding the targetAfter party, then choose from the structured data of all attributes for describing Ontological concept oneself needs for describing the meshMark some or all of the attribute of party structured data；Finally according to some or all of selection structured data, the mesh is constructedMark party's portrait of party.In addition, in the actual operation process, user is when constructing party's portrait, it is also possible to can be defeatedEnter judicial domain ontology originally no Ontological concept and its association attributes, at this time, on the one hand, system starts to construct party's pictureAs after, structured data that can need user and Ontological concept and its association attributes that judicial domain ontology just had originally is exportedIt is presented to user interface, and by way of figure/table；On the other hand, system can also export user demand but judicial neckThe list of the originally no Ontological concept of domain ontology and its association attributes.
Due to judicial domain ontology saved based on judicial document extract it is many can accurately describe party andThe structured data of association attributes, thus party accurately and accurately can be constructed based on judicial document using above-mentioned technical proposalPortrait.
Through the embodiment of the present invention, by the way of based on judicial domain ontological construction party portrait, reached and passed throughSelect the dependency structure data of the attribute of the party in judicial domain ontology to construct the purpose of the portrait of the party, thusThe technical effect of automatic building party portrait is realized, and then solves and can not construct party's portrait automatically in the related technologyThe technical issues of.
Optionally, according to selection or received for describing the structured data of the attribute of target party, building meshThe party of mark party draws a portrait
S2, in the case where target party is individual party, according to selection for describing the category of target partyProperty some or all of structured data, construct party's portrait of individual party；
S4, in the case where target party is group party, according to selection for describing the category of target partyProperty some or all of structured data, building group party party portrait.
Since group party includes multiple individual parties with incidence relation, building group party'sWhen party draws a portrait, in addition to needing in building group party other than each individual, it is also necessary to construct the pass between these individualsConnection relationship.
It should be noted that the embodiment of the present invention is a kind of using domain body and machine learning techniques, computer is realizedThe method for being automatically based upon judicial document building party's portrait.Certain human-computer interaction is needed in building process, such as by userThe various demands of input and adjustment building party's portrait, what other work were automatically performed by computer.The present invention can efficiently,It is accurately finished the processing and analysis of a large amount of judicial documents, building party's portrait；And the sum of user's adjustment can be timely responded toNewly-increased demand shows the difference of party's portrait from result data, meets the needs of user constantly excavates new information.ThisIt invents while being suitable for building party's individual portrait and group and draw a portrait.
Optionally, judicial domain ontology is constructed by following steps:
S6, the attribute according to Ontological concept and for describing Ontological concept determine the syntax for parsing judicial documentFeature Words and the corresponding selective variable of Feature Words；
S8, according to the determining corresponding selective variable of Feature Words and Feature Words, the building syntax；
S10, using the syntax of building, parsing needs the judicial document parsed, obtains judicial document parsing result；
Judicial document parsing result is filled into judicial domain ontology by S12.
That is, in order to enrich, expand existing judicial domain ontology, judicial domain ontology can be used, specifically when implementingThe Ontological concept in judicial domain ontology and the attribute for describing Ontological concept can be used, building computer can solve automaticallyAnalyse the syntax of judicial document, wherein the syntax of judicial document are the frame mode of language, composition and Bianization the ﹐ phrase including wordWith the tissue of sentence.And using the syntax of building, the parsing judicial document that more newly-increased needs parse, and then by judicial documentParsing result is filled into judicial domain ontology, can specifically be filled into pair of the correspondence Ontological concept in judicial domain ontologyIt answers in attribute.Wherein, which is based on context-free grammar.When parsing document using the syntax, mainly with judgement documentIn parsed based on single sentence (hereinafter referred to as simple sentence), obtained from simple sentence related required for building party's portraitInformation.The term of the Feature Words of the syntax and selective variable both is from judicial domain ontology.
It should be noted that system can be independent to each judicial document parsing result after parsing judicial document every timeIt saves, while the data of all accumulations can also be uniformly saved together.For unified preserving type, due to allThe judicial document parsing result that the secondary judicial document of parsing obtains will be fused together, so that result set is constantly accumulated, be used for structureThe structured data for building party's portrait (including individual party portrait and all party's portraits) is enriched constantly, is increased, thusFine and comprehensive party's portrait can be formed.Specifically, when constructing party's portrait, user can according to need choosingSelect this, former each data or even all time data accumulation results.Meanwhile technical solution of the present invention can also utilizeThe various data for constructing party's portrait constantly enhance system analysis document and building party's portrait with the method for machine learningAbility.
Optionally, according to the corresponding selective variable of determining Feature Words and Feature Words, after constructing the syntax, onState method further include:
S14 obtains the style of writing feature of judicial document；
S16 constructs grammatical paragraph feature templates and grammatical paragraph position is special according to the style of writing feature of the syntax and judicial documentTemplate is levied, comprising corresponding template characteristic and grammatical subset in each template,
Correspondingly, parsing needs the judicial document parsed using the syntax of building, judicial document parsing result packet is obtainedIt includes:
S18 uses the grammatical paragraph feature templates of building, or grammatical paragraph feature templates and grammatical paragraph position featureTemplate, parsing needs the judicial document parsed paragraph by paragraph, obtains judicial document parsing result.
Generally, judicial document all can think paragraph comprising appealing paragraph, adjudicating paragraph, true paragraph and law court, andEvery kind of paragraph can all have the exclusive style of writing feature of oneself.Different paragraphs often has different style of writing features, for example, appealing sectionFalling is plaintiff describes to prosecute the paragraph of defendant why, is case " reason paragraph "；Judgement paragraph is according to after law article legal principleThe paragraph made decisions is case " result paragraph "；True paragraph is the description paragraph that thing occurs between former defendant, is case" the objective description paragraph " of part；Law court thinks that paragraph is the paragraph that judge does reason according to prosecution content, the fact, evidence, is case" reason things out paragraph " of part.
Thus according to the style of writing feature of judicial document and and description with different characteristic style of writing used in the syntax, structureGrammatical paragraph feature templates and grammatical paragraph position feature template are built, each template includes template characteristic and corresponding syntaxCollect two parts.In use, the two templates will guide computer software on fixed paragraph and paragraph position using most suitably usedGrammatical subset so that the syntax parse the performance of judicial document and accuracy all greatly improves.
Optionally, using grammatical paragraph feature templates and grammatical paragraph position feature template, parsing needs to parse paragraph by paragraphJudicial document, obtaining judicial document parsing result includes:
S20, from the target paragraph for needing to extract in the judicial document parsed；
S22, for the corresponding grammatical paragraph feature templates of target paragraph matching；
S24, using the grammatical paragraph feature templates matched, parses target paragraph, obtains corresponding if successful matchParsing result, and jump to the process of analysis of next target paragraph；
S26 matches corresponding grammatical paragraph position feature template if it fails to match for target paragraph, if successful match,Then using the grammatical paragraph position feature template matched, target paragraph is parsed, obtains corresponding parsing result, and jump to downThe process of analysis of one target paragraph.
That is, so that each paragraph for needing the judicial document parsed is matched grammatical paragraph feature templates when implementing, ifSuccessful match then calls the grammatical paragraph feature templates matched, parses the paragraph with the grammatical subset in the template, and will solutionThe information separated out is filled into the correspondence attribute in judicial domain ontology.If it fails to match, then makes when previous paragraphs matching textMethod paragraph position feature template parses the paragraph using the grammatical subset of the template, and will parsing at this time if successful matchInformation out is filled into the correspondence attribute in judicial domain ontology, at this time if it fails to match, enters next paragraphProcess of analysis, until having handled whole paragraphs of the paperwork.
Further, after all judicial documents are parsed, system can also belong to according to the party parsedProperty, the incidence relation between party is constructed, and count the Numeric Attributes of each party, i.e. attribute value, based on thisTo count the Numeric Attributes of party group.After the completion of the building of party's incidence relation, each party's correlativity is countedEach dimension statistical value, based on this come count party group relationship each dimension statistical value, and by theseAttribute value, relationship and statistical value are all stored in database.
Further, it is possible to show that party's individual is drawn a portrait using all data in user interface and above-mentioned databaseIt draws a portrait with group.When the attribute that user selects specific individual, group to gather in interface, system can pass through OLAP technologyThe specific dimension data and aggregated data of party's individual and group is presented.
It should be noted that including but not limited to template matching and synonym column can be used in above-mentioned various matchingsThe matched method of table is matched.
Optionally, during parsing target paragraph, the above method further include:
S28 at least records the sequence number and target phase for needing the judicial document parsed if corresponding parsing result is skyIt falls；Record result is filled into judicial domain ontology.
That is, in resolving, if grammatical paragraph feature templates or grammatical paragraph position feature template matching success,But the information parsed is sky, then it can be by the Noumenon property set of document sequence number, sentence and be likely to require fillingAll note is in log.In this way, system can be with the shape of list when attribute oneself select when the user clicks but barrenCorresponding judicial document and specific paragraph therein, sentence etc. is presented in formula.
Further, for either with or without successful match paragraph and/or sentence, system can be to these paragraphs and/or sentenceSon carries out data mining, merges identical paragraph and/or sentence, and attempt with the other attribute datas obtained to these sectionsIt falls and/or sentence is matched, count the Ontological concept and association attributes that may include in these paragraphs and/or sentence.SeparatelyOutside, the automatic study that Frequent episodes method carries out the syntax can also be used but be not limited to, so that system developer and guardian are setMeter writes the new syntax.
Optionally, in the syntax using building, parsing needs the judicial document that parses, obtain judicial document parsing result itAfterwards, the above method further include:
S30 constructs the incidence relation between all parties according to judicial document parsing result；
S32 counts the Numeric Attributes of each party；
S34 counts each dimension of the incidence relation of each party after the completion of the incidence relation between all parties constructsThe statistical value of degree；
S36, by the Numeric Attributes of incidence relation, each party between all parties and the pass of each partyThe statistical value of each dimension of connection relationship is filled into judicial domain ontology.
Embodiment 2
According to embodiments of the present invention, a kind of Installation practice of building party portrait is provided.
Fig. 2 is a kind of schematic diagram of the device of optional building party portrait according to an embodiment of the present invention, such as Fig. 2 instituteShow, which includes: searching unit 202, for searching target party from the judicial domain ontology constructed in advance, whereinStructured data in judicial domain ontology comprising Ontological concept and the attribute for describing Ontological concept, Ontological concept include working as thingPeople；Processing unit 204, for after finding target party, from the structured data of the attribute for describing Ontological conceptChoose or receive the structured data of the attribute for describing target party in the structured data of user's input；First buildingUnit 206, for the structured data according to selection the or received attribute for being used to describe target party, building target is worked asThe party of thing people draws a portrait.
It should be noted that according to the professional standards such as China's laws and regulations and people's court's Information System configuration technical specification(abbreviation method mark) can construct the judicial domain ontology centered on judicial party.Wherein, judicial domain ontology includes ontologyThe structured data of concept and the attribute for describing Ontological concept.Herein, the Ontological concept of core is in addition to including in judicial documentExcept the party being related to, can also include case, applicable law, case by, accept the concepts such as law court, time, and judicial textThe other information being related in book can be used as the attribute of these concepts.When implementation, method mark and law related data can be usedThe type and value of specification Ontological concept and the attribute for describing Ontological concept, reach in judicial document main concept andExpress consistent purpose.For example, judicial domain ontology can be database, Ontological concept be can be " case ", describe ontologyThe structured data of the attribute of concept " case " can be the type of case, e.g., criminal case, civil case etc..It is led in the administration of justiceIn the ontology of domain, Ontological concept storage corresponding with the description structured data of attribute of Ontological concept.
Based on technical solution provided by the invention, when user needs to construct the party of certain party (i.e. target party)When portrait, target party first can be found out from all Ontological concepts of judicial domain ontology；Finding the targetAfter party, then choose from the structured data of all attributes for describing Ontological concept oneself needs for describing the meshMark some or all of the attribute of party structured data；Finally according to some or all of selection structured data, the mesh is constructedMark party's portrait of party.In addition, in the actual operation process, user is when constructing party's portrait, it is also possible to can be defeatedEnter judicial domain ontology originally no Ontological concept and its association attributes, at this time, on the one hand, system starts to construct party's pictureAs after, structured data that can need user and Ontological concept and its association attributes that judicial domain ontology just had originally is exportedIt is presented to user interface, and by way of figure/table；On the other hand, system can also export user demand but judicial neckThe list of the originally no Ontological concept of domain ontology and its association attributes.
Due to judicial domain ontology saved based on judicial document extract it is many can accurately describe party andThe structured data of association attributes, thus party accurately and accurately can be constructed based on judicial document using above-mentioned technical proposalPortrait.
Through the embodiment of the present invention, by the way of based on judicial domain ontological construction party portrait, reached and passed throughSelect the dependency structure data of the attribute of the party in judicial domain ontology to construct the purpose of the portrait of the party, thusThe technical effect of automatic building party portrait is realized, and then solves and can not construct party's portrait automatically in the related technologyThe technical issues of.
Optionally, above-mentioned first construction unit includes: the first building module, for being individual party in target partyIn the case where, according to some or all of the attribute for describing target party of selection structured data, building individual works as thingThe party of people draws a portrait；Second building module, is used in the case where target party is group party, according to the use of selectionIn some or all of the attribute of description target party structured data, party's portrait of building group party.
Optionally, above-mentioned apparatus further include: determination unit is used for after obtaining judicial domain ontology, general according to ontologyConsidering the attribute for describing Ontological concept, Feature Words and the corresponding choosing of Feature Words of the syntax for parsing judicial document are determinedSelecting property variable；Second construction unit, for according to the determining corresponding selective variable of Feature Words and Feature Words, the building syntax；Resolution unit, for using the syntax of building, parsing needs the judicial document parsed, obtains judicial document parsing result；FirstFills unit, for judicial document parsing result to be filled into judicial domain ontology.
Optionally, above-mentioned apparatus further include: acquiring unit, for according to the determining corresponding selection of Feature Words and Feature WordsProperty variable, after the building syntax, obtain the style of writing feature of judicial document；Third construction unit, for according to the syntax and administration of justice textThe style of writing feature of book constructs grammatical paragraph feature templates and grammatical paragraph position feature template, includes to correspond in each templateTemplate characteristic and grammatical subset, wherein resolution unit is also used to: using the grammatical paragraph feature templates of building, or the syntaxParagraph feature templates and grammatical paragraph position feature template, parsing needs the judicial document parsed paragraph by paragraph, obtains judicial document solutionAnalyse result.
Optionally, above-mentioned resolution unit includes: extraction module, the target for extracting from the judicial document that needs parseParagraph；Matching module, for matching corresponding grammatical paragraph feature templates for target paragraph；First parsing module, is used for: ifWith success, then using the grammatical paragraph feature templates matched, target paragraph is parsed, obtains corresponding parsing result, and jumpTo the process of analysis of next target paragraph；Second parsing module, is used for: if it fails to match, matching and corresponds to for target paragraphGrammatical paragraph position feature template, if successful match, use the grammatical paragraph position feature template matched, parse targetParagraph obtains corresponding parsing result, and jumps to the process of analysis of next target paragraph.
Optionally, above-mentioned apparatus further include: recording unit is used for during parsing target paragraph, if corresponding solutionAnalysing result is sky, then at least records the sequence number and target paragraph for needing the judicial document parsed；Second fills unit, being used for willRecord result is filled into judicial domain ontology.
Optionally, above-mentioned apparatus further include: the 4th construction unit, in the syntax using building, parsing to need to parseJudicial document, after obtaining judicial document parsing result, according to judicial document parsing result, construct the pass between all partiesConnection relationship；First statistic unit, for counting the Numeric Attributes of each party；Second statistic unit, for respectively working as thingAfter the completion of incidence relation building between people, the statistical value of each dimension of the incidence relation of each party is counted；Third fillingUnit, for by the association of the Numeric Attributes of incidence relation, each party between all parties and each partyThe statistical value of each dimension of relationship is filled into judicial domain ontology.
It should be noted that device section Example is same or like with corresponding method section Example；Device partWorking principle, the function of realization and the technical effect reached of each functional unit/module in embodiment respectively with it is correspondingCorrespondence step in method section Example is same or like, and details are not described herein.
The device of above-mentioned building party portrait includes processor and memory, above-mentioned searching unit, processing unit and theOne construction unit etc. stores in memory as program unit, executes above procedure stored in memory by processorUnit.
Include kernel in processor, is gone in memory to transfer corresponding program unit by kernel.Kernel can be set oneOr more, content of text is parsed by adjusting kernel parameter.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/Or the forms such as Nonvolatile memory, if read-only memory (ROM) or flash memory (flash RAM), memory include that at least one is depositedStore up chip.
Present invention also provides a kind of embodiments of computer program product, when executing on data processing equipment, fitIn the program code for executing initialization there are as below methods step: searching target from the judicial domain ontology constructed in advance and work as thingPeople, wherein the structured data in judicial domain ontology comprising Ontological concept and the attribute for describing Ontological concept, Ontological conceptIncluding party；After finding target party, from the structured data of the attribute for describing Ontological concept choose orReceive the structured data of the attribute for describing target party in the structured data of user's input；According to selection or connectThe structured data for the attribute for describing target party received, party's portrait of building target party.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodimentThe part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through othersMode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke YiweiA kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine orPerson is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutualBetween coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or moduleIt connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unitThe component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multipleOn unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unitIt is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated listMember both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent productWhen, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantiallyThe all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other wordsIt embodies, which is stored in a storage medium, including some instructions are used so that a computerEquipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole orPart steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are depositedReservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program codeMedium.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the artFor member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answeredIt is considered as protection scope of the present invention.