Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with this specification.On the contrary, they are only and such as institute
The example of the consistent device and method of some aspects be described in detail in attached claims, this specification.
It is only to be not intended to be limiting this explanation merely for for the purpose of describing particular embodiments in the term that this specification uses
Book.The "an" of used singular, " described " and "the" are also intended to packet in this specification and in the appended claims
Most forms are included, unless the context clearly indicates other meaning.It is also understood that term "and/or" used herein is
Refer to and includes that one or more associated any or all of project listed may combine.
It will be appreciated that though various information may be described using term first, second, third, etc. in this specification, but
These information should not necessarily be limited by these terms.These terms are only used to for same type of information being distinguished from each other out.For example, not taking off
In the case where this specification range, the first information can also be referred to as the second information, and similarly, the second information can also be claimed
For the first information.Depending on context, word as used in this " if " can be construed to " ... when " or
" when ... " or " in response to determination ".
The automatic resume assessment technology of target actually also rests on resume identification and resolution phase.Resume identification and parsing are
It realizes and printed page analysis is carried out to resume, and according to preset aiming field, job hunter's corresponding information is extracted in identification, completes
The process of resume automatic reading.It has been found that identifying and parsing for resume, the method based on pure rule can be according to preparatory
The keyword of setting is segmented resume, it is believed that multiple continuous rows constitute a section, such as essential information, education experience, work warp
It goes through.Each point of good paragraph is matched according to regular expressions, the matched mode of pattern rule extracts key message.So
And the method for pure rule only can be only achieved higher accuracy rate when resume has fixed form, this is not in practical application
Reality.In order to cope with resume structure complicated and changeable and reach certain order of accuarcy, need manual compiling largely regular, together
When need to safeguard a large amount of keywords and thesaurus, with the continuous rising of resume parsing demand, later maintenance is also increasingly number
It is tired.The case where can not being covered for rule, the poor recall rate of parsing.It is assessed for resume, current resume assessment technology is not
Really automatic talent's purpose of appraisals can be reached, the information of only a few can only be included in scope of assessment, be mainly easy extraction
Structured message, the keyword being matched to etc..It is big to be included in work experience, project experience etc. for many information actually in resume
It is the important evidence for evaluating applicant's comprehensive strength and ability to work in the natural text description of section.But in current technology,
These information can not be just extracted first in resolving, generally can only be at whole section it is even more impossible to be included in final resume assessment
After input system, still to be assessed by the mode of manual read and judge, human cost is high.
In consideration of it, this specification embodiment provides a kind of scheme of knowledge based map progress resume assessment, for pure rule
The then insurmountable big segment description text resolution problem of method carries out semantic level to it by natural language processing technique
Understand, refine the contents of a project, the degree of participation of applicant, performance level etc., and combines the correlation of biographic information and post, answers
The every ability for the person of engaging scores, and realizes that resume content truly fully assesses.Such as, it can use and known based on resume
Know the resume that the data that map carries out processing acquisition to resume to be assessed indicate and data content goes fitting to assess, works as fitting
Resume meet the requirements after, can be used to assess new resume.
As shown in Figure 1, being a kind of this specification knowledge based map progress resume shown according to an exemplary embodiment
The flow chart of the method for assessment advances with the history resume assessed and generates resume knowledge mapping, the resume knowledge graph
Include correlation information of the biographic information of history resume relative to positions demand in spectrum, includes unstructured in resume to be assessed
Information, which comprises
In a step 102, it is carried out using semantic component of the resume knowledge mapping to unstructured information in resume to be assessed
Mark obtains the first structure label that can indicate the unstructured information;
At step 104, the positions demand of target post is obtained;
In step 106, it is opposite that the unstructured information is obtained from resume knowledge mapping by first structure label
In the correlation information of the positions demand, and determined using the correlation information as evaluation factor resume to be assessed relative to
The correlation of target post.
The method that knowledge based map provided in this embodiment carries out resume assessment can be executed by software, can also be led to
It crosses the mode that software and hardware combines or hardware executes and realizes that related hardware can be by two or more physical entities
It constitutes, can also be made of a physical entity.The present embodiment method can be applied to the electronic equipment with processing capacity.Its
In, electronic equipment can be the equipment such as PC, tablet computer, laptop, desktop computer.
Knowledge mapping is substantially semantic network, is a kind of data structure based on figure, can be by node (Point)
It is formed with side (Edge).In knowledge mapping, each node is indicated present in real world " entity ", each edge be entity with
" relationship " between entity.Knowledge mapping is the effective representation of relationship.Popular understanding, knowledge mapping can be all
A relational network obtained from different types of information links together.Knowledge mapping is provided to be gone point from the angle of " relationship "
The ability of analysis problem.Assuming that describing a fact with knowledge mapping: Zhang San is the father of Li Si.Here entity is Zhang San and Li
Four, relationship is father.Certainly, Zhang San and Li Si be possible to can with other people there is certain type of relationships, wouldn't examine herein
Consider.After telephone number is also added to knowledge mapping as node, telephone number is also entity, can also between people and phone
To define a kind of relationship has_phone, i.e., some telephone number belongs to someone.It can be added to using the time as attribute
Indicate to put telephone lines into operation in has_phone relationship time of number, this attribute not only may be added in relationship, can also be added to
Entity is when medium.
Resume knowledge mapping is the knowledge mapping using the relevant information architecture of resume.It is whole that resume knowledge mapping can be one
Set realizes the frame of knowledge representation and reasoning, including knowledge mapping entity, relationship, word woods (synonym, upper hyponym), vertically knows
Know that map (field profession map), knowledge maintenance module, (upper the next and equipotential reasoning inconsistent pushes away machine learning inference engine
Reason, Knowledge Discovery reasoning, Ontological concept reasoning) etc..On the one hand the inference mechanism of knowledge mapping plays auxiliary in resume parsing
Recognition reaction;On the other hand in information evaluation, the functions such as entity positioning, matching degree identification is realized, are commented for final resume
Offer support is provided.
In one embodiment, it can use the history resume assessed and generate resume knowledge mapping.It had assessed
History resume may include the resume for having applied for successful job hunter, can also include not applying for successful job hunter
Resume.The history resume assessed can be and carry out the resume after whole scoring to history resume, is also possible to for resume
It is middle one or more biographic information scored after resume.The biographic information of history resume is included at least in resume knowledge mapping
The correlation information of positions demand relative to post.Positions demand can be determined by recruitment needs and field positioning.For example, can
To include skill requirement, academic demand, length of service demand, industrial characteristic demand etc..Biographic information, which can be in resume, to be remembered
The information of record, it may for example comprise individual's description, learning experiences description, work experience description etc..Node in resume knowledge mapping with
And the relationship between node, it can configure according to demand.For example, the node in resume knowledge mapping may include post node and letter
Joint-running point etc..Post node can be used to indicate that positions demand, resume node can be used to indicate that information relevant to resume.Letter
The node gone through in knowledge mapping connects side for indicating there is incidence relation between connected node.Correlation information can be association
Degree, scoring or matching degree etc. are for evaluating the information of relevance.Exemplary, the node of resume node and post node connects the category on side
Property, it may include the property of value of the resume node relative to post node.The property of value can be by modes such as score value/degrees of association
To embody.In some examples, certain resume nodes also have the property of value.For example, the expression of some node obtains Nobel
Prize, which has the property of value, for describing the value of the node.It, can be by node when determining the correlation between node
Between connect the property of value on side to determine, can also be determined by the property of value of node.
There are many kinds of the modes for constructing resume knowledge mapping, and a kind of resume knowledge mapping construction method introduced below is described
Method may include: depth printed page analysis process and map construction process.Depth printed page analysis process includes the interdependent pass of the space of a whole page again
It is extraction process and information extraction and understanding process.
It may include: the sky using image segmentation algorithm to the history resume space of a whole page in space of a whole page dependence extraction process
Between logic region divided;Identify the area type of region content corresponding to spatial logic region, the area type packet
It includes: image, title or content of text;The logic of region content is obtained according to the geometry site between different spaces logic region
Dependence;Text region will be carried out with the region content of logic dependence and area type, and generates context dependent
Tree.
Image, title, content of text etc. are frequently included in resume, it, can be in order to which the space of a whole page to resume carries out depth analysis
It is divided using spatial logic region of the image segmentation algorithm to the history resume space of a whole page.It is a kind of resume as shown in A in Fig. 2
Schematic diagram.It can be multiple regions by the spatial logic region division of the resume space of a whole page by image segmentation algorithm, and combine space
The provincial characteristics of logic region identifies the area type of the region content corresponding to it, to distinguish to region content.Example
Such as, which spatial logic region is image, which spatial logic region is title, which spatial logic region is content of text
Deng.By the space geometry positional relationship analysis to each region, the logic dependence between each region is judged, and then will
Space geometry relationship map is at logical tree structure.For content of text, Character segmentation is carried out to content of text, can use optics
Character recognition technologies identify content of text, in conjunction with recognition result and the logic dependence and region class of region content
Type generates context dependent tree.It is the context dependent tree generated using space of a whole page dependence extracting method as shown in B in Fig. 2.In
In context dependent tree, Text region both was carried out to content of text, moreover it is possible to embody the logic dependence between different information.It can
With understanding, B is only a kind of example in Fig. 2, and context dependent tree can also exist otherwise, as long as region can be embodied
The logic dependence and particular content of content.
After obtaining context dependent tree, biographic information can be extracted from context dependent tree.Biographic information may be structure
Change information, it is also possible to unstructured information.It for structured message, can directly extract, and obtain from context dependent tree
Structured message adds it to existing resume and knows relative to the value of positions demand or the property of value of structured message
Know in map.For unstructured information, mark can be carried out to the semantic component of text sequence by Sequence Learning model.
Text fragment in resume is often unstructured information, and text fragment is usually the description to personal story and experience, because
Here sequence labelling can be the identification to event body ingredient, such as the time, place, identity, apply by behavior, target entity
Deng.The different field recognized needs to carry out different post-processing operations, for example time standard, applies by behavior normalization, real
Body inquiry and certification etc..Entity mark is carried out to the information after post-processing, and entity will have been marked is added to existing resume and known
Know map.Resume instruction map is constantly updated using the history resume largely assessed, there is resume knowledge mapping rich
Rich relationship.It is understood that when constructing resume knowledge mapping, in addition to the history resume that will have been assessed is as data source
Resume knowledge mapping is constructed, auxiliary building can also be carried out using other data sources, to obtain more perfect resume knowledge mapping.
It is understood that resume knowledge mapping can also be constructed using other means, will not repeat them here.
Next it introduces and how resume to be assessed to be assessed using resume knowledge mapping.
It may include unstructured information in resume to be assessed, can also include structured message.In one embodiment,
Resume to be assessed can be converted into context dependent tree, structured message/unstructured information of the resume to be assessed in advance
It can be obtained from context dependent tree.For example, each node to context dependent tree traverses, it can indicate non-structural to obtain
Change the first structure label of information.Context dependent tree, which can be, can embody in resume to be assessed in image, title and text
The tree of logic dependence between appearance etc..Exemplary, the generating process of the context dependent tree may include:
It is divided using spatial logic region of the image segmentation algorithm to the resume space of a whole page to be assessed;
Identify the area type of region content corresponding to spatial logic region, the area type includes: image, title
Or content of text;
The logic dependence of region content is obtained according to the geometry site between different spaces logic region;
Text region will be carried out with the region content of logic dependence and area type, and generates context dependent
Tree.
The present embodiment can realize that resume paragraph structure dependence mentions by traditional printed page analysis in conjunction with deep learning
It takes.The context dependent tree generating process that the resume knowledge mapping application stage carries out is carried out with resume knowledge mapping generation phase
Context dependent tree generating process is related, and details are not described herein.As shown in Fig. 2, providing the example that a kind of resume is transformed into dependency tree.
For unstructured information, resume knowledge mapping can use to the semanteme of unstructured information in resume to be assessed
Ingredient is labeled, and obtains the first structure label that can indicate the unstructured information.
To the purpose that the semantic component of unstructured information in resume to be assessed is labeled, semantic classification can be, with
Just first structure tag representation unstructured information can be utilized, realizes the label that unstructured information is converted into structuring, with
Just subsequent operation.In one embodiment, first structure label may include the annotation results of semantic component and semantic component.
Correspondingly, described be labeled using semantic component of the resume knowledge mapping to unstructured information in resume to be assessed, obtain
The first structure label that can indicate the unstructured information may include:
Identify the bulk composition of unstructured information in resume to be assessed, the main body cost includes: time, place, body
Part is applied by behavior;
The bulk composition identified is post-processed using resume knowledge mapping, acquisition can indicate the unstructured information
First structure label;It include treated bulk composition and its annotation results, the post-processing in the first structure label
Including time standard, apply by behavior normalization or object query and certification.
In this embodiment it is possible to the identification of event body ingredient be carried out to unstructured information, such as time, place, body
Part is applied by behavior, target entity etc..The different field recognized needs to carry out different post-processing operations, such as time standard
Change, apply by behavior normalization, object query and certification etc..Involved in this part largely with knowledge base in resume knowledge mapping
Interaction, can also use the inference engine in knowledge mapping.By taking object query and certification as an example, identification depends on semantic structure
Rather than independent of entity dictionary, the entity being tagged to submits resume knowledge mapping to navigate to relevant body node, for knowing
Know the information not having in library, new knowledge is obtained by internet.For recognizing semantic component " Hangzhou ", resume knowledge is utilized
Inference engine in map can carry out unification to semantic component.Inference engine may include logical language conversion, it is upper the next and
Reasoning of equal value, Ontological concept reasoning, Inconsistent reasoning, Knowledge Discovery reasoning etc..As it can be seen that it is not only able to achieve entity unification, for
Unknown entity name (such as company, periodical) or unknown word can be identified by the Knowledge Discovery inference engine of knowledge mapping,
No longer need the dictionaries such as manual maintenance large scale key word, synonym, physical name.For entity " Hangzhou ", annotation results can
To be " Location ".By taking unstructured information " I completes A project in W company in July, 2019 " as an example, identify to be assessed
The bulk composition of unstructured information in resume, such as: the time: 2019.7;Place: W company;Identity: I;Behavior: A are completed
Mesh.Available first structure label after progress post-processing operation: (I, algorithm expert), (W company, work unit) (completes A
Project, action).The expression-form of first structure label is not limited to this expression way, is intended merely to carry out example here.
Judge whether there is bulk composition to be present in knowledge mapping with node in first structure label using structure label, it is assumed that resume is known
Know and there is " action " this label in map, then judges whether there is " A project " from node corresponding to action
This action can be by node where node where A project in resume knowledge mapping and post B if inquiring A project
Company side value one of of the value as this unstructured information relative to post B, or by the valence of A project place node
Value one of of the value attribute as this unstructured information relative to post B.
Being labeled to the semantic component of unstructured information can be by deep learning model realization.Such as, by sequence
Model is practised to be labeled the semantic component of text sequence.Automatic marking is carried out to word sequence ingredient by deep learning model,
Realize the height precisely parsing of unstructured information;Avoid expensive artificial Rulemaking and maintenance cost.
Further, if there is no the bulk compositions in first structure label in resume knowledge mapping, pass through semantic mould
Type determines correlation information of the bulk composition relative to the positions demand in first structure label, and updates resume knowledge graph
Spectrum.For example, connecting side using the bulk composition and the positions demand as node, using the correlation information as node, more
The new resume knowledge mapping.
There is no the bulk compositions in first structure label can be in resume knowledge mapping, even utilizes Knowledge Discovery
After reasoning, the bulk composition being still not present in resume knowledge mapping can then obtain new knowledge using other modes at this time, and
It updates in resume knowledge mapping.Still by taking above-mentioned example as an example, if do not inquired from knowledge mapping by unstructured letter
The information such as the bulk composition that parses are ceased, then analyze in the unstructured information certain bulk composition relative to hilllock using semantic model
The value of position B, and by the relationship of the bulk composition and post B in unstructured information, it is added in knowledge mapping.
About target post, in the scene that job hunter delivers target post, target post can be job hunter and be delivered
Post.In the scene for recommending resume to the side of employing, target post can be the post that the side of employing is recruited.Positions demand can
It, can also be from its of the side's of employing history recruitment information or identical post to be obtained from the presently disclosed recruitment information in the side of employing
It obtains in the recruitment information of his side of employing, can specifically configure according to demand.Positions demand can be skill requirement, educational background needs
It asks, length of service demand, industrial characteristic demand etc..Further, can also according to the correlation of resume and target post to
Recommend resume etc. in the side of employing.
After obtaining the positions demand of first structure label and target post of unstructured information, first can be passed through
Structure label obtains correlation information of the unstructured information relative to the positions demand from resume knowledge mapping, and
Correlation of the resume to be assessed relative to target post is determined using the correlation information as evaluation factor.
Wherein, first structure label may include that treated bulk composition and its annotation results.Certain bulk compositions can
It can be entity.Correlation information of the unstructured information relative to positions demand, can be based on entity institute in first structure label
It is obtained in the relationship of node where node and positions demand, it can also be based on the valence of node where entity in first structure label
Value attribute obtains.May there are an entity or multiple entities to be present in resume knowledge mapping in one first structure label, it can
With using the value for all entities being present in resume knowledge mapping in first structure label or with the relationship of positions demand,
The correlation information of the first structure label relative to positions demand is obtained, which can be considered unstructured
Correlation information of the information relative to the positions demand, and then can be corresponding by unstructured information each in resume to be assessed
Correlation information determines correlation of the resume to be assessed relative to target post as evaluation factor.
In one embodiment, resume to be assessed is other than unstructured information, it is also possible to including structured message, in view of
This, correlation correlation also with the structured message relative to the positions demand of the resume to be assessed relative to target post
For property information as evaluation factor, the structured message can be from resume knowledge relative to the correlation information of the positions demand
It is obtained in map.
In this embodiment, both using unstructured information relative to positions demand correlation information as evaluation factor,
Again using structured message relative to positions demand correlation information be used as evaluation factor, realization comprehensive assessment.
It is exemplary, the structured message relative to the positions demand correlation information using the second structure label from
It is obtained in resume knowledge mapping, the second structure label pushes away structured message and its title using inference engine of equal value
It is obtained after reason.For example, structured message can also be obtained from dependency tree, traversed to the progress of context dependent tree each node
Cheng Zhong obtains the corresponding second structure label of structured message.
The structured messages such as educational background, paper, Patent Output can be extracted using the methods of template matching.It utilizes
Valence inference engine makes inferences structured message and its title, and information unification may be implemented, to obtain the second structure label.
Entity corresponding to structured message can be quickly navigated to from knowledge mapping using the second structure label, to obtain structure
Change correlation information of the information relative to positions demand.
Structured message such as educational background, paper, Patent Output and part unstructured information such as text fragment parse
Critical entities title, according to the correlation of knowledge mapping and value reasoning, be converted into scoring.Remaining unstructured letter
Breath, such as description to project participation process can also be encoded by semantic model, it is intended to be carried out auxiliary to the value of event and be commented
Valence.Final assessment models provide resume according to above-mentioned all information and integrally score.Wherein, all assessment models can be according to going through
History resume and history evaluation training obtain.
Various technical characteristics in embodiment of above can be arbitrarily combined, as long as the combination between feature is not present
Conflict or contradiction, but as space is limited, it is not described one by one, therefore the various technical characteristics in above embodiment is any
It is combined the range for also belonging to this disclosure.
It is illustrated below with one of which combination.
As shown in figure 3, providing a kind of architecture diagram of the system of knowledge based map progress resume assessment.The system can be with
Including three parts: depth printed page analysis module, information integration and evaluation module and knowledge mapping.This specification is by knowledge graph
Spectrum is applied to automatic resume and assesses, and emphasizes its important function in the whole process.The inference mechanism of knowledge mapping is on the one hand
Play the role of assisting in identifying in space of a whole page parsing;On the other hand in information evaluation, entity positioning, matching degree identification etc. are realized
Function provides support for final resume assessment.Depth printed page analysis module is divided into two submodules according to the function of realization again:
Space of a whole page dependence extracting sub-module, information extraction and understand submodule.Space of a whole page dependence extracting sub-module can pass through sky
Between logic region divide, area type identification, geometrical relationship to the hands such as logical relation structure mapping, Character segmentation, character recognition
Section obtains context dependent tree.Information extraction and understand that submodule can use knowledge mapping and structured message is converted to the second knot
Text fragment in unstructured information is carried out semantic component mark, to obtain first structure label by structure label.In resume
It may include inference engine, knowledge acquisition, knowledge maintenance, knowledge base etc. in knowledge mapping.Inference engine may include logic again
Language conversion, upper the next and of equal value reasoning, Inconsistent reasoning, Ontological concept reasoning, Knowledge Discovery reasoning etc..Logical language conversion
For carrying out the conversion between different language.Upper bottom reasoning can be used to implement hyponym reasoning.Ontological concept reasoning can be with
It is the ontology for deriving identical meanings.Inconsistent reasoning, which can be, derives different entities.Knowledge Discovery reasoning is to utilize node
And node relationships infer new node or node relationships.Knowledge acquisition can be from internet and obtain knowledge, can also be by different
Structure Knowledge conversion obtains.It is, of course, also possible to be that other modes obtain.Knowledge maintenance include knowledge base maintenance, index maintenance and
Collection and self study, to realize the maintenance of map.It may include entity, relationship, upper the next, word woods in knowledge base and vertically know
Know library.Vertical repository can be field profession map.In information integration with evaluation module, resume knowledge mapping can use
Relational query is carried out to first structure label, the second structure label and positions demand and correlation is abstract, and effectively by other
Semantic text inputs semantic information encoding model, carries out auxiliary evaluation with the value to event, in summary information, may be implemented
Assessment to resume to be assessed.This specification embodiment proposes the concept of depth printed page analysis, by traditional printed page analysis and depth
Study combines, (general for the complex information region that can not be handled in previous resume parsing and so-called resume assessment technology
Based on big segment description text) problem, unified with nature language understanding and knowledge mapping technology carry out secondary analysis, greatly promote
Comprehensive and the degree of automation of resume assessment.It proposes to apply to knowledge mapping into resume parsing and automatic assessment for the first time
In, and it is demonstrated for the importance and validity of information extraction and understanding.
Corresponding with the aforementioned knowledge based map progress embodiment of method of resume assessment, this specification additionally provides base
The device of resume assessment and its embodiment of applied electronic equipment are carried out in knowledge mapping.
The embodiment that this specification knowledge based map carries out the device of resume assessment can be applied in computer equipment.Dress
Setting embodiment can also be realized by software realization by way of hardware or software and hardware combining.It is implemented in software to be
Example, as the device on a logical meaning, being will be in nonvolatile memory by the processor of computer equipment where it
Corresponding computer program instructions are read into memory what operation was formed.For hardware view, as shown in figure 4, being this explanation
A kind of hardware structure diagram of computer equipment where book knowledge based map carries out the device of resume assessment, in addition to shown in Fig. 4
Except processor 410, network interface 420, memory 430 and nonvolatile memory 440, knowledge based map in embodiment
The computer equipment at 431 place of device of resume assessment is carried out generally according to the actual functional capability of the equipment, can also include other
Hardware repeats no more this.
As shown in figure 5, being a kind of this specification knowledge based map progress resume shown according to an exemplary embodiment
The block diagram of the device of assessment advances with the history resume assessed and generates resume knowledge mapping, the resume knowledge mapping
In include history resume correlation information of the biographic information relative to positions demand, include unstructured letter in resume to be assessed
Breath, described device include:
Information labeling module 52, for using resume knowledge mapping in resume to be assessed unstructured information it is semantic at
Divide and be labeled, obtains the first structure label that can indicate the unstructured information;
Demand obtains module 54, for obtaining the positions demand of target post;
Resume evaluation module 56, for obtaining the unstructured letter from resume knowledge mapping by first structure label
Manner of breathing and determines resume to be assessed using the correlation information as evaluation factor for the correlation information of the positions demand
Correlation relative to target post.
In one embodiment, the information labeling module 52 is specifically used for:
Identify the bulk composition of unstructured information in resume to be assessed, the main body cost includes: time, place, body
Part is applied by behavior;
The bulk composition identified is post-processed using resume knowledge mapping, acquisition can indicate the unstructured information
First structure label;It include treated bulk composition and its annotation results, the post-processing in the first structure label
Including time standard, apply by behavior normalization or object query and certification.
In one embodiment, the unstructured information of the resume to be assessed is obtained from context dependent tree, the dress
Setting further includes dependency tree generation module (Fig. 5 is not shown), is used for:
It is divided using spatial logic region of the image segmentation algorithm to the resume space of a whole page to be assessed;
Identify the area type of region content corresponding to spatial logic region, the area type includes: image, title
Or content of text;
The logic dependence of region content is obtained according to the geometry site between different spaces logic region;
Text region will be carried out with the region content of logic dependence and area type, and generates context dependent
Tree.
In one embodiment, described device further includes map update module (Fig. 5 is not shown), is used for:
If determining first by semantic model there is no the bulk composition in first structure label in resume knowledge mapping
Correlation information of the bulk composition relative to the positions demand in structure label, and needed with the bulk composition and the post
It asks as node, using the correlation information as node and connects side, update the resume knowledge mapping.
In one embodiment, the biographic information of the resume to be assessed further includes structured message, resume phase to be assessed
For target post correlation also using the structured message relative to the positions demand correlation information as assessment
The factor, the structured message are obtained from resume knowledge mapping relative to the correlation information of the positions demand.
Further, the structured message is to utilize the second structure mark relative to the correlation information of the positions demand
Label are obtained from resume knowledge mapping, the second structure label using inference engine of equal value to structured message and its title into
It is obtained after row reasoning.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The module of explanation may or may not be physically separated, and the component shown as module can be or can also be with
It is not physical module, it can it is in one place, or may be distributed on multiple network modules.It can be according to actual
The purpose for needing to select some or all of the modules therein to realize this specification scheme.Those of ordinary skill in the art are not
In the case where making the creative labor, it can understand and implement.
Correspondingly, this specification embodiment also provides a kind of computer equipment, including memory, processor and it is stored in
On reservoir and the computer program that can run on a processor, wherein the processor realizes above-mentioned when executing described program
The method that one knowledge based map carries out resume assessment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for equipment reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
Correspondingly, this specification embodiment also provides a kind of computer storage medium, journey is stored in the storage medium
Sequence instruction, described program instruction realize that knowledge based map described in any of the above-described carries out resume assessment when being executed by processor
Method.
This specification embodiment can be used one or more wherein include the storage medium of program code (including but not
Be limited to magnetic disk storage, CD-ROM, optical memory etc.) on the form of computer program product implemented.Computer is available to be deposited
Storage media includes permanent and non-permanent, removable and non-removable media, can be accomplished by any method or technique letter
Breath storage.Information can be computer readable instructions, data structure, the module of program or other data.The storage of computer is situated between
The example of matter includes but is not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory
Device (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), the read-only storage of electrically erasable
Device (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM), digital versatile disc
(DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices or any other non-biography
Defeated medium, can be used for storage can be accessed by a computing device information.
Those skilled in the art will readily occur to this specification after considering specification and practicing the invention applied here
Other embodiments.This specification is intended to cover any variations, uses, or adaptations of this specification, these modifications,
Purposes or adaptive change follow the general principle of this specification and do not apply in the art including this specification
Common knowledge or conventional techniques.The description and examples are only to be considered as illustrative, the true scope of this specification and
Spirit is indicated by the following claims.
It should be understood that this specification is not limited to the precise structure that has been described above and shown in the drawings,
And various modifications and changes may be made without departing from the scope thereof.The range of this specification is only limited by the attached claims
System.
The foregoing is merely the preferred embodiments of this specification, all in this explanation not to limit this specification
Within the spirit and principle of book, any modification, equivalent substitution, improvement and etc. done should be included in the model of this specification protection
Within enclosing.