CN106022599A - Industrial design talent level evaluation method and system - Google Patents
Industrial design talent level evaluation method and system Download PDFInfo
- Publication number
- CN106022599A CN106022599A CN201610329208.7A CN201610329208A CN106022599A CN 106022599 A CN106022599 A CN 106022599A CN 201610329208 A CN201610329208 A CN 201610329208A CN 106022599 A CN106022599 A CN 106022599A
- Authority
- CN
- China
- Prior art keywords
- talent
- classification
- ability
- industrial design
- vector
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an industrial design talent level evaluation method and system. The method comprises the steps: enabling a computer system to automatically extract the attribute feature of an industrial design talent; carrying out the field dividing of the talent through employing a computer automatic classification algorithm, and determining the industrial design area where the talent belongs; carrying out the calculation of the cost of the talent according to a talent classification result and the attribute feature of the talent; carrying out the design capability matching calculation of the talent; and recommending the talent to a specific demanding user according to the cost of the talent and a talent capability matching calculation result. The method carries out the machine reading and analysis of large-scale talent data, carries out the matching of the talent and the market demands and the intelligent recommendation of the talent through employing an algorithm, builds a talent evaluation mode based on the algorithm and the technology, and can save a large amount of manpower cost.
Description
Technical field
The present invention relates to a kind of horizontal evaluating method of the talent and system, belong to information automation process field, be specifically related to
A kind of horizontal evaluating method of the industrial design talent and system.
Background technology
Industrial design is one emerging application-directed specialty of China.In recent years, along with government and enterprise are to industrial design ground
Position and the great attention of effect, the most therefore industrial design has obtained unprecedented development.Industrial design production is carried out in enterprise
During generally require to introduce there is the industrial design talent of certain capabilities level, need the talent is carried out effectively, fixed accurately
Position, it is a requisite link that the talent screens evaluation and test.
Industrial design talent's Level Evaluation relates to all many-sides and link, different with evaluation and test content according to evaluating object, deposits
In multiple evaluating method and the selection of standard, though for same evaluating object, evaluation and test content, it is also possible to due to evaluation and test main body,
The difference of the aspects such as evaluation and test time and space, causes the difference of evaluation result.
Current existing talent assessment system, Evaluation index system, the technical method of test and appraisal all tend to simplifying, and pay attention to achievement
Effect is checked and rated and is ignored quality assessment, especially for large-scale talents information data, lacks effective process, lacks intellectuality
, the talent's Level Evaluation automated process for industrial design field of automatization and system.
The present invention combines text mining and big data processing technique and processes large-scale talents information data, based on
Efficient machine capability and clear and definite parser, set big data and artificial intelligence as instrument, the industry setting up automatization
Meter talent's evaluating method and system, it is achieved the talent classifies automatically, price, ability are mated and intelligent recommendation automatically.
Summary of the invention
The present invention mainly solves effectively locating for large-scale talents information data deficiency existing for prior art
Manage and lack the technology for talent's Level Evaluation method and system in industrial design field intelligentized, automatization and ask
Topic, it is proposed that a kind of horizontal evaluating method of the industrial design talent and system.The method and system can be for large-scale industry
Design field talent's data carry out the process of automatization, provide talent's evaluation result of quantization；Simultaneously can be according to evaluation result
By talent recommendation to the user of particular demands, set up incidence relation between the talent and demand user, quick, effective；This
Outward, moreover it is possible to according to the variation characteristic of user's request so that it is have Active Learning ability, during improving talent's evaluation and test, the talent belongs to
Property extraction, classification of qualified personnel, talent recommendation parameter arrange so that the talent evaluation and test and recommend more accurate
The above-mentioned technical problem of the present invention is mainly addressed by following technical proposals:
A kind of horizontal evaluating method of the industrial design talent, comprises the following steps:
Characteristic extraction step, for obtaining from talents information data according to the talent's attributes extraction rule pre-build
Industrial design talent's attribute character；
Field partiting step, is used for utilizing sorting algorithm that the talent is carried out field division, determines industrial design belonging to the talent
Category；
Calculation of price step, for calculating talent's price according to industrial design category belonging to the talent and talent's attribute character；
Ability coupling step, for designing energy according to the talent ability Matching Elements COMPREHENSIVE CALCULATING reflection talent pre-build
The talent ability matching value of power；
Talent recommendation step, for recommending talents information according to talent's price and talent ability matching value to user.
Optimizing, above-mentioned a kind of horizontal evaluating method of the industrial design talent, in described characteristic extraction step: utilize text
Participle technique obtains talent's attribute character from talents information data；Described talents information data include: personnel resume, Ren Caiwen
Volume, the talent's logging in log information in systems one or more；Described talent's attribute character includes: the age, sex,
One or more in district, marital status, educational background, specialty, graduation universities and colleges, length of service.
Optimize, above-mentioned a kind of horizontal evaluating method of the industrial design talent, in the partiting step of described field, build the talent
Attribute feature vector, be subordinate to based on layering and the talent classified, determine its affiliated industrial design category.
Optimize, above-mentioned a kind of horizontal evaluating method of the industrial design talent, in described calculation of price step, set the talent
Price range 0-A and Dynamic gene B, calculate talent's price lishu according to base attribute interval, carry out the talent in conjunction with Dynamic gene
Calculation of price；Wherein talent's Dynamic gene is according to including that macroeconomic environment, industry average salary level, talent's relation between supply and demand exist
One or more interior factors periodically adjust；Described base attribute includes: age, sex, area, marital status, educational background, specially
Industry, one or more graduated in universities and colleges, length of service.Talent's calculation of price is adopted and is calculated with the following method:
Wherein k represents the talent's number of attributes chosen, αkWeight for each attribute
Value, meets 0≤αk≤ 1, simple method is to take 1/k, nkFor talent bank has talent's number of identical same alike result k with talent P
Amount, β is talent's price adjustment factor.These computational methods are designed relative to talent's quantity based on extensive talent's data.
Optimize, above-mentioned a kind of horizontal evaluating method of the industrial design talent, in described ability coupling step, use comprehensive
Pass judgment on computational methods and calculate the ability matching value of the talent according to the talent ability Matching Elements pre-build；The described talent mates energy
Power Matching Elements includes primary ability Matching Elements；Described primary ability Matching Elements includes: Knowledge Capability coupling, experience ability
One or more in coupling, Quality ability coupling；Described Comprehensive Evaluation computational methods include that fuzzy comprehensive evoluation, main constituent divide
One or more in analysis, step analysis；Described ability matching value includes comprehensive matching ability value and the ability of each subitem key element
Matching value.
In order to solve the problems referred to above, according to a further aspect in the invention, it is provided that a kind of industrial design talent's level evaluation and test
System, including:
Characteristic extracting module, for obtaining from talents information data according to the talent's attributes extraction rule pre-build
Industrial design talent's attribute character；
Field divides module, is used for utilizing sorting algorithm that the talent is carried out field division, determines industrial design belonging to the talent
Category；
Calculation of price module, for calculating talent's price according to industrial design category belonging to the talent and talent's attribute character；
Ability matching module, for designing energy according to the talent ability Matching Elements COMPREHENSIVE CALCULATING reflection talent pre-build
The talent ability matching value of power；
Talent recommendation module, for recommending talents information according to talent's price and talent ability matching value to user.
Optimizing, above-mentioned a kind of horizontal evaluating system of the industrial design talent, in described characteristic extracting module: utilize text
Participle technique obtains talent's attribute character from talents information data；Described talents information data include: personnel resume, Ren Caiwen
Volume, the talent's logging in log information in systems one or more；Described talent's attribute character includes: the age, sex,
One or more in district, marital status, educational background, specialty, graduation universities and colleges, length of service.
Optimizing, above-mentioned a kind of horizontal evaluating system of the industrial design talent, described field divides in module, builds the talent
Attribute feature vector, utilize layering to be subordinate to and the talent classified, determine its affiliated industrial design category.
Optimize, above-mentioned a kind of horizontal evaluating system of the industrial design talent, in described calculation of price module, set the talent
Price range 0-A and Dynamic gene B, calculate talent's price lishu according to base attribute interval, carry out the talent in conjunction with Dynamic gene
Calculation of price；Wherein talent's Dynamic gene is according to including that macroeconomic environment, industry average salary level, talent's relation between supply and demand exist
One or more interior factors periodically adjust；Described base attribute includes: age, sex, area, marital status, educational background, specially
Industry, one or more graduated in universities and colleges, length of service.
Optimize, above-mentioned a kind of horizontal evaluating system of the industrial design talent, in described ability matching module, use comprehensive
Pass judgment on computational methods and calculate the ability matching value of the talent according to the talent ability Matching Elements pre-build；The described talent mates energy
Power Matching Elements includes primary ability Matching Elements；Described primary ability Matching Elements includes: Knowledge Capability coupling, experience ability
One or more in coupling, Quality ability coupling；Described Comprehensive Evaluation computational methods include that fuzzy comprehensive evoluation, main constituent divide
One or more in analysis, step analysis；Described ability matching value includes comprehensive matching ability value and the ability of each subitem key element
Matching value.
Therefore, present invention have the advantage that and extensive talent's data are carried out machine reading and analysis, it is right to come with algorithm
The talent and the market demand carry out mating and intelligent recommendation, set up and evaluate and test pattern based on algorithm and the technical talent, it is possible to save
Substantial amounts of human cost.
Accompanying drawing explanation
Fig. 1 is the workflow schematic diagram of embodiment of the present invention；
Fig. 2 is the system hierarchy figure of the present invention.
Fig. 3 is the classification of qualified personnel flow chart of the present invention.
Detailed description of the invention
Below by embodiment, and combine accompanying drawing, technical scheme is described in further detail.
Embodiment 1:
Fig. 1 is the workflow schematic diagram of embodiment of the present invention, and its implementation process is as follows:
Step one: system read talent's data, as personnel resume, talent's questionnaire, the talent in systems log in daily record etc.
Content of text messages.System applicating text participle technique extracts the attribute character of the talent, including: age, sex, area, wedding
The base attributes such as relation by marriage situation, educational background, specialty, graduation universities and colleges, length of service.
Step 2: application layering is subordinate to classifies to the talent, determines its affiliated industrial design category, industrial design model
Farmland includes following classification.1. planar design: photography design, package design, typesetting designs, font design, CI Design, boundary
Face is designed, interaction design, Vision Design, image procossing, and Digital Media designs, content design, printed design, publishes design, extensively
Accusing design, website design, Experience design, propaganda material designs, card design, Exhibition Design, and calligraphy designs, Visual Communication Design.
2. Fashion Design: fabric pattern design, design of plus material, Fashion Design, appearance designs, design, material design, and jewel design is worn
Wear design.3. product design: Environmental Ceramic Design, Car design, furniture design, house ornamentation designs, Home Fashion & Design Shanghai, and household electrical appliances design, toy
Design, game design, design of electronic products, artware designs, kitchenware design, and gift designs, and energy products design, and yacht sets
Meter, dresses design, and disposable products design, product design.4. Environment Design: exhibition designs, indoor design, city is advised
Drawing, urban design, landscape design, landscape design, architectural design, structure designs, Lighting Design, agricultural design, traffic design.⑤
Engineering design: tool design, design of material, electronic engineering design, structure engineering design, design of electronic products, area of computer aided
Design, computer aided manufacturing, 3D models, and scientific instrument designs, and armarium designs, and research equipment designs, Machine Design, boat
Empty aerospace design, biological engineering designs, 3D printed design, product structure design.
Concrete, classification of qualified personnel is comprised the following steps:
1, industrial design hierarchical classification and each stratigraphic classification characteristic vector thereof are set up.
A, the industrial design classifying text of lowest hierarchical level is described, describe as atom.Then carry out participle, go to stop at word
Reason.
B, set up lowest hierarchical level classification upper level characteristic of division vector.Each atom is described as a KNN algorithm
In one " training sample ", after feature selection, each atom words of description vector constitutes in this hierarchy characteristic vector space
Individual point.With { V1,V2…Vn}。
C, set up two-stage on lowermost layer frame category feature vector.Lowest hierarchical level is classified upper level all atoms describe
Merge, as one " training sample ", after feature selection, constitute big class hierarchy characteristic vector.With { V1 1,V1 2…V1 nRepresent.
D, set up on lowermost layer frame three grades category feature vector.All atoms of upper for lowest hierarchical level classification two-stage are described
Merge, as " training " sample, constituting portion level characteristics vector after feature selection.With { V2 1,V2 2…V2 nRepresent.
By that analogy, the characteristic vector { V of the classification of each level is builtm 1,Vm 2…Vm n, m is classification level grade, and n is each
Hierarchical category quantity.
2, industrial design category feature vector corrected training.
A, read in training talent text data X, participle, go to stop word, feature selection.
B, the classification of distribution highest level.If X is not null vector, calculate X and { Vm 1,Vm 2…Vm nThe phase of each vector in }
Like degree:I.e. included angle cosine.Calculate X belong to highest level classification degree of membership:
ClassjFor specific name
The talent is assigned in that classification that weight is maximum.If distribution is correct, then proceed to next stage classification, record every
The wrong divided data of each classification of individual level, after these data reach some, the center vector calculating them joins phase
Answering in level respective classes vector, the characteristic vector of industrial design classification is revised, can be as the basis of classification later.
Step 3: carry out talent's calculation of price according to classification of qualified personnel result and talent's attribute character.Set talent's price district
Between 0-100 and Dynamic gene, according to the base such as age, sex, area, marital status, educational background, specialty, graduation universities and colleges, length of service
This property calculation talent's price lishu is interval, carries out talent's calculation of price in conjunction with Dynamic gene.
Concrete, talent's calculation of price is adopted and is calculated with the following method:
Wherein k represents the talent's number of attributes chosen, αkWeight for each attribute
Value, meets 0≤αk≤ 1, simple method is to take 1/k, nkFor talent bank has talent's number of identical same alike result k with talent P
Amount, β is talent's price adjustment factor.These computational methods are designed relative to talent's quantity based on extensive talent's data.
Step 4: use step analysis to carry out talent ability matching primitives, use two-stage index.
A Knowledge Capability is mated.Specifically include: 1. industrial design theoretical basis: Summary for Design, design history, industrial design are thought
Think basis, value engineering, ergonomics.2. plastic arts basis: Design Sketch, design color, design literary sketch, design base
Plinth.3. Aesthetic Basis: design esthetics, the at home and abroad history of arts and craft.4. thinking and methodology basis: design methodology, THE CREATIVE SCIENCE, figure
Shape creative design, the procedures of product design and method.5. basis is expressed in design: have an X-rayed with chart, product design quickly shows, calculate
Machine Aided Design, product model make.6. engineering basis: Fundamentals of Machine Design, Electrical Technology Basis Held, material and molding work
Skill, advanced manufacturing technology.7. psychology basis: design psychology, consumer psychology.8. humanistic community basis: design philosophy,
Design sociology, modern science and technology outline, Patent Law and design patent application.9. market and management basis: design and use
Research, design management, the marketing.
B experience ability is mated.Specifically include: 1. design information obtains and disposal ability.2. computer professional software application
Ability.3. technical expression ability.4. specialized aesthetic ability.5. Design Thinking ability.6. cooperation ability.7. product comprehensively sets
Meter ability.8. design object planning and operating capability.
C Quality ability mates.Specifically include: 1. ideological and ethical standard.2. professionalism.3. humanistic quality.4. health element
Matter.5. psychological diathesis.
Step 5: give specific demand user by talent recommendation according to talent's price and talent ability matching primitives result.
Step 6: talent's attribute identification, classification of qualified personnel, talent recommendation relevant parameter are carried out excellent according to historical feedback result
Change so that evaluation and test and recommendation to the talent are more accurate.
Embodiment 2:
Fig. 2 is the system hierarchy figure of the present invention.System includes the automatic sort module of sample training module, the talent, people
Just calculation of price module, talent's designed capacity matching module, talent recommendation module, parameters revision module.
Sample training module, for automatically classifying by sample data training talent's attribute identification, the talent, talent recommendation.
The automatic sort module of the talent, for carrying out classification process by the disaggregated model trained to the talent.
Talent's calculation of price module, for calculating the price of the talent.
Talent's designed capacity matching module, for calculating the talent's competent ability to all kinds of design scenario, with the number quantified
Value is represented.
Talent recommendation module, uses proposed algorithm by talent recommendation to the user of particular demands.
Parameters revision module, is optimized according to the parameter of the housing choice behavior of user module each to system so that the talent belongs to
Property extraction, classification of qualified personnel, talent recommendation are more accurate.
Specific embodiment described herein is only to present invention spirit explanation for example.Technology neck belonging to the present invention
Described specific embodiment can be made various amendment or supplements or use similar mode to replace by the technical staff in territory
Generation, but without departing from the spirit of the present invention or surmount scope defined in appended claims.
Claims (10)
1. the horizontal evaluating method of the industrial design talent, it is characterised in that comprise the following steps:
Characteristic extraction step, for obtaining industry according to the talent's attributes extraction rule pre-build from talents information data
Designing talents attribute character；
Field partiting step, is used for utilizing sorting algorithm that the talent is carried out field division, determines industrial design category belonging to the talent；
Calculation of price step, for calculating talent's price according to industrial design category belonging to the talent and talent's attribute character；
Ability coupling step, for according to the talent ability Matching Elements COMPREHENSIVE CALCULATING reflection talent's designed capacity pre-build
Talent ability matching value；
Talent recommendation step, for recommending talents information according to talent's price and talent ability matching value to user.
A kind of horizontal evaluating method of the industrial design talent the most according to claim 1, it is characterised in that described feature extraction
In step: utilize text participle technique to obtain talent's attribute character from talents information data；Described talents information data include:
One or more of personnel resume, talent's questionnaire, the talent logging in log information in systems；Described talent's attribute character bag
Include: one or more in age, sex, area, marital status, educational background, specialty, graduation universities and colleges, length of service.
A kind of horizontal evaluating method of the industrial design talent the most according to claim 1, it is characterised in that described field divides
In step, building the attribute feature vector of the talent, that classifies the talent specifically comprises the following steps that
Step 1, sets up industrial design hierarchical classification and each stratigraphic classification characteristic vector thereof, including following sub-step:
Step 101, describes the industrial design classifying text of lowest hierarchical level, describes as atom, then carries out participle, goes to stop word
Process；
Step 102, sets up the characteristic of division vector of lowest hierarchical level classification upper level, is described by each atom and calculates as a KNN
One " training sample " in method, each atom words of description vector V after feature selectionnConstitute in this hierarchy characteristic vector space
A point, with { V1,V2…VnDescribe；
Step 103, sets up the category feature vector of two-stage on lowermost layer frame, and all atoms of upper level of lowest hierarchical level being classified are retouched
State merging, as one " training sample ", after feature selection, constitute big class hierarchy characteristic vector Vn 1, with { V1 1,V1 2…V1 nTable
Show；
Step 104, sets up the category feature vector of on lowermost layer frame three grades, is retouched by all atoms of upper for lowest hierarchical level classification two-stage
State merging, as " training " sample, constituting portion level characteristics vector V after feature selectionn 2, with { V2 1,V2 2…V2 nRepresent；
Step 105, by that analogy, builds the characteristic vector { V of the classification of each levelm 1,Vm 2…Vm n, m is classification level grade, n
For each hierarchical category quantity.
Step 2, the training of industrial design category feature vector corrected, specifically include:
Step 201, read in training talent text data X, participle, go to stop word, feature selection；
Step 202, the classification of distribution highest level.If X is not null vector, calculate X and { Vm 1,Vm 2…Vm nEach vector in }
Similarity:I.e. included angle cosine, calculate X belong to highest level classification degree of membership:
ClassjFor specific name
The talent is assigned in that classification that weight is maximum, if distribution is correct, then proceeds to next stage classification, record each level
The wrong divided data of each classification, after these data reach some, the center vector calculating them joins corresponding level
In respective classes vector, the characteristic vector of industrial design classification is revised, can be as the basis of classification later.
A kind of horizontal evaluating method of the industrial design talent the most according to claim 1, it is characterised in that described calculation of price
In step, set talent's price range 0-A and Dynamic gene B, calculate talent's price lishu according to base attribute interval, in conjunction with adjusting
Integral divisor carries out talent's calculation of price；Wherein talent's Dynamic gene according to include macroeconomic environment, industry average salary level,
Talent's relation between supply and demand periodically adjusts in one or more interior factors；Described base attribute includes: age, sex, area, marriage
One or more in situation, educational background, specialty, graduation universities and colleges, length of service；
Further, described talent's calculation of price is based on following formula
Wherein k represents the talent's number of attributes chosen, αkFor the weighted value of each attribute, meet 0≤αk≤ 1, simple method is
Take 1/k, nkFor having talent's quantity of same alike result k in talent bank with talent P, β is talent's price adjustment factor.This calculating side
Method is designed relative to talent's quantity based on extensive talent's data.
A kind of horizontal evaluating method of the industrial design talent the most according to claim 1, it is characterised in that described ability is mated
In step, Comprehensive Evaluation computational methods are used to calculate the ability coupling of the talent according to the talent ability Matching Elements pre-build
Value；Described talent's matching capacity Matching Elements includes primary ability Matching Elements；Described primary ability Matching Elements includes: knowledge
One or more in ability coupling, experience ability coupling, Quality ability coupling；Described Comprehensive Evaluation computational methods include obscuring
One or more in Comprehensive Evaluation, principal component analysis, step analysis；Described ability matching value include comprehensive matching ability value and
The ability matching value of each subitem key element.
6. the horizontal evaluating system of the industrial design talent, it is characterised in that including:
Characteristic extracting module, for obtaining industry according to the talent's attributes extraction rule pre-build from talents information data
Designing talents attribute character；
Field divides module, is used for utilizing sorting algorithm that the talent is carried out field division, determines industrial design category belonging to the talent；
Calculation of price module, for calculating talent's price according to industrial design category belonging to the talent and talent's attribute character；
Ability matching module, for according to the talent ability Matching Elements COMPREHENSIVE CALCULATING reflection talent's designed capacity pre-build
Talent ability matching value；
Talent recommendation module, for recommending talents information according to talent's price and talent ability matching value to user.
A kind of horizontal evaluating system of the industrial design talent the most according to claim 6, it is characterised in that described feature extraction
In module: utilize text participle technique to obtain talent's attribute character from talents information data；Described talents information data include:
One or more of personnel resume, talent's questionnaire, the talent logging in log information in systems；Described talent's attribute character bag
Include: one or more in age, sex, area, marital status, educational background, specialty, graduation universities and colleges, length of service.
A kind of horizontal evaluating system of the industrial design talent the most according to claim 6, it is characterised in that described field divides
Module performs following step and classifies the talent, determines its affiliated industrial design category:
Step 1, sets up industrial design hierarchical classification and each stratigraphic classification characteristic vector thereof, including following sub-step:
Step 101, describes the industrial design classifying text of lowest hierarchical level, describes as atom, then carries out participle, goes to stop word
Process；
Step 102, sets up the characteristic of division vector of lowest hierarchical level classification upper level, is described by each atom and calculates as a KNN
One " training sample " in method, each atom words of description vector V after feature selectionnConstitute in this hierarchy characteristic vector space
A point, with { V1,V2…VnDescribe；
Step 103, sets up the category feature vector of two-stage on lowermost layer frame, and all atoms of upper level of lowest hierarchical level being classified are retouched
State merging, as one " training sample ", after feature selection, constitute big class hierarchy characteristic vector Vn 1, with { V1 1,V1 2…V1 nTable
Show；
Step 104, sets up the category feature vector of on lowermost layer frame three grades, is retouched by all atoms of upper for lowest hierarchical level classification two-stage
State merging, as " training " sample, constituting portion level characteristics vector V after feature selectionn 2, with { V2 1,V2 2…V2 nRepresent；
Step 105, by that analogy, builds the characteristic vector { V of the classification of each levelm 1,Vm 2…Vm n, m is classification level grade, n
For each hierarchical category quantity.
Step 2, the training of industrial design category feature vector corrected, specifically include:
Step 201, read in training talent text data X, participle, go to stop word, feature selection；
Step 202, the classification of distribution highest level.If X is not null vector, calculate X and { Vm 1,Vm 2…Vm nEach vector in }
Similarity:I.e. included angle cosine, calculate X belong to highest level classification degree of membership:
ClassjFor specific name
The talent is assigned in that classification that weight is maximum, if distribution is correct, then proceeds to next stage classification, record each level
The wrong divided data of each classification, after these data reach some, the center vector calculating them joins corresponding level
In respective classes vector, the characteristic vector of industrial design classification is revised, can be as the basis of classification later.
A kind of horizontal evaluating system of the industrial design talent the most according to claim 6, it is characterised in that described calculation of price
In module, set talent's price range 0-A and Dynamic gene B, calculate talent's price lishu according to base attribute interval, in conjunction with adjusting
Integral divisor carries out talent's calculation of price；Wherein talent's Dynamic gene according to include macroeconomic environment, industry average salary level,
Talent's relation between supply and demand periodically adjusts in one or more interior factors；Described base attribute includes: age, sex, area, marriage
One or more in situation, educational background, specialty, graduation universities and colleges, length of service；
Further, described talent's calculation of price is based on following formula
Wherein k represents the talent's number of attributes chosen, αkFor the weighted value of each attribute, meet 0≤αk≤ 1, simple method is
Take 1/k, nkFor having talent's quantity of same alike result k in talent bank with talent P, β is talent's price adjustment factor.This calculating side
Method is designed relative to talent's quantity based on extensive talent's data.
A kind of horizontal evaluating system of the industrial design talent the most according to claim 6, it is characterised in that described ability
Join in module, use Comprehensive Evaluation computational methods to calculate the ability coupling of the talent according to the talent ability Matching Elements pre-build
Value；Described talent's matching capacity Matching Elements includes primary ability Matching Elements；Described primary ability Matching Elements includes: knowledge
One or more in ability coupling, experience ability coupling, Quality ability coupling；Described Comprehensive Evaluation computational methods include obscuring
One or more in Comprehensive Evaluation, principal component analysis, step analysis；Described ability matching value include comprehensive matching ability value and
The ability matching value of each subitem key element.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610329208.7A CN106022599A (en) | 2016-05-18 | 2016-05-18 | Industrial design talent level evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610329208.7A CN106022599A (en) | 2016-05-18 | 2016-05-18 | Industrial design talent level evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106022599A true CN106022599A (en) | 2016-10-12 |
Family
ID=57098667
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610329208.7A Pending CN106022599A (en) | 2016-05-18 | 2016-05-18 | Industrial design talent level evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106022599A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107977794A (en) * | 2017-12-14 | 2018-05-01 | 方物语（深圳）科技文化有限公司 | Data processing method, device, computer equipment and the storage medium of industrial products |
| CN108596420A (en) * | 2018-02-02 | 2018-09-28 | 武汉文都创新教育研究院（有限合伙） | A kind of talent assessment system and method for Behavior-based control |
| CN109324963A (en) * | 2018-08-22 | 2019-02-12 | 中国平安人寿保险股份有限公司 | The method and terminal device of automatic test profitable result |
| CN110753939A (en) * | 2017-06-07 | 2020-02-04 | 三菱电机大楼技术服务株式会社 | Data name classification support device and data name classification support program |
| CN110942180A (en) * | 2019-11-12 | 2020-03-31 | 广州泽沐信息科技有限责任公司 | Industrial design matching service party prediction method based on xgboost algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106262A (en) * | 2013-01-28 | 2013-05-15 | 新浪网技术(中国)有限公司 | Method and device of file classification and generation of support vector machine model |
| CN104700190A (en) * | 2014-09-17 | 2015-06-10 | 国家电网公司 | Method and device for matching item and professionals |
| CN105160498A (en) * | 2015-10-21 | 2015-12-16 | 北京普猎创新网络科技有限公司 | Personal value calculation method based on big data |
| CN105574788A (en) * | 2016-02-05 | 2016-05-11 | 北京普猎创新网络科技有限公司 | Competency label-based training providing method |
-
2016
- 2016-05-18 CN CN201610329208.7A patent/CN106022599A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106262A (en) * | 2013-01-28 | 2013-05-15 | 新浪网技术(中国)有限公司 | Method and device of file classification and generation of support vector machine model |
| CN104700190A (en) * | 2014-09-17 | 2015-06-10 | 国家电网公司 | Method and device for matching item and professionals |
| CN105160498A (en) * | 2015-10-21 | 2015-12-16 | 北京普猎创新网络科技有限公司 | Personal value calculation method based on big data |
| CN105574788A (en) * | 2016-02-05 | 2016-05-11 | 北京普猎创新网络科技有限公司 | Competency label-based training providing method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110753939A (en) * | 2017-06-07 | 2020-02-04 | 三菱电机大楼技术服务株式会社 | Data name classification support device and data name classification support program |
| CN110753939B (en) * | 2017-06-07 | 2024-03-01 | 三菱电机楼宇解决方案株式会社 | Data name classification auxiliary device |
| CN107977794A (en) * | 2017-12-14 | 2018-05-01 | 方物语（深圳）科技文化有限公司 | Data processing method, device, computer equipment and the storage medium of industrial products |
| CN108596420A (en) * | 2018-02-02 | 2018-09-28 | 武汉文都创新教育研究院（有限合伙） | A kind of talent assessment system and method for Behavior-based control |
| CN109324963A (en) * | 2018-08-22 | 2019-02-12 | 中国平安人寿保险股份有限公司 | The method and terminal device of automatic test profitable result |
| CN109324963B (en) * | 2018-08-22 | 2024-03-15 | 中国平安人寿保险股份有限公司 | Method for automatically testing profit result and terminal equipment |
| CN110942180A (en) * | 2019-11-12 | 2020-03-31 | 广州泽沐信息科技有限责任公司 | Industrial design matching service party prediction method based on xgboost algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111178624B (en) | New product demand prediction method | |
| CN106022599A (en) | Industrial design talent level evaluation method and system | |
| Riddington et al. | Comparison of gravity model, survey and location quotient-based local area tables and multipliers | |
| Sharif | Basis for techno-economic policy analysis | |
| CN106845717A (en) | A kind of energy efficiency evaluation method based on multi-model convergence strategy | |
| CN109543359B (en) | Artificial intelligence package design method and system based on Internet of things big data | |
| CN108388955A (en) | Customer service strategies formulating method, device based on random forest and logistic regression | |
| CN108710663A (en) | A kind of data matching method and system based on ontology model | |
| CN108399287A (en) | Using the appraisal procedure of the machine tool beam design scheme of Fuzzy Level Analytic Approach | |
| CN107301604A (en) | Multi-model fusion estimation system | |
| CN109948913A (en) | A kind of multi-source feature power consumer composite portrait system based on double-deck xgboost algorithm | |
| Liu et al. | Short-term load forecasting based on elastic net improved GMDH and difference degree weighting optimization | |
| CN109377049A (en) | A kind of " industry-industry-enterprise " whole process development zone industrial selection method | |
| Cui et al. | Deep integration of health information service system and data mining analysis technology | |
| CN110298693A (en) | A kind of big data system recommended based on vegetable price prediction | |
| Sun et al. | Evaluation model and empirical research on the green innovation capability of manufacturing enterprises from the perspective of ecological niche | |
| Pang et al. | Selection of wind turbine based on fuzzy analytic network process: A case study in China | |
| CN101404073B (en) | Complex product fuzz hierarchy collocation method | |
| Gehlhar | Economic growth and trade in the Pacific rim: an analysis of trade patterns | |
| CN107194727A (en) | The user's request factor based on Kano model is characterized and extracting method | |
| Han et al. | Analysis of global competitiveness of engineering modeling and simulation technology for next-manufacturing innovation: Using quantitative analysis of patents and papers | |
| CN102855354A (en) | Industrial-enterprise-oriented online multi-dimensional energy consumption data statistics modeling method | |
| Öztürk Akbıyık et al. | Analysis of Global Research Trends on BIM Studies in the Field of Architecture | |
| Yang et al. | Modelling and forecast of future growth for Shandong’s small industrial towns: A scenario-based interactive approach | |
| Fang et al. | [Retracted] Analysis of Human Resource Allocation Scheme for Digital Media Big Data Based on Recurrent Neural Network Model |
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 |
Application publication date: 20161012 |
|
| RJ01 | Rejection of invention patent application after publication |