CN115455205A - Time sequence knowledge graph-based occupational development planning method - Google Patents
Time sequence knowledge graph-based occupational development planning method Download PDFInfo
- Publication number
- CN115455205A CN115455205A CN202211152160.9A CN202211152160A CN115455205A CN 115455205 A CN115455205 A CN 115455205A CN 202211152160 A CN202211152160 A CN 202211152160A CN 115455205 A CN115455205 A CN 115455205A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- time
- development planning
- graph
- candidate
- 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.)
- Granted
Links
- 238000011161 development Methods 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013528 artificial neural network Methods 0.000 claims abstract description 31
- 239000013598 vector Substances 0.000 claims abstract description 21
- 238000013507 mapping Methods 0.000 claims abstract description 10
- 230000001755 vocal effect Effects 0.000 claims description 23
- 230000008859 change Effects 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000008520 organization Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 6
- 230000018109 developmental process Effects 0.000 description 88
- 238000010586 diagram Methods 0.000 description 9
- 238000011156 evaluation Methods 0.000 description 5
- 230000007704 transition Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012015 optical character recognition Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000007115 recruitment Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 238000013481 data capture Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a time sequence knowledge graph-based occupational development planning method, which comprises the following steps of: s1: analyzing and extracting mass information, and constructing nodes and relations of a knowledge graph; s2: mapping each node into a high-dimensional vector according to the constructed knowledge graph, constructing a graph neural network, inputting the graph neural network into a multilayer fully-connected neural network for classification, and constructing an occupational development planning model; s3: and extracting the information of the occupational development planning candidate, mapping the information of the occupational development planning candidate to a high-dimensional vector according to a graph neural network of a knowledge graph, inputting the information into the occupational development planning model, and outputting an occupational development path and a target with a background experience similar to that of the occupational development planning candidate. According to the invention, career planning analysis based on the time sequence knowledge graph of mass data can give career development suggestions of related candidates relatively comprehensively and quickly.
Description
Technical Field
The invention relates to the field of computer software, in particular to a time sequence knowledge graph-based occupational development planning method.
Background
China has tens of millions of college graduates to the society every year, and a good career plan not only has important guiding significance for individuals, but also is beneficial to the coming of the society for making macroscopic regulation and control of talents and promoting the healthy and stable development of national economy. Due to human complexity, career planning for an individual is a very complex problem, and a candidate is easy to cause a large amount of mismatching in the market if the candidate does not have clear systematic cognition and planning for the professional development of the candidate. Data reports show that more than 2/3 of staff repent to miss the first occupation, and meanwhile, a large number of middle-high-end personnel cannot find suitable work, such as college president and masters which learn seven-year design and do not do design work at last or learn communication but do programming-related work following the hot spots at that time, and finally do not know how to promote related skills or competence for related work. Therefore, it is necessary for the on-duty staff or the students about to take the job to help them form the correct professional cognition, complete the systematic professional planning and design the scientific development path.
However, the current implementation of professional planning services is mainly divided into three categories:
the first category is based primarily on offline traditional expert knowledge-based professional development planning. The evaluation and planning are carried out in an off-line manual docking mode by experts, and the defects of high price, low efficiency and the like exist, so that the evaluation and planning method has the effect in a small range and is difficult to popularize quickly. Meanwhile, the professional planning service is influenced by the artificial knowledge range and the professional level of the professional planning service, even due to the influence of subjective factors, certain planning deviation is easily caused, the influence on an individual is reflected to be amplified to cause bad experience, and therefore the extreme sample is prevented from having strong practical significance by deeply exploring the characteristics of personal characters, education, skills and the like.
The second category is that the key phrases extracted through the job database on line are combined with the key phrases input by the user to recommend the job. The technology mainly considers the matching degree of a user and a certain position at the time, does not consider the career growth of the user, the character characteristics of people and the soft characteristics which are not reflected on a resume, and cannot well support the relatively long-term planning problem of career development.
The third category is professional planning recommendations with big data or deep learning. This technique is mainly limited to entering information for isolated positions only, and does not consider the associated transitions and changes between positions. Therefore, only the fitness of employment of a single node or the skill and the ability required to be supplemented can be guided in this way, but dynamic changes of professional development transition are not considered, and a comprehensive professional development plan is not formed.
The massive resume data naturally has information of diversified professional development paths, if the common characteristics and attributes of the related professional developments of different user groups can be obtained based on the professional development evolution information provided by the massive data, a targeted professional planning suggestion is extracted, and the information of the related academic histories, skills, abilities and the like which are required to be possessed is summarized, so that the method is efficient and accurate.
Thus, the prior art is deficient and needs improvement.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for career development planning based on the time sequence knowledge graph can comprehensively and quickly give career development suggestions of related candidates.
The technical scheme of the invention is as follows: a method for career development planning based on a time-series knowledge graph comprises the following steps: s1: analyzing and extracting mass information, and constructing nodes and relations of a knowledge graph; s2: mapping each node into a high-dimensional vector according to the constructed knowledge graph, constructing a graph neural network, inputting the graph neural network into a multilayer fully-connected neural network for classification, and constructing an occupational development planning model; s3: and extracting the information of the occupational development planning candidate, mapping the information of the occupational development planning candidate to a high-dimensional vector according to a graph neural network of a knowledge graph, inputting the information into the occupational development planning model, and outputting an occupational development path and a target with a background experience similar to that of the occupational development planning candidate.
In the technical solution, in the method for vocational development planning based on the time-series knowledge graph, in step S1, the construction of the nodes and the relations of the knowledge graph includes: s11: constructing a skill knowledge node and a relation; s12: constructing probability knowledge nodes and relations; s13: constructing knowledge nodes and relations of a company school; s14: and constructing the time-sequence knowledge nodes and the relationship of the talents.
The method for career development planning based on the time-series knowledge map is applied to each technical scheme, wherein the skill knowledge is the skill knowledge of all-industry and all-post, and comprises primary industry, secondary function, tertiary post, professional skill and soft skill; the probability knowledge is the probability of various characteristics; the company school knowledge comprises logos, incomes, web sites, properties, numbers of people, organization structures, departments, teams and posts of a company, and grades, starting time, official networks, courtyards and professions of schools; the talent time sequence knowledge comprises resume of the massive candidate talents in the massive information and time for the massive candidate talents to be added into the knowledge graph.
In the method for vocational development planning based on the time sequence knowledge graph, step S14 is performed to update the talent time sequence knowledge nodes and relationships in real time according to the addition of the massive candidate talent resumes or the update of the massive candidate resumes.
The method for professional development planning based on the time-series knowledge graph is applied to the technical schemes, wherein the nodes and the relations of the knowledge graph comprise the characteristics of the nodes, the similarity among the nodes and the implicit relations among the nodes.
In the method for vocational development planning based on the time-series knowledge graph, in step S2, each node is mapped into a high-dimensional vector by using an attention mechanism through a plurality of layers of GATs, and a graph neural network of the knowledge graph is constructed.
In the method for professional development planning based on the time sequence knowledge graph, the talent time sequence knowledge nodes mapped into the high-dimensional vectors are input into the multilayer fully-connected neural network for classification in step S2, and a professional development planning model is constructed.
In step S3, the professional development planning model automatically analyzes and fuses professional experiences of a plurality of massive candidate talents, and outputs professional development paths and targets with background experiences most similar to those of the candidate talents for professional development planning.
The method for vocational development planning based on the time sequence knowledge graph is applied to the technical schemes, and the output vocational development path and the target are vocational development planning reports in preset formats.
The method for vocational development planning based on the time-series knowledge graph is applied to the technical schemes, and further comprises the following steps of S4: and monitoring the change of the information of the candidate for the professional development planning in real time, and repeating the step S3.
The invention has the beneficial effects that:
the invention constructs the nodes and the relations of the knowledge graph by analyzing and extracting mass information, constructs the graph neural network and the career development planning model, and can give career development suggestions of related candidates more comprehensively and quickly based on the career planning analysis of the time sequence knowledge graph of mass data, and also can give a career staff with a similar background of the candidates for the reference of the candidates. The invention is different from the prior occupational development suggestion mode based on offline manual work, is not limited to the manual knowledge boundary, and avoids the subjective factors and the deviation of different professional levels; compared with the mode based on the existing position library or single-point matching degree on other lines, the method considers dynamic position transition, can dynamically depict the position development trend by combining a time sequence map, and can deduce the position path by integrating the position development and growth processes; moreover, due to the real-time property and the rapidity of the method, the follow-up occupation selection of the candidate can be tracked, and the related suggestions can be dynamically adjusted; more effective and professional to assist the professional development of the candidate.
Drawings
FIG. 1 is a knowledge graph update flow diagram of the present invention;
FIG. 2 is a flowchart of the vocational development planning prediction of the present invention;
FIG. 3 is a model diagram of a model of career development planning according to the present invention;
fig. 4 is a diagram of a career development planning suggestion of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and the specific embodiments.
The embodiment provides a method for vocational development planning based on a time sequence knowledge graph, which comprises the following steps: first, step S1: and analyzing and extracting mass information, and constructing nodes and relations of the knowledge graph.
Before constructing nodes and relations of a knowledge graph, massive information needs to be collected, analyzed and extracted, and the method comprises the following steps: resume information extraction, network information extraction and user information collection, in particular to techniques of resume analysis, OCR, named entity recognition and the like. The resume information can describe information of a candidate most directly, resume content is analyzed, automatically classified and added into a knowledge graph, the resume content comprises schools, companies and the like of the candidate, webpage information extraction is also a very important method for obtaining a cow, most of people who build trees for self-work can leave traces in the internet environment actively or passively, and the most direct method such as the candidate has a publicly published paper, a source code of githu, a blog of CSDN and the like can further improve the information of the candidate.
Moreover, the method can be used for collecting user data in real time, and specifically comprises the following steps: and (3) performing real-time dynamic capture on public information on a network through a data capture tool, such as company information: industry, scale, organizational structure, service scope, location, etc., institution information: school, specialty and level, job information: company, recruitment information, demand, etc. In addition to these hard information, soft indicators are also very important, so we will collect user information, i.e. user information obtained after contacting with candidates (e.g. hunter or HR) using our recruiting platform, such as communication ability, expression ability, etc.
Moreover, the nodes and relations for constructing the knowledge graph specifically include: s11: constructing skill knowledge nodes and relations; s12: constructing probability knowledge nodes and relations; s13: constructing knowledge nodes and relations of a company school; s14: and constructing a talent time sequence knowledge node and a relationship.
The skill knowledge is the skill knowledge of all-industry all-post, and comprises a first-level industry, a second-level function, a third-level post, professional skills and soft skills; the probability knowledge is the probability of various characteristics; the company school knowledge comprises logos, incomes, websites, properties, people numbers, organization structures, departments, teams and posts of a company, and grades, starting time, official networks, hospital systems and specialties of schools; the talent time sequence knowledge comprises resumes of a large number of candidate talents in the mass information and time for adding the large number of candidate talents into the knowledge graph.
Thus, the method combines a large amount of information of the candidate, and combines the information of the current market environment and the like, including education background, classmates information, colleagues information, company information, industry, project experience and some basic information, such as the city, income curve and position information. And forming a knowledge graph structure according to the mass information content.
Wherein, the first stage: a skill knowledge graph; and integrating the network information and performing manual arrangement by an expert consultant. And generating a skill knowledge map covering all posts in the whole industry. The nodes comprise primary industry, secondary function, tertiary post and skill, wherein the skill comprises professional skill and soft skill.
And a second stage: probabilistic knowledgemaps, i.e., with the knowledge map of skills, we add more features to the second stage, such as gender, age, academic calendar, work experience, etc. And meanwhile, calculating the probability of various characteristics, such as male and female proportion and the like, under corresponding posts.
And a third stage: the company school knowledge graph at this stage constitutes a static company knowledge graph. Including but not limited to companies: logo, income, address, nature, number of people, organizational structure, department, team, post, etc., school: level, time of creation, official website, department of hospital, specialty, etc.
A fourth stage: talent time-series knowledge-graphs; at this stage, various kinds of information of talents and time are added. And constructing a talent time sequence diagram. This part is the dynamic knowledge map; wherein, the knowledge-graph updating process is shown in fig. 1; in step S14, updating the talent timing sequence knowledge nodes and relationships in real time according to the newly added or updated mass candidate biographies of the mass candidate biographies; and the nodes and the relations of the knowledge graph comprise the characteristics of each node, the similarity between the nodes and the implicit relation between the nodes. Thus, each time a candidate resume is added or updated, the corresponding node and edge are added and the corresponding time node is stored. Additional information such as interview evaluations, character tests, and events may be added. Meanwhile, other nodes and edges can be supplemented, perfected and updated.
After the knowledge-graph is constructed, step S2 is performed: mapping each node into a high-dimensional vector according to the constructed knowledge graph, constructing a graph neural network, inputting the graph neural network into a multilayer fully-connected neural network for classification, and constructing a professional development planning model; moreover, each node is mapped into a high-dimensional vector by using an attention mechanism of the multi-layer GAT, and a graph neural network of the knowledge graph is constructed; and inputting the talent time sequence knowledge nodes mapped into the high-dimensional vectors into the multilayer fully-connected neural network for classification, and constructing an occupational development planning model.
When the knowledge graph is constructed and used, each node in the knowledge graph is an entity, such as a candidate, a school, a company and the like, label information, such as 211/985/Shuangyi and the like, from a website is also contained in the knowledge graph, and behavior and evaluation information of a user on a recruitment system, such as information completion of soft strength of a candidate, is also contained in the knowledge graph. Constructs include, but are not limited to, company information: industry, scale, service scope, location, etc. Candidate information: salary, school, specialty, post, etc., school: location, specialty, 211/985/biclass, etc. tab, website home page: the gimhiub, CSDN, hopkins, patent nets, and the like disclose talent dynamic knowledge maps of legitimate information channels. In addition, new entities or new entity attributes are discovered through network information, and the coverage rate of the knowledge graph is continuously improved. Then, the data in the knowledge map is calculated in a timed, full and incremental mode, including the extraction of the characteristics of all nodes, the similarity between the nodes, the implicit relation between the nodes and the like, and finally, each candidate node is mapped into a high-dimensional vector to be used as downstream model prediction. Based on the method, the talent knowledge map is dynamically updated, and instantaneity and accuracy are guaranteed.
When the occupational development planning model is constructed, the occupational development planning model is as shown in fig. 3, nodes in a Graph are firstly subjected to multilayer GAT (Graph Attention Network), each node is mapped into a high-dimensional vector by using an Attention mechanism of the GAT, rich relevant information is aggregated and abstracted by redirection, then all nodes of 'people' are selected, namely talent time sequence knowledge nodes are input into a lower-layer multilayer fully-connected neural Network for classification, the classification is mainly aimed at finding one or more target characters, and multidimensional information such as relevant experience, skills and the like of the target characters is used as a target of a candidate for the predicted occupational development planning. In the schematic diagram of fig. 3, dots with different shades and grays may be used to respectively represent "company", "person", and "school" (deep to shallow), which are only schematic for convenience of understanding, and the actual diagram includes many nodes, so that there are not examples.
And finally, step S3: and extracting the information of the occupational development planning candidate, mapping the information of the occupational development planning candidate to a high-dimensional vector according to a graph neural network of a knowledge graph, inputting the information into the occupational development planning model, and outputting an occupational development path and a target with a background experience similar to that of the occupational development planning candidate. The professional development planning model automatically analyzes and fuses professional experiences of a plurality of massive candidate talents, and outputs professional development paths and targets with background experiences most similar to those of the professional development planning candidates. And outputting the professional development path and a professional development planning report with a preset format.
When the occupational development planning prediction is carried out by applying the neural network of the diagram and the occupational development planning model, the flow steps are as follows, wherein the occupational development planning prediction flow diagram is shown in fig. 2.
Firstly, inputting all information of candidates for planning professional development, including past histories and expected position information, and comparing the information formats obtained in the step with various formats, wherein the technologies of resume analysis, OCR (optical character recognition), named entity recognition and the like applied in construction of a knowledge graph are used.
And further, mapping the information of the candidate for professional development planning into a high-dimensional vector by combining a diagram neural network embedding scheme of a latest knowledge graph, inputting the information into a professional development planning model of a multi-layer neural network of fine-tune to predict the latest professional planning of the candidate, and automatically analyzing and fusing a plurality of candidate professional experience output background experiences to output professional development paths which are most similar to the candidate for the candidate to refer to.
After the professional development planning model outputs a plurality of target tasks, a more complex fusion model integrating expert knowledge and natural language processing is needed, and information of a plurality of target characters is aggregated and used as a final professional planning. And outputting the result as a professional development planning report, and summarizing the information of the related academic calendar, skills, abilities, professions and the like, wherein the information comprises the required time, the learned skills and the direction to which the next work is to be searched for more appropriately. More particularly, a career development path is given in the database that most closely resembles the candidate's context, wherein the career development path may be real or abstract for reference by the candidate. Moreover, the dynamics of the candidate can be tracked in real time, and the suggested professional development plan can be correspondingly adjusted along with the professional development change of the candidate. The final suggestion part in the career planning outputs as shown in fig. 4, fig. 4 is a radar map example of a certain career planning in a certain industry, which contains a plurality of indexes, a polygon 1 in the radar map is a result to be achieved, and a middle deep line 2 is the current state of a certain career development planning candidate, so that the difference between the career development planning candidate and a target result can be well shown through the radar map, a comprehensive analysis result can be provided for the career development planning candidate, the candidate can be helped to pertinently fill up corresponding short boards, and the target state can be quickly reached.
In conclusion, the talent dynamic knowledge map constructed by the mass data can provide career planning data which are most timely and effective for the candidates for career development planning, the industry development dynamics can be rapidly mastered, and the candidates can clearly know the direction and the mode of the labor force of the candidates through the career benchmarks. It may be specific to which skills (including professional skills and soft skills) need to be mastered at what point in time, which type of company is more suitable for itself, what level the company can reach according to the existing planning, and so on.
Therefore, after a time sequence knowledge graph supported by massive resumes is provided, all information of a candidate for career development planning is mapped into a high-dimensional vector by combining a time sequence neural network embedding scheme of a latest knowledge graph, the vector comprises all information (companies, schools and other candidates) in the graph, the information is input into a trained neural network to predict the latest career planning of the candidate, and information of related academic histories, skills, capabilities, industries and the like which are required to be provided is summarized, wherein the information comprises the required time, the learned skills and the direction in which the next work is more suitable to find, and more particularly, a career development path (which may be real or abstracted) of a career staff in a database and the most similar to the background of the candidate is provided for the candidate to refer. Moreover, the dynamics of the candidate can be tracked in real time, and the suggested professional development plan can be correspondingly adjusted along with the professional development change of the candidate; namely, the information change of the candidate for vocational development planning can be monitored in real time, and the step S3 is repeated.
In short, this scheme can be summarized in two stages:
1. a knowledge graph construction stage: massive information (resumes, public information and user information) is collected, analyzed and extracted, nodes and relations are enriched, a knowledge graph is constructed, and updating is carried out at irregular time;
2. and (3) vocational planning stage: giving candidate information, entering nodes where the candidates are located into high-dimensional vectors through a neural network of the graph, inputting the high-dimensional vectors into a trained neural network model to predict other multiple candidates which are possibly close to the candidates in the future, automatically analyzing and fusing information such as current professional experiences, skills and the like of the multiple candidates, and taking the information as professional development paths and targets of the given candidates.
Therefore, the invention of the embodiment aims to excavate a corresponding career development path for each candidate in a time-series knowledge graph with massive information. The career planning problem is very complex, the current work has a lot of crossovers, and the transition between careers needs more data and technical support, so that a complete time sequence dynamic knowledge graph needs to be constructed by combining various factors and the historical information of the current candidate; the knowledge graph may contain information about each candidate, such as educational background, choice of work experience, city change, interview experience and evaluation, etc., as well as company information, including the size, nature, industry, etc., of the company, information about the position of the company, information about the institution, etc. As the knowledge graph can lead entities and the relationship among the entities to be changed continuously along with the time lapse and the occupational transition of the candidates or the development change of the era, in order to acquire knowledge comprehensively, a dynamic knowledge graph is built, time dimension is added into knowledge graph data, and the variation and trend of the career of the candidates with different backgrounds along with the time are analyzed by utilizing a time sequence analysis technology and a graph neural network technology, so that the suggestion of key career planning is summarized for the candidates to refer.
Compared with the traditional professional development planning scheme based on expert knowledge, the method considers more information dimensions, more comprehensive information and more objective market analysis results, and is faster and more personalized. A single expert experience-based scheme is often not objective enough, and different experts have different hunting directions, so that a very comprehensive scheme cannot be given.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalents and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for planning occupational development based on a time-series knowledge graph is characterized by comprising the following steps:
s1: analyzing and extracting mass information, and constructing nodes and relations of a knowledge graph;
s2: mapping each node into a high-dimensional vector according to the constructed knowledge graph, constructing a graph neural network, inputting the graph neural network into a multilayer fully-connected neural network for classification, and constructing a professional development planning model;
s3: and extracting the information of the occupational development planning candidate, mapping the information of the occupational development planning candidate to a high-dimensional vector according to a graph neural network of a knowledge graph, inputting the information into the occupational development planning model, and outputting an occupational development path and a target with a background experience similar to that of the occupational development planning candidate.
2. The method for time-series knowledgeable graph-based vocational development planning of claim 1, wherein: in step S1, the nodes and relationships for constructing the knowledge graph include:
s11: constructing skill knowledge nodes and relations;
s12: constructing probability knowledge nodes and relations;
s13: constructing knowledge nodes and relations of a company school;
s14: and constructing the time-sequence knowledge nodes and the relationship of the talents.
3. The method for vocational development planning based on a time series knowledge graph of claim 2, wherein the skill knowledge is industry wide job skill knowledge comprising primary industry, secondary job, tertiary job, professional skill and soft skill; the probability knowledge is the probability of various characteristics; the company school knowledge comprises logos, incomes, websites, properties, people numbers, organization structures, departments, teams and posts of a company, and grades, starting time, official networks, hospital systems and specialties of schools; the talent time sequence knowledge comprises resumes of a large number of candidate talents in the mass information and time for adding the large number of candidate talents into the knowledge graph.
4. The method for vocational development planning based on the time-series knowledge-graph of claim 3, wherein in step S14, the talent time-series knowledge nodes and relationships are updated in real time according to the addition of the large number of candidate talent resumes or the update of the large number of candidate talent resumes.
5. The method for career development planning based on time-series knowledge graph of claim 4, wherein the nodes and relations of the knowledge graph comprise the characteristics of each node, the similarity between the nodes and the implicit relation between the nodes.
6. The method for career development planning based on time-series knowledge graph of claim 3, wherein in step S2, a graph neural network of the knowledge graph is constructed by mapping each node into a high-dimensional vector through a plurality of layers of GATs by using an attention mechanism thereof.
7. The method for vocational development planning based on the time-series knowledge graph of claim 6, wherein in step S2, the talent time-series knowledge nodes mapped into the high-dimensional vectors are input into the multi-layer fully-connected neural network for classification, and a vocational development planning model is constructed.
8. The method for vocational development planning based on time-series knowledge graph of claim 7, wherein in step S3, the vocational development planning model automatically analyzes and fuses the vocational experiences of a plurality of mass candidate talents and outputs the vocational development path and goal with the background experience most similar to the vocational development planning candidates.
9. The method for vocational development planning based on time-series knowledgebase map of claim 8, wherein the outputted vocational development path and goal are professional development planning reports in a preset format.
10. The method for time-series knowledgebase planning for professional development according to claim 1, further comprising step S4: and monitoring the change of the information of the candidate for the professional development planning in real time, and repeating the step S3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211152160.9A CN115455205B (en) | 2022-09-21 | 2022-09-21 | Occupational development planning method based on time sequence knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211152160.9A CN115455205B (en) | 2022-09-21 | 2022-09-21 | Occupational development planning method based on time sequence knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115455205A true CN115455205A (en) | 2022-12-09 |
| CN115455205B CN115455205B (en) | 2023-06-30 |
Family
ID=84305048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211152160.9A Active CN115455205B (en) | 2022-09-21 | 2022-09-21 | Occupational development planning method based on time sequence knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115455205B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117371625A (en) * | 2023-12-07 | 2024-01-09 | 中科软股教育科技(北京)股份有限公司 | Occupational development prediction system and method based on big data analysis |
| CN117828194A (en) * | 2024-03-04 | 2024-04-05 | 武汉华林梦想科技有限公司 | Occupational recommendation method based on knowledge graph |
| CN118071032A (en) * | 2024-04-18 | 2024-05-24 | 贵州优特云科技有限公司 | Personalized occupation planning method and device based on artificial intelligence and storage medium |
| CN118132616A (en) * | 2024-05-08 | 2024-06-04 | 江西工业贸易职业技术学院(江西省粮食干部学校、江西省粮食职工中等专业学校) | Auxiliary analysis method and system for professional planning |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082356A1 (en) * | 2008-09-30 | 2010-04-01 | Yahoo! Inc. | System and method for recommending personalized career paths |
| CN109460479A (en) * | 2018-11-19 | 2019-03-12 | 广州合摩计算机科技有限公司 | A kind of prediction technique based on reason map, device and system |
| US20190228288A1 (en) * | 2018-01-22 | 2019-07-25 | Kliq.Ca Inc. | Systems and methods for decision modelling of a temporal path |
| US20190303836A1 (en) * | 2018-03-30 | 2019-10-03 | Accenture Global Solutions Limited | Determining optimal workforce types to fulfill occupational roles in an organization based on occupational attributes |
| US20200034479A1 (en) * | 2018-07-26 | 2020-01-30 | JANZZ Ltd | Classifier system and method |
| CN112417165A (en) * | 2020-11-18 | 2021-02-26 | 杭州电子科技大学 | Method and system for constructing and inquiring lifetime planning knowledge graph |
| CN113159330A (en) * | 2021-04-30 | 2021-07-23 | 嘉应学院 | Professional learning path system and method based on hierarchical task network planning model learning |
| CN113240400A (en) * | 2021-06-02 | 2021-08-10 | 北京金山数字娱乐科技有限公司 | Candidate determination method and device based on knowledge graph |
| CN114331380A (en) * | 2021-12-31 | 2022-04-12 | 北京百度网讯科技有限公司 | Method, system, equipment and storage medium for predicting occupational flow relationship |
| CN114691892A (en) * | 2022-04-21 | 2022-07-01 | 君之福(北京)科技有限公司 | Retired soldier occupation planning method and retired soldier occupation planning system based on big data |
-
2022
- 2022-09-21 CN CN202211152160.9A patent/CN115455205B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082356A1 (en) * | 2008-09-30 | 2010-04-01 | Yahoo! Inc. | System and method for recommending personalized career paths |
| US20190228288A1 (en) * | 2018-01-22 | 2019-07-25 | Kliq.Ca Inc. | Systems and methods for decision modelling of a temporal path |
| US20190303836A1 (en) * | 2018-03-30 | 2019-10-03 | Accenture Global Solutions Limited | Determining optimal workforce types to fulfill occupational roles in an organization based on occupational attributes |
| US20200034479A1 (en) * | 2018-07-26 | 2020-01-30 | JANZZ Ltd | Classifier system and method |
| CN109460479A (en) * | 2018-11-19 | 2019-03-12 | 广州合摩计算机科技有限公司 | A kind of prediction technique based on reason map, device and system |
| CN112417165A (en) * | 2020-11-18 | 2021-02-26 | 杭州电子科技大学 | Method and system for constructing and inquiring lifetime planning knowledge graph |
| CN113159330A (en) * | 2021-04-30 | 2021-07-23 | 嘉应学院 | Professional learning path system and method based on hierarchical task network planning model learning |
| CN113240400A (en) * | 2021-06-02 | 2021-08-10 | 北京金山数字娱乐科技有限公司 | Candidate determination method and device based on knowledge graph |
| CN114331380A (en) * | 2021-12-31 | 2022-04-12 | 北京百度网讯科技有限公司 | Method, system, equipment and storage medium for predicting occupational flow relationship |
| CN114691892A (en) * | 2022-04-21 | 2022-07-01 | 君之福(北京)科技有限公司 | Retired soldier occupation planning method and retired soldier occupation planning system based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 苏叶健: "基于知识图谱的职业规划推荐系统构建研究", 《企业科技与发展》, pages 69 - 72 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117371625A (en) * | 2023-12-07 | 2024-01-09 | 中科软股教育科技(北京)股份有限公司 | Occupational development prediction system and method based on big data analysis |
| CN117371625B (en) * | 2023-12-07 | 2024-02-06 | 中科软股教育科技(北京)股份有限公司 | Occupational development prediction system and method based on big data analysis |
| CN117828194A (en) * | 2024-03-04 | 2024-04-05 | 武汉华林梦想科技有限公司 | Occupational recommendation method based on knowledge graph |
| CN117828194B (en) * | 2024-03-04 | 2024-06-21 | 武汉华林梦想科技有限公司 | Occupational recommendation method based on knowledge graph |
| CN118071032A (en) * | 2024-04-18 | 2024-05-24 | 贵州优特云科技有限公司 | Personalized occupation planning method and device based on artificial intelligence and storage medium |
| CN118071032B (en) * | 2024-04-18 | 2024-06-28 | 贵州优特云科技有限公司 | Personalized occupation planning method and device based on artificial intelligence and storage medium |
| CN118132616A (en) * | 2024-05-08 | 2024-06-04 | 江西工业贸易职业技术学院(江西省粮食干部学校、江西省粮食职工中等专业学校) | Auxiliary analysis method and system for professional planning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115455205B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230174B (en) | Online interactive learning system and method based on network | |
| US10331764B2 (en) | Methods and system for automatically obtaining information from a resume to update an online profile | |
| Karthikeyan et al. | Towards developing hybrid educational data mining model (HEDM) for efficient and accurate student performance evaluation | |
| CN115455205B (en) | Occupational development planning method based on time sequence knowledge graph | |
| Kermanshachi et al. | Development of the project complexity assessment and management framework for heavy industrial projects | |
| Xia | Interaction recognition and intervention based on context feature fusion of learning behaviors in interactive learning environments | |
| US20150317602A1 (en) | Scoring model methods and apparatus | |
| Almgerbi et al. | A systematic review of data analytics job requirements and online-courses | |
| US20150317605A1 (en) | Scoring model methods and apparatus | |
| Närman et al. | Using enterprise architecture to analyse how organisational structure impact motivation and learning | |
| Kour et al. | Analysis of student performance using Machine learning Algorithms | |
| Darko et al. | Service quality in football tourism: an evaluation model based on online reviews and data envelopment analysis with linguistic distribution assessments | |
| Yahya et al. | A data‐mining‐based approach to informed decision‐making in engineering education | |
| José-García et al. | C3-IoC: A career guidance system for assessing student skills using machine learning and network visualisation | |
| Muhamedyev et al. | The use of machine learning “black boxes” explanation systems to improve the quality of school education | |
| Rahayu et al. | Defining e-portfolio factor for competency certification using fuzzy delphi method | |
| Bian et al. | Research on an Artificial Intelligence‐Based Professional Ability Evaluation System from the Perspective of Industry‐Education Integration | |
| Ruppert | Visual analytics to support evidence-based decision making | |
| KR20210058376A (en) | Learning system using video offering answer based on artificial intelligence | |
| Khan et al. | Measuring economic resilience of manufacturing organization leveraging integrated data envelopment analysis (DEA)-machine learning approach | |
| Hogo | The design of academic programs using rough set association rule mining | |
| Ayyoub et al. | Learning Style Identification Using Semi-Supervised Self-Taught Labeling | |
| KR20210001703A (en) | Method for operating learner guiding service, system and computer-readable medium recording the method | |
| Rosário | Generative ai and generative pre-trained transformer applications: Challenges and opportunities | |
| Ajgaonkar et al. | EduKrishnaa: A Career Guidance Web Application Based on Multi-intelligence Using Multiclass Classification Algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |