CN117828194A - Occupational recommendation method based on knowledge graph - Google Patents
Occupational recommendation method based on knowledge graph Download PDFInfo
- Publication number
- CN117828194A CN117828194A CN202410240353.2A CN202410240353A CN117828194A CN 117828194 A CN117828194 A CN 117828194A CN 202410240353 A CN202410240353 A CN 202410240353A CN 117828194 A CN117828194 A CN 117828194A
- Authority
- CN
- China
- Prior art keywords
- node
- professional
- target person
- path
- learning ability
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000000875 corresponding effect Effects 0.000 claims description 54
- 230000006870 function Effects 0.000 claims description 53
- 230000001186 cumulative effect Effects 0.000 claims description 28
- 238000011144 upstream manufacturing Methods 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000001419 dependent effect Effects 0.000 claims description 9
- 230000002596 correlated effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 2
- 241001672694 Citrus reticulata Species 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000012797 qualification Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing. In particular to a professional recommendation method based on a knowledge graph. Comprising the following steps: constructing a knowledge graph according to professional information of the employment; generating professional paths based on the knowledge graph and combining the professions possibly engaged in the future of the target personnel and corresponding post requirements, and acquiring nodes passed by each professional path; calculating the time spent from each node to the target personnel of the adjacent downstream nodes in the professional path; calculating the passing rate between adjacent nodes of the occupational path; and calculating the cost required by each occupational path according to the passing rate, the time required to be spent and the number of nodes in the occupational path. The professional recommendation method based on the knowledge graph can greatly improve the efficiency and rationality of professional planning.
Description
Technical Field
The present invention relates generally to the field of data processing technology. More particularly, the invention relates to a professional recommendation method based on a knowledge graph.
Background
Professional planning refers to the process of individuals in the development of the business by making goals, taking actions and making decisions to achieve career success and satisfaction. With the continuous development of socioeconomic and the continuous change of professional architecture, professional planning is increasingly important for the professional success of individuals. However, many people face confusion and challenges in job planning due to the vast complexities of professional information, different personal capabilities and interests, difficulty in future predictions, and the like. The existing occupation recommendation method generally comprises the steps that an individual searches massive post information through the Internet according to the current academic and profession of the individual, and plans the future occupation of the individual through manually analyzing the corresponding requirements of each post.
A Knowledge Graph (knowledgegraph) is a structured tool for representing and displaying Knowledge, and is composed of nodes (entities) and edges (relationships), wherein the nodes represent the entities, the edges represent the relationships between the entities, and the Knowledge is presented in a graphical manner by using a visualization technology so as to help a user to better understand and utilize the Knowledge. The knowledge graph aims at describing concepts, entities, events and relations among the concepts, entities and events in the objective world, is an advanced technical means, can be used for storing and managing various kinds of knowledge and intuitively displaying the various kinds of knowledge, and therefore, in order to improve the efficiency of professional planning, the knowledge graph can be adopted for automatically carrying out the professional planning. When the professional planning is carried out on the individual, the knowledge graph can be adopted to integrate, correlate and analyze massive professional information, and more accurate and comprehensive professional planning suggestions are provided for the individual.
Disclosure of Invention
To solve one or more of the above technical problems, the present invention provides aspects as follows.
In a first aspect, the present invention provides a professional recommendation method based on a knowledge graph, including: constructing a knowledge graph according to career information of careers, wherein the career information comprises age, profession, academic, career type and corresponding post requirements; generating professional paths based on the knowledge graph and combining the professions possibly engaged in the future of the target personnel and corresponding post requirements, and acquiring nodes passed by each professional path; each node in the job path respectively represents various post requirements which are needed to be reached by the target personnel to engage in the corresponding occupation; acquiring the time spent from each node to a downstream node employment person in each occupation path, and calculating the time spent from each node to a target person of an adjacent downstream node in the occupation path by combining the learning ability value of the target person; for each occupational path, calculating the passing rate between adjacent nodes of the occupational path according to the adjacent upstream nodes and the adjacent downstream nodes of each node in the knowledge graph; and calculating the cost required by each professional route according to the passing rate, the time spent and the node number in the professional route, and recommending the professional route with the minimum cost to a target person.
In one embodiment, the time spent in calculating the professional path from each node to its neighboring downstream node target person includes: establishing a cumulative probability density function according to the time spent by each node in the professional path to the career of the downstream node; the independent variable of the cumulative probability density function is the time spent by a certain node to a career of a downstream node, and the dependent variable is the cumulative probability corresponding to the time; for each occupational path, calculating the time spent from each node to the target person of the adjacent downstream node in the occupational path according to the inverse function of the cumulative probability density function and the learning ability value of the target person.
In one embodiment, the time it takes for a target person to travel from one node in the professional path to its neighboring downstream node is positively correlated with the inverse function value of the cumulative probability density function and negatively correlated with the learning ability value of the target person.
In one embodiment, for adjacent nodes k and k+1 in the professional path, the time the target person spends from node k to node k+1 is calculated as:
wherein,representing the time from the kth node to the kth+1th node of the target person, +.>An inverse function value representing the probability cumulative density function obtained by the career from the kth node to the kth+1th node,/o>A learning ability value representing the target person.
In one embodiment, the occupation acquisition method that the target person may engage in the future includes:
acquiring professional information of a target person, and calculating the similarity between the professional data of the target person and the professional data of each person in all careers, wherein the similarity calculation formula is as follows:
in the method, in the process of the invention,representing the similarity of the professional data of the target person and the i-th career professional data, +.>Attribute value of jth Attribute representing professional data of target person, < >>An attribute value representing a j-th attribute of i-th employment data,representing the attribute number of the professional data of the target person; />() Representing a value function, wherein two values in brackets are equal, 1 is taken, and 0 is taken if the two values are not equal;
and determining the similarity with the maximum value from all the calculated similarities, and taking the occupation which is occupied by the employment corresponding to the similarity with the maximum value as the occupation which is likely to be occupied by the target personnel in the future.
In one embodiment, the method for calculating the learning ability value of the target person includes:
establishing a learning ability Gaussian function Gus according to data related to learning ability in professional data of an employment, wherein independent variables of the learning ability Gaussian function are learning ability values, and the dependent variables are probabilities of occurrence of the learning ability values;
calculating the learning ability value of the target person according to the learning time of the target person and the learning ability Gaussian function, wherein the calculation expression is as follows:
wherein,learning ability value representing the target person, +.>Indicating the learning time of the target person, +.>Representing a learning ability gaussian function; the learning ability value of the target person can reflect the learning ability of the target person, the learning ability value ranges from 0 to 1, and the closer to 1, the stronger the learning ability is.
In one embodiment, if the neighboring node downstream of the node a in the professional path is the node B, the calculating the passing rate between neighboring nodes of the professional path includes:
acquiring all adjacent upstream nodes of a D node and all adjacent downstream nodes of the D node in a knowledge graph, then respectively calculating the probability that professional data corresponding to each adjacent upstream node contains the D node and recording the probability as upstream node containing probability, and the probability that professional data corresponding to the D node contains each adjacent downstream node and recording the probability as downstream node containing probability;
and multiplying the sum of the probabilities contained in each upstream node by the sum of the probabilities contained in each downstream node to obtain the passing rate from the node A to the node D.
In one embodiment, for a professional path, calculating the cost required for it includes: respectively calculating the costs corresponding to each pair of adjacent nodes, and then adding the costs corresponding to each pair of adjacent nodes to obtain the cost required by the occupational path; for a neighboring node, its corresponding cost is inversely proportional to the pass rate corresponding to the pair of neighboring nodes and proportional to the time spent corresponding to the pair of neighboring nodes.
In one embodiment, the computational expression of the cost required for the job path is:
wherein,cost for representing professional path, +.>Representing the time from the kth node to the (k+1) th node in the professional path that the target person is likely to engage in,/for>Indicating the pass rate of the (k+1) th node in the professional path that the target person is likely to engage in,/-)>Representing the adjustment factors, the cost of the professional planning reflects the cost required for the target person from now on to his professional goal.
The invention has the technical effects that: when the professional recommendation method based on the knowledge graph is used for performing professional planning on target personnel, firstly, the professional knowledge graph is built through collecting professional information of the people, and a possible professional path of the target personnel is obtained according to the similarity of professional information of the target personnel and the careers, the learning ability value of the target personnel is calculated according to information related to learning in the professional information of the target personnel and the careers, then the time and passing rate of each node on the path of the target personnel are calculated, finally, the cost of each possible professional path is calculated, and the professional planning is performed on the target personnel according to the cost, and because the learning ability of the target personnel and the professional similarity of the target personnel and the careers are considered and the professional planning is automatically performed during the professional planning, the professional recommendation method based on the knowledge graph is used for greatly improving the efficiency and rationality of the professional planning and automatically matching and recommending the optimal professional planning result for the target personnel.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the invention are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is a flow chart schematically illustrating a knowledge-based professional recommendation method of an embodiment of the present invention;
FIG. 2 is a partial schematic diagram schematically illustrating a constructed knowledge-graph of an embodiment of the invention;
FIG. 3 is a node relationship diagram schematically illustrating an embodiment of the present invention;
FIG. 4 is a flow chart of a time calculation method that schematically illustrates the costs of a targeted persona in accordance with an embodiment of the invention;
FIG. 5 is a schematic diagram of a generated professional path that schematically illustrates an embodiment of the present invention;
FIG. 6 is a flow chart schematically illustrating a job acquisition method that a target person of an embodiment of the present invention may engage in the future;
fig. 7 is a flowchart schematically showing a calculation method of the learning ability value of the target person according to the embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Professional recommendation method embodiment based on knowledge graph:
as shown in fig. 1, the professional recommendation method based on the knowledge graph of the present invention includes:
s101, constructing a knowledge graph according to career information of careers, wherein the career information comprises ages, professions, academia, career types and corresponding post requirements.
The employment includes a plurality of staff members at various posts in various industries, and preferably, the plurality of professions includes professions in which various subjects are at the present society, such as teachers, doctors, construction workers, cost engineers, electrical engineers, programmers, public officers, business staff, sales consultants, lawyers, accountants, and the like. The academy may include a high school, a college, a university, a family, and a study. The occupation type includes the occupation of each industry, and may include, for example, education industry, medical industry, financial industry, legal industry, construction industry, and the like. The occupation types may include teachers, doctors, accountants, engineering costs, lawyers, human resources, and the like. Post requirements include requirements for age ranges, academia, professions, and various skill certificates. Skill certificates such as accounting certificates, lawyer qualifications, and the like.
As shown in FIG. 2, the knowledge graph is constructed by taking careers as entities, taking ages, professions, academia, occupation types and corresponding post requirements as attributes, and taking specific information corresponding to each attribute as attribute values. For example: for the attribute "age", if the information corresponding thereto is 23 years old, the attribute value corresponding thereto is "23 years old", and for the attribute "academy", if the information corresponding thereto is "high-middle", the attribute value corresponding thereto is "high-middle".
In this embodiment, the professional information includes age, specialty, academic, professional type, and corresponding post requirements, and in other embodiments, the nodes in the knowledge graph may further include salary, company type, and the like.
S102, generating professional paths, and acquiring nodes passed by each professional path, wherein the nodes specifically comprise: generating professional paths based on the knowledge graph and combining the professions possibly engaged in the future of the target personnel and corresponding post requirements, and acquiring nodes passed by each professional path; each node in the job path represents various post requirements that the target person needs to reach to engage in the corresponding job.
For example: if the occupation possibly engaged in by the target personnel in the future is a teacher, the post requirement of the teacher is a graduate graduation and a teacher qualification is held, and the current school of the target personnel is a senior high school, the generated occupation path is a senior high school to a graduate graduation, and the graduate graduation is a teacher qualification.
S103, calculating time spent from each node to target personnel of adjacent downstream nodes in the professional path, wherein the time is specifically as follows: acquiring the time spent from each node to a downstream node employment person in each occupation path, and calculating the time spent from each node to a target person of an adjacent downstream node in the occupation path by combining the learning ability value of the target person;
in general, the more learning a target person has, the faster it can reach a post requirement, and therefore the shorter it takes from one node to the next adjacent node; if the employment spends a long time from that node to the next adjacent node, the targeted personnel will also spend a long time. Thus, the time spent by the target person from each node to its neighboring downstream nodes in the professional path can be calculated based on the time spent by the employment and the learning ability value of the target person.
S104, for each occupational path, calculating the passing rate between adjacent nodes of the occupational path according to the adjacent upstream nodes and the adjacent downstream nodes of each node in the knowledge graph.
For a certain pair of adjacent nodes A and D, assuming that the D node is a downstream node of the A node in the occupation path, all adjacent upstream nodes of the D node and all adjacent downstream nodes of the D node in the knowledge graph can be firstly obtained, then the probability that occupation data corresponding to each adjacent upstream node contains the D node is calculated and recorded as the upstream node containing probability, and the probability that occupation data corresponding to the D node contains each adjacent downstream node is recorded as the downstream node containing probability, and the passing rate from the A node to the D node is calculated according to all the upstream node containing probabilities and the downstream node containing probabilities. And multiplying the sum of the probabilities contained in each upstream node by the sum of the probabilities contained in each downstream node to obtain the passing rate from the node A to the node D.
For example: as shown in fig. 3, assuming that the adjacent upstream node of the D node is an a node and a B node, and the adjacent downstream node of the D node is a C node, the calculation expression of the passing rate from the a node to the D node is:
（1）
in the method, in the process of the invention,represents the pass rate through the D node, +.>Representing the number of professional data containing both nodes A and D, < >>Representing the number of professional data comprising node a, +.>Representing the number of professional data including both node B and node D +.>Indicating the number of professional data comprising node B, < +.>Representing the number of professional data including both D and C nodes. The pass rate of a node is determined by the likelihood of success of the other nodes to the node and the likelihood of success of the node to the other nodes.
And S105, calculating the cost required by each professional route according to the passing rate, the time spent and the node number in the professional route, and recommending the professional route with the minimum cost to a target person.
For a professional path, the costs corresponding to each pair of adjacent nodes can be calculated respectively, and then the costs corresponding to each pair of adjacent nodes are added to obtain the cost required by the professional path. For a pair of neighboring nodes, its corresponding cost is inversely proportional to the pass rate for the pair of neighboring nodes and directly proportional to the time spent for the pair of neighboring nodes.
Assuming that there are k nodes to pass through in the professional path, the calculation expression of the cost required by the professional path is:
（2）
wherein,cost for representing professional path, +.>Representing the time from the kth node to the (k+1) th node in the professional path that the target person is likely to engage in,/for>Indicating the pass rate of the (k+1) th node in the professional path that the target person is likely to engage in,/-)>Representing adjustment factorsTypically set to 0.5; the cost of professional planning reflects the cost required for the target person from now on to his professional goals.
When the professional recommendation method based on the knowledge graph is used for performing professional planning on target personnel, firstly, the professional knowledge graph is built through collecting professional information of the people, and a possible professional path of the target personnel is obtained according to the similarity of professional information of the target personnel and the careers, the learning ability value of the target personnel is calculated according to information related to learning in the professional information of the target personnel and the careers, then the time and passing rate of each node on the path of the target personnel are calculated, finally, the cost of each possible professional path is calculated, and the professional planning is performed on the target personnel according to the cost, and because the learning ability of the target personnel and the professional similarity of the target personnel and the careers are considered and the professional planning is automatically performed during the professional planning, the professional recommendation method based on the knowledge graph is used for greatly improving the efficiency and rationality of the professional planning and automatically matching and recommending the optimal professional planning result for the target personnel.
As can be seen from the above embodiments, the calculation of the time spent from each node to its adjacent downstream node target person in the professional path is based on the time spent from each node to its downstream node careers in the professional path and the learning ability value of the target person, and as shown in fig. 4, in one embodiment, the calculation of the time spent from each node to its adjacent downstream node target person in the professional path includes:
s401, establishing a cumulative probability density function, specifically: acquiring the time spent from each node in each occupation path to a downstream node employment person according to the knowledge graph, and further establishing a cumulative probability density function according to the time spent by the employment person; the independent variable of the cumulative probability density function is the time spent by a node to a downstream node employment, and the dependent variable is the cumulative probability corresponding to the time.
As shown in fig. 5, for example, the target person is currently a senior high school student who may be engaged in profession in the future, the post requirement of the accounting is that the school is a family graduation, the primary accounting certificate, the middle accounting certificate and the mandarin certificate need to be held, and the generated professional path may be from senior high university to holding the primary accounting certificate, from holding the primary accounting certificate to holding the middle accounting certificate, and from holding the middle accounting certificate to holding the mandarin certificate. In other embodiments, the holding mandarin chinese license may also be arranged before or between the holding of the primary accounting license and the holding of the secondary accounting license in the professional path.
The total of four cumulative probability density functions are established, wherein the independent variable of the first cumulative probability density function is the time spent by the university to the university, the independent variable of the second cumulative probability density function is the time spent by the university to the primary accounting evidence, the independent variable of the third cumulative probability density function is the time spent by the primary accounting evidence to the middle accounting evidence, and the independent variable of the fourth cumulative probability density function is the time spent by the middle accounting evidence to the mandarin evidence.
If five of the history people engaged in the accounting profession spent 3 years from the middle school to the university and two spent 4 years from the middle school to the university, the corresponding factor value when the argument takes 3 is 3/5 and the corresponding factor value when the argument takes 4 is 2/5 for the first cumulative probability density function. Assuming that two of the five persons take 5 years from entrance to holding the primary accounting for university and three persons take 6 years from entrance to holding the primary accounting for university, the corresponding value of the factor at the time of taking 5 from the variable is 2/5 and the corresponding value of the factor at the time of taking 6 from the variable is 3/5 for the second cumulative probability density function. Assuming that one of the five persons takes 2 years from holding the primary accounting certificate to holding the intermediate accounting certificate and four persons take 3 years from holding the primary accounting certificate to holding the intermediate accounting certificate, for the third cumulative probability density function, the corresponding dependent variable takes 1/5 when the self-variable takes 2, and takes 4/5 when the self-variable takes 3. Assuming that four of the five persons take 1 year from holding the middle-level accounting document to holding the mandarin chinese document and one person takes 2 years from holding the middle-level accounting document to holding the mandarin chinese document, for the fourth cumulative probability density function, the corresponding dependent variable value when the independent variable value takes 1 is 4/5, and when the independent variable value takes 2, the corresponding dependent variable value takes 1/5.
S402, calculating time spent by each node in the professional path to target personnel of adjacent downstream nodes, wherein the time is specifically as follows: for each occupational path, calculating the time spent from each node to the target person of the adjacent downstream node according to the inverse function value of the cumulative probability density function and the learning ability value of the target person.
From the above embodiments, it is known that the time taken from one node in the professional path to its neighboring downstream node is positively correlated with the inverse function value of the cumulative probability density function and negatively correlated with the learning ability value of the target person. In one embodiment, for adjacent nodes k and k+1 in the professional path, the time the target person spends from node k to node k+1 is calculated as:
（3）
wherein,representing the time from the kth node to the kth+1th node of the target person, +.>An inverse function value representing the cumulative probability density function obtained by the career from the kth node to the kth+1th node,/v>A learning ability value representing the target person.
From the above embodiments, it is known that, in generating the occupation path, the occupation path is generated according to the occupation likely to be performed by the target person in the future, and in one embodiment, as shown in fig. 6, the occupation obtaining method likely to be performed by the target person in the future includes:
s601, acquiring professional information of a target person, and calculating the similarity between the professional data of the target person and the professional data of each person in all careers, wherein the similarity calculation formula is as follows:
（4）
in the method, in the process of the invention,representing the similarity of the professional data of the target person and the i-th career professional data, +.>Attribute value of jth Attribute representing professional data of target person, < >>An attribute value representing a j-th attribute of i-th employment data,attribute number representing professional data of target person +.>() The value function is represented, and if two values in brackets are equal, 1 is taken, and if the two values are not equal, 0 is taken.
For attribute values that do not contain numerical values, they may be assigned, for example: if the attribute is an academic, the attribute value is high and medium, 1 can be assigned to the academic, the attribute value is a college family, 2 can be assigned to the academic, and 3 can be assigned to the academic.
S602, determining the similarity with the maximum value from all the calculated similarities, and taking the occupation which is possibly performed by the employment person corresponding to the similarity with the maximum value as the occupation which is possibly performed by the target person in the future.
As can be seen from the above embodiments, for each professional path, it is necessary to calculate the time taken for the target person from each node to its adjacent downstream node according to the learning ability value of the target person, and in one embodiment, as shown in fig. 7, the method for calculating the learning ability value of the target person includes:
s701, establishing a learning ability Gaussian function, wherein the learning ability Gaussian function is specifically: and establishing a learning ability Gaussian function Gus according to data related to learning ability in professional data of the employment, wherein an independent variable of the learning ability Gaussian function is a learning ability value, and the dependent variable is the probability of the learning ability value.
The data related to learning ability includes an academic, a time to go to school, the number of times a prize is acquired, and the like. The learning ability gaussian function Gus may be built based on only one of the above types of data, or may be combined with a plurality of types of data. If the learning ability gaussian function Gus is constructed only according to the academy, the value of the academy is assigned according to the height of the academy, and the value corresponding to the academy is used as an independent variable, for example, the value corresponding to the academy is 1 when the academy is in the middle of the school, the value corresponding to the academy is 2 when the academy is in the university, and the value corresponding to the academy is 3 when the academy is in the research student. The probability of occurrence of the learning ability value means that the employment who reaches the learning ability value is a proportion of the total number of employment among all employment.
S702, calculating a learning ability value of the target person according to the learning time of the target person and the learning ability Gaussian function, wherein the calculation expression is as follows:
（5）
wherein,learning ability value representing the target person, +.>Indicating the learning time of the target person, +.>Representing a learning ability gaussian function. The learning ability value of the target person can reflect the learning ability of the target person, the learning ability value ranges from 0 to 1, and the closer to 1, the stronger the learning ability is.
In the description of the present specification, the meaning of "a plurality", "a number" or "a plurality" is at least two, for example, two, three or more, etc., unless explicitly defined otherwise.
While various embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Many modifications, changes, and substitutions will now occur to those skilled in the art without departing from the spirit and scope of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention.
Claims (9)
1. The professional recommendation method based on the knowledge graph is characterized by comprising the following steps of:
constructing a knowledge graph according to career information of careers, wherein the career information comprises age, profession, academic, career type and corresponding post requirements;
generating professional paths based on the knowledge graph and combining the professions possibly engaged in the future of the target personnel and corresponding post requirements, and acquiring nodes passed by each professional path; each node in the job path respectively represents various post requirements which are needed to be reached by the target personnel to engage in the corresponding occupation;
acquiring the time spent from each node to a downstream node employment person in each occupation path, and calculating the time spent from each node to a target person of an adjacent downstream node in the occupation path by combining the learning ability value of the target person;
for each occupational path, calculating the passing rate between adjacent nodes of the occupational path according to the adjacent upstream nodes and the adjacent downstream nodes of each node in the knowledge graph;
and calculating the cost required by each professional route according to the passing rate, the time spent and the node number in the professional route, and recommending the professional route with the minimum cost to a target person.
2. The knowledge-graph-based professional recommendation method according to claim 1, wherein the calculating time spent from each node to its neighboring downstream node target person in the professional path includes:
establishing a cumulative probability density function according to the time spent by each node in the professional path to the career of the downstream node; the independent variable of the cumulative probability density function is the time spent by a certain node to a career of a downstream node, and the dependent variable is the cumulative probability corresponding to the time;
for each occupational path, calculating the time spent from each node to the target person of the adjacent downstream node in the occupational path according to the inverse function of the cumulative probability density function and the learning ability value of the target person.
3. The knowledge-graph-based professional recommendation method according to claim 2, wherein the time spent from one node in the professional path to the target person of the adjacent downstream node thereof is positively correlated with the inverse function value of the cumulative probability density function and negatively correlated with the learning ability value of the target person.
4. The knowledge-graph-based professional recommendation method as claimed in claim 3, wherein for the adjacent nodes k and k+1 in the professional path, the time spent by the target person from the node k to the node k+1 is calculated as:
wherein,representing the time from the kth node to the kth+1th node of the target person, +.>An inverse function value representing a probability cumulative density function obtained by the career from the kth node to the kth+1th node,/>a learning ability value representing the target person.
5. The knowledge-based professional recommendation method as claimed in claim 1, wherein the method for acquiring professions that the target person may engage in the future comprises:
acquiring professional information of a target person, and calculating the similarity between the professional data of the target person and the professional data of each person in all careers, wherein the similarity calculation formula is as follows:
in the method, in the process of the invention,representing the similarity of the professional data of the target person and the i-th career professional data, +.>Attribute value of jth Attribute representing professional data of target person, < >>Attribute value of the j-th Attribute representing the i-th career occupational data, +.>Attribute number representing professional data of target person +.>() Representing a value function, wherein two values in brackets are equal, 1 is taken, and 0 is taken if the two values are not equal;
and determining the similarity with the maximum value from all the calculated similarities, and taking the occupation which is occupied by the employment corresponding to the similarity with the maximum value as the occupation which is likely to be occupied by the target personnel in the future.
6. The professional recommendation method based on a knowledge graph according to claim 1, wherein the calculation method of the learning ability value of the target person includes:
establishing a learning ability Gaussian function Gus according to data related to learning ability in professional data of an employment, wherein independent variables of the learning ability Gaussian function are learning ability values, and the dependent variables are probabilities of occurrence of the learning ability values;
calculating the learning ability value of the target person according to the learning time of the target person and the learning ability Gaussian function, wherein the calculation expression is as follows:
wherein,learning ability value representing the target person, +.>Indicating the learning time of the target person, +.>Representing a learning ability gaussian function; the learning ability value of the target person can reflect the learning ability of the target person, the learning ability value ranges from 0 to 1, and the closer to 1, the stronger the learning ability is.
7. The knowledge-graph-based professional recommendation method according to claim 1, wherein if the downstream neighboring node of the node a in the professional path is the node B, the calculating the passing rate between neighboring nodes of the professional path includes:
acquiring all adjacent upstream nodes of a D node and all adjacent downstream nodes of the D node in a knowledge graph, then respectively calculating the probability that professional data corresponding to each adjacent upstream node contains the D node and recording the probability as upstream node containing probability, and the probability that professional data corresponding to the D node contains each adjacent downstream node and recording the probability as downstream node containing probability;
and multiplying the sum of the probabilities contained in each upstream node by the sum of the probabilities contained in each downstream node to obtain the passing rate from the node A to the node D.
8. The professional recommendation method based on knowledge graph according to any one of claims 1 to 7, wherein for a certain professional path, calculating the required cost includes: respectively calculating the costs corresponding to each pair of adjacent nodes, and then adding the costs corresponding to each pair of adjacent nodes to obtain the cost required by the occupational path; for a neighboring node, its corresponding cost is inversely proportional to the pass rate corresponding to the pair of neighboring nodes and proportional to the time spent corresponding to the pair of neighboring nodes.
9. The knowledge-based professional recommendation method according to claim 8, wherein the calculation expression of the cost required for the job path is:
wherein,cost for representing professional path, +.>Representing the time from the kth node to the (k+1) th node in the professional path that the target person is likely to engage in,/for>Indicating the pass rate of the (k+1) th node in the professional path that the target person is likely to engage in,/-)>Representing the adjustment factors, the cost of the professional planning reflects the cost required for the target person from now on to his professional goal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410240353.2A CN117828194B (en) | 2024-03-04 | 2024-03-04 | Occupational recommendation method based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410240353.2A CN117828194B (en) | 2024-03-04 | 2024-03-04 | Occupational recommendation method based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117828194A true CN117828194A (en) | 2024-04-05 |
| CN117828194B CN117828194B (en) | 2024-06-21 |
Family
ID=90521327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410240353.2A Active CN117828194B (en) | 2024-03-04 | 2024-03-04 | Occupational recommendation method based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117828194B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180285823A1 (en) * | 2017-04-04 | 2018-10-04 | Linkedin Corporation | Ranking job offerings based on growth potential within a company |
| CN109002906A (en) * | 2018-06-25 | 2018-12-14 | 上海学民网络科技有限公司 | A kind of occupational planning path architecture system and processing method |
| US20190266497A1 (en) * | 2018-02-23 | 2019-08-29 | Microsoft Technology Licensing, Llc | Knowledge-graph-driven recommendation of career path transitions |
| US20190303798A1 (en) * | 2018-03-30 | 2019-10-03 | Microsoft Technology Licensing, Llc | Career path recommendation engine |
| KR20200129024A (en) * | 2019-05-07 | 2020-11-17 | (주)스마트소셜 | Apparatus and method for recommending job |
| CN115455205A (en) * | 2022-09-21 | 2022-12-09 | 深圳今日人才信息科技有限公司 | Time sequence knowledge graph-based occupational development planning method |
| CN116258397A (en) * | 2023-01-06 | 2023-06-13 | 上海卓越睿新数码科技股份有限公司 | Professional energy map construction method based on skill guidance |
| CN116452162A (en) * | 2023-04-04 | 2023-07-18 | 平安科技（深圳）有限公司 | Path planning method, device, equipment and storage medium |
| US20230245067A1 (en) * | 2022-01-31 | 2023-08-03 | John Baker | System and method for recommending potential careers or career paths |
| CN116596494A (en) * | 2023-05-18 | 2023-08-15 | 数应科技(浙江)有限公司 | Person post matching method and system based on knowledge map deep learning |
| CN117172978A (en) * | 2023-11-02 | 2023-12-05 | 北京国电通网络技术有限公司 | Learning path information generation method, device, electronic equipment and medium |
| US20240046393A1 (en) * | 2022-08-02 | 2024-02-08 | Montana Butsch | Individualized path recommendation engine based on personal characteristics |
| CN117541202A (en) * | 2023-11-14 | 2024-02-09 | 中电科新型智慧城市研究院有限公司 | Employment recommendation system based on multi-mode knowledge graph and pre-training large model fusion |
-
2024
- 2024-03-04 CN CN202410240353.2A patent/CN117828194B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180285823A1 (en) * | 2017-04-04 | 2018-10-04 | Linkedin Corporation | Ranking job offerings based on growth potential within a company |
| US20190266497A1 (en) * | 2018-02-23 | 2019-08-29 | Microsoft Technology Licensing, Llc | Knowledge-graph-driven recommendation of career path transitions |
| US20190303798A1 (en) * | 2018-03-30 | 2019-10-03 | Microsoft Technology Licensing, Llc | Career path recommendation engine |
| CN109002906A (en) * | 2018-06-25 | 2018-12-14 | 上海学民网络科技有限公司 | A kind of occupational planning path architecture system and processing method |
| KR20200129024A (en) * | 2019-05-07 | 2020-11-17 | (주)스마트소셜 | Apparatus and method for recommending job |
| US20230245067A1 (en) * | 2022-01-31 | 2023-08-03 | John Baker | System and method for recommending potential careers or career paths |
| US20240046393A1 (en) * | 2022-08-02 | 2024-02-08 | Montana Butsch | Individualized path recommendation engine based on personal characteristics |
| CN115455205A (en) * | 2022-09-21 | 2022-12-09 | 深圳今日人才信息科技有限公司 | Time sequence knowledge graph-based occupational development planning method |
| CN116258397A (en) * | 2023-01-06 | 2023-06-13 | 上海卓越睿新数码科技股份有限公司 | Professional energy map construction method based on skill guidance |
| CN116452162A (en) * | 2023-04-04 | 2023-07-18 | 平安科技（深圳）有限公司 | Path planning method, device, equipment and storage medium |
| CN116596494A (en) * | 2023-05-18 | 2023-08-15 | 数应科技(浙江)有限公司 | Person post matching method and system based on knowledge map deep learning |
| CN117172978A (en) * | 2023-11-02 | 2023-12-05 | 北京国电通网络技术有限公司 | Learning path information generation method, device, electronic equipment and medium |
| CN117541202A (en) * | 2023-11-14 | 2024-02-09 | 中电科新型智慧城市研究院有限公司 | Employment recommendation system based on multi-mode knowledge graph and pre-training large model fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117828194B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bairagi et al. | Research methodology: A practical and scientific approach | |
| Wu et al. | Sampling theory and practice | |
| Boukhelifa et al. | How data workers cope with uncertainty: A task characterisation study | |
| Choo et al. | Information seeking | |
| Davis et al. | Indicators as a technology of global governance | |
| Adams et al. | Research methods for graduate business and social science students | |
| Harper | Identity, intersectionality, and mixed-methods approaches | |
| Fişek et al. | Status characteristics and expectation states theory: A priori model parameters and test | |
| Asanbe et al. | Teachers’ performance evaluation in higher educational institution using data mining technique | |
| Cestari et al. | A case study extension methodology for performance measurement diagnosis in nonprofit organizations | |
| Koreff | Are auditors' reliance on conclusions from data analytics impacted by different data analytic inputs? | |
| Kale et al. | Introduction to research | |
| Williams et al. | Drivers of social engagement: Employee voice–advice sharing relationship | |
| CN117828194B (en) | Occupational recommendation method based on knowledge graph | |
| Chockkalingam et al. | Which one’s more work? Predicting effective credit hours between courses | |
| Juanamasta et al. | An integrative review of Indonesia’s quality of care | |
| Usman et al. | E-assessment and computer-aided prediction methodology for student admission test score | |
| Hackman | Seven maxims for institutional researchers: Applying cognitive theory and research | |
| Williams | Boundedness, Belonging and Becoming: Primary School Children's Perspectives of Education and Learning in the South Wales Valleys | |
| Ozonder | Longitudinal Analysis of Out-Of-Home Activity Generation in the Greater Toronto and Hamilton Area | |
| CN112686057B (en) | College student psychological consultation appointment recommendation method and system | |
| Li et al. | PLEAS: A Decision Support System for Personnel Evaluation and Selection | |
| Jones | A Case Study Exploration of Representative Bureaucracy and Self-Efficacy within the Senior Executive Service | |
| Yasmin | Enhancement in process mining: guideline for process owner and process analyst | |
| MacDonald | The information sharing behaviour of health service managers: a three-part study |
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 |