WO2020187168A1 - 求职简历推送方法与装置以及任务推送方法与装置 - Google Patents

求职简历推送方法与装置以及任务推送方法与装置 Download PDF

Info

Publication number
WO2020187168A1
WO2020187168A1 PCT/CN2020/079346 CN2020079346W WO2020187168A1 WO 2020187168 A1 WO2020187168 A1 WO 2020187168A1 CN 2020079346 W CN2020079346 W CN 2020079346W WO 2020187168 A1 WO2020187168 A1 WO 2020187168A1
Authority
WO
WIPO (PCT)
Prior art keywords
task
job
resume
job application
matching degree
Prior art date
Application number
PCT/CN2020/079346
Other languages
English (en)
French (fr)
Inventor
吴晓军
Original Assignee
河北冀联人力资源服务集团有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 河北冀联人力资源服务集团有限公司 filed Critical 河北冀联人力资源服务集团有限公司
Publication of WO2020187168A1 publication Critical patent/WO2020187168A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management

Definitions

  • the present disclosure relates to the field of data processing technology, and in particular, to a method and device for pushing job application resumes and a method and device for pushing tasks.
  • the purpose of the present disclosure includes providing a method and device for pushing a job application resume and a method and device for pushing a task, which can improve the accuracy of pushing resumes and tasks.
  • the embodiment of the present disclosure provides a method for pushing a job application resume, including:
  • each job resume in at least one job resume calculate the match between the task to be processed and the job resume as the first match, and calculate the match between at least one first job resume and the job resume as the first Second matching degree, and determining the matching score of the job application resume according to the first matching degree and the second matching degree, wherein each first job application resume is to complete a task with the same or similar content as the task within a historical period Of job applicants’ resumes; and
  • the job application resume pushing method further includes:
  • the degree of matching between the task to be processed and the job resume is calculated as the first degree of matching
  • the difference between the at least one first job resume and the job resume is calculated
  • the matching degree is regarded as the second matching degree, including:
  • For each job application resume category calculate the degree of match between the task and the center of the job application resume category, and use the degree of match between the task and the center of the job application resume category as the task and the job resume category
  • the matching degree between each job application resume is calculated, and the matching degree between the at least one first job application resume and the center of the job application resume is calculated, and the at least one first job application resume and the center of the job application resume are calculated.
  • the matching degree between the two is used as the matching degree between the at least one first job application resume and each job application resume in the job application resume category.
  • clustering the at least one job application resume to obtain at least one job application resume category includes:
  • the calculating the matching degree between the task and the center of the job application resume category includes:
  • the calculating the matching degree between the at least one first job application resume and the center of the job application resume includes:
  • For each first job application resume obtain the skill label of the first job application resume, and search for the score of the skill label in the center skill label of the job application resume class that matches any skill label of the first job application resume with;
  • the determining the matching score of the job application resume according to the first matching degree and the second matching degree includes:
  • a weighted summation is performed on the first matching degree and the second matching degree, and the result of the weighted summation is used as the matching score of the job resume.
  • the method of obtaining the skill label of the center of the job application resume category and the skill label of the task includes:
  • the job seeker is added by the system, the recruitment website or the task issuer after completing the task.
  • the process of extracting the work experience of the job applicant includes:
  • Extract keywords from the work experience of the job seeker perform semantic matching on all the extracted keywords with the skill tags in the preset skill tag library, and use the skill tags on the semantic matching as the reference from the job seeker
  • the skill tags in the skill tag library are tags related to job competence, including gender tags, age tags, professional tags, educational background tags, working years tags, and tags related to work content.
  • the process of adding the job seeker himself includes:
  • the job seeker may first select a task, and then the system pushes related skill tags to the job seeker for the job seeker to choose according to the task selected by the job seeker.
  • the skill tag added by the system, the recruitment website or the task issuer after the job seeker completes the task is the skill tag of the task completed by the job seeker, and the skill tag of the task is added by the task issuer.
  • Each skill label of the task can be set with different scores according to its importance.
  • the matching degree between the task and each first job application resume is greater than the first value.
  • the embodiment of the present disclosure also provides a task pushing method, including:
  • the matching degree between the job application resume to be processed and the task as the third matching degree
  • the matching score of the task is determined according to the third matching degree and the fourth matching degree, wherein each second job application resume is completed within a certain period of time with the same or similar content as the task The job applicant's resume of the job applicant;
  • the task pushing method before calculating the matching degree between the job application resume to be processed and the task as the third matching degree for each task in the at least one task, the task pushing method further includes:
  • the matching degree between the job application resume to be processed and the task is calculated as the third matching degree
  • the at least one second job application resume corresponding to the task is calculated.
  • the degree of matching between is regarded as the fourth degree of matching, including:
  • For each task category calculate the matching degree between the job application resume and the center of the task category as the matching degree between the job application resume and each task in the task category, and calculate the job application resume and the task
  • the matching degree between the at least one second job application resume corresponding to the center of the class is taken as the matching degree between the job application resume and the at least one second job application resume corresponding to each task in the task category.
  • clustering the at least one task to obtain at least one task class includes:
  • the vector corresponding to the at least one task is clustered, wherein all the vectors corresponding to the task can extract keywords from the content of the task and vectorize the keywords to calculate the vectorized result Combine it.
  • the calculating the matching degree between the job application resume and the center of the task category as the matching degree between the job application resume and each task in the task category includes:
  • the calculation of the degree of matching between the job application resume and the at least one second job application resume corresponding to the center of the task category is taken as one of the at least one second job application resume corresponding to each task in the task category.
  • determining the matching score of the task according to the third matching degree and the fourth matching degree includes:
  • the method of obtaining the skill label of the job application resume and the skill label of the center of the task category includes:
  • the task pushing method further includes:
  • the fourth matching degree corresponding to the task for obtaining the maximum matching score if it is determined that the fourth matching degree is less than the second value, the publisher of the job application resume is prompted to improve the job application resume information.
  • the matching degree between the task and each second job resume is greater than the first value.
  • the embodiment of the present disclosure also provides a device for pushing a job application resume, including:
  • the first calculation unit is used for calculating the matching degree between the task to be processed and the job application resume as the first matching degree for each job resume in at least one job application resume, and calculating at least one first job application resume and the job application resume
  • the matching degree between the two is regarded as the second matching degree
  • the matching score of the job application resume is determined according to the first matching degree and the second matching degree, wherein each first job application resume is completed in a historical period of time with the Resume of job applicants with the same or similar tasks;
  • the first pushing unit is configured to select a target job application resume from the at least one job application resume according to the matching score of the at least one job application resume, and push the target job application resume to the publisher of the task.
  • the embodiment of the present disclosure also provides a task pushing device, including:
  • the second calculation unit is configured to calculate the matching degree between the job application resume to be processed and the task as the third matching degree for each task in at least one task, and calculate the at least one first matching degree between the job application resume and the task.
  • the matching degree between job application resumes is taken as the fourth matching degree
  • the matching score of the task is determined according to the third matching degree and the fourth matching degree, where each second job application resume is completed and matched within a historical period of time. Resume of job applicants for tasks with the same or similar task content;
  • the second pushing unit is configured to select a target task from the at least one task according to the matching score of the at least one task, and push the target task to the publisher of the job application resume.
  • FIG. 1 is a schematic flowchart of an embodiment of a method for pushing a job application resume in the present disclosure
  • FIG. 2 is a schematic diagram of a part of the process of another embodiment of the method for pushing a job application resume of the present disclosure
  • Fig. 3 is a schematic flowchart of an embodiment of the task pushing method of the present disclosure
  • FIG. 4 is a schematic diagram of a part of the process of another embodiment of the task pushing method of the present disclosure.
  • FIG. 5 is a schematic structural diagram of an embodiment of a job application resume pushing device of the present disclosure
  • FIG. 6 is a schematic structural diagram of an embodiment of the task pushing device of the present disclosure.
  • the present disclosure provides a method for pushing job application resumes, including:
  • the aforementioned at least one job application resume may be obtained by obtaining a resume posted by a user on a recruitment website.
  • the number of first job application resumes can be set as required, for example, set to 5, which will not be repeated here.
  • the period of history can be set as needed, for example, set to the past 1 year, which will not be repeated here.
  • the degree of similarity or similarity between task contents can be measured by the similarity between task contents (the value can be a percentage). When the similarity is greater than a certain value (such as 80%), it indicates the contents of the corresponding two tasks Similarity: When the similarity is greater than a larger value (such as 99%), it means that the contents of the corresponding two tasks are the same.
  • the calculation of the similarity can be obtained by using mature algorithms in the prior art. For example, by comparing the keywords in the description of the two tasks, it can be calculated from the industry field, location, time arrangement, qualification requirements, etc. of the task. Comparing aspects and comprehensively comparing the results.
  • the selecting a target job application resume from the at least one job application resume according to the matching score of the at least one job application resume may include:
  • the matching scores of the at least one job application resume are sorted in descending order, and the job application resume corresponding to at least one of the previously preset number of matching scores is used as the target job application resume.
  • the preset value can be set as required, for example, set to 10, which will not be repeated here.
  • the job application resume pushing method provided by the embodiments of the present disclosure is based on the matching degree of the job applicant’s job application resume and the task, and the job applicant’s job application resume and the job application resume of the job applicant who has completed the same or similar tasks as the task content over a period of time.
  • the matching degree calculates the matching score of the job applicant’s job application resume, and pushes the resume based on the matching score of the job applicant’s job resume.
  • Figure 2 is a schematic diagram of a part of the process of another job application resume push method provided by the present disclosure.
  • the tasks to be processed and Before the matching degree between the job application resumes is used as the first matching degree it can also include:
  • the degree of matching between the task to be processed and the job resume is calculated as the first degree of matching
  • the difference between the at least one first job resume and the job resume is calculated
  • the matching degree, as the second matching degree can include:
  • clustering the at least one job application resume may be implemented by clustering the vector corresponding to the at least one job application resume.
  • the vector corresponding to the job resume can be obtained by extracting keywords from the content of the job resume, vectorizing the keywords, and combining the vectorized results.
  • the above-mentioned clustering calculation process can be realized with the help of mature clustering algorithms in the field of data mining, for example, k-means clustering algorithm, Canopy algorithm, hierarchical clustering algorithm, LDA algorithm, etc.
  • the matching degree between the at least one first job application resume and the center of the job application resume category is taken as one of each job application resume in the job application resume category to which the center of the at least one first job application resume and the job application resume category belongs Therefore, when calculating the matching degree between the task and the at least one job application resume and the matching degree between the at least one first job application resume and the at least one job application resume, it is not necessary to calculate the The matching degree between the task and each job resume in the at least one job application resume, and the matching degree between the at least one first job resume and each job resume in the at least one job resume, greatly reducing the first The calculation amount of the first matching degree and the second matching degree makes the method for pushing a resume for job opening provided by this embodiment improve the efficiency of pushing resumes compared with the method provided by the foregoing embodiment
  • the calculation of the matching degree between the task and the center of the job resume category may include:
  • said calculating the matching degree between the at least one first job application resume and the center of the job application resume category may include:
  • For each first job application resume obtain the skill label of the first job application resume, and search for the score of the skill label in the center skill label of the job application resume class that matches any skill label of the first job application resume with;
  • determining the matching score of the job application resume according to the first matching degree and the second matching degree may include:
  • a weighted summation is performed on the first matching degree and the second matching degree, and the result of the weighted summation is used as the matching score of the job resume.
  • the skill label of the job resume comes from at least one of the following three aspects: the first aspect is extracted from the job applicant’s work experience; the second aspect is added by the job applicant; the third aspect is After the job seeker completes the task, the system, recruitment website or task publisher will add it.
  • the process of extracting skill tags from the job seeker’s work experience includes: extracting keywords from the job seeker’s work experience, and combining the extracted keywords with those in the preset skill tag library.
  • the skill tags perform semantic matching, and the skill tags on the semantic matching are used as the skill tags extracted from the job seeker's work experience.
  • the skill tags in the skill tag library are tags related to the ability to be competent for the job, including gender tags, age tags, professional tags, academic qualification tags, working years tags, and tags related to job content. For example, assuming that the job content of a task includes managing and maintaining an Oracle database, the skill label of the task may include Oracle database management and Oracle database maintenance.
  • a skill tag library for tasks can be set.
  • the job seeker can first select a task, and then the system pushes related skill tags to the job seeker for the job seeker to choose according to the task selected by the job seeker. For example, if the job seeker chooses the task of confinement, then the website will push the health certificate, personality and other skill tags related to confinement to the job seeker for the job seeker to choose.
  • the skill tag added in the third aspect is the skill tag of the task completed by the job seeker, and the skill tag of the task can be added by the task publisher.
  • Each skill label of the task can be set with different scores according to its importance. For a certain skill label T1 of the job resume, if the skill label T1 matches a certain skill label T2 of the task, then the skill label T1 The score of is the score of skill label T2. If the skill label T1 matches certain two skill labels T20 and T21 of the task, the score of the skill label T1 is the greater of the score of the skill label T20 and the score of the skill label T21. If the skill tag T1 does not match any skill tag of the task, the score of the skill tag T1 is 0.
  • the score of T1 is the ratio of the score of the type to which T3 belongs, and the ratio of the number of skill tags under the type to which T3 belongs.
  • a certain skill label in the center of a job application resume is Oracle database management
  • the skill label related to the job content of the first job application resume includes Oracle database management and Oracle database maintenance, and it is related to the job content in advance.
  • the skill label of the Oracle database management center of the job resume category is assigned a value of 40. If the skill label T1 matches certain two skill labels T30 and T31 of the first job resume, the score of T1 is the ratio of the score of the type to which T30 belongs to the number of skill labels in the type to which T30 belongs, and T31 The larger the ratio of the score of the category to which T31 belongs to the number of skill tags in the category to which T31 belongs. If the skill label T1 does not match any skill label of the first job application resume, the score of the skill label T1 is 0.
  • a certain skill label in the center of a certain job resume category is MySQL database management
  • the above are just examples.
  • various existing algorithms can also be used to customize and optimize the assignment process, so as to customize targeted assignment schemes for different users or recruitment situations.
  • the weight of the first matching degree and the weight of the second matching degree are values in the range of 0 to 1, and the sum of the two is 1.
  • the matching degree between the task and each first job application resume is greater than the first value.
  • the first value can be set as required, and will not be repeated here. Limiting the matching degree between the task and each first job application resume to be greater than the first value can ensure that the first job application resume is a job application resume that matches the task relatively, so that the calculated second matching degree is more accurate, This makes the present job application resume push method more accurate than the aforementioned methods.
  • the present disclosure also provides a task pushing method, including:
  • At least one task can be obtained by obtaining a task posted by a user or an enterprise on a recruitment website.
  • the number of second job application resumes can be set as required, for example, set to 5, which will not be repeated here.
  • the period of history can be set as needed, for example, set to the past 1 year, which will not be repeated here.
  • the same or similar task content can be measured by the similarity between the task content (the value can be a percentage). When the similarity is greater than a certain value (such as 80%), it means that the content of the corresponding two tasks are similar. ; When the similarity is greater than a larger value (such as 99%), it means that the content of the corresponding two tasks is the same.
  • the calculation of the similarity between tasks here can refer to the description in the foregoing embodiment, which will not be repeated here.
  • the selecting a target task from the at least one task according to the matching score of the at least one task may include:
  • the matching scores of the at least one task are sorted in descending order, and a task corresponding to at least one matching score among the previously preset number of matching scores is taken as the target task.
  • the preset value can be set as required, for example, set to 10, which will not be repeated here.
  • the task push method provided by the embodiments of the present disclosure is based on the matching degree of the job applicant’s job application resume with the task, and the match between the job applicant’s job resume and the job applicant’s resume of the job applicant who has completed the same or similar tasks as the task content over a period of time. Calculate the matching score of the task and push the task based on the matching score of the task.
  • the whole solution not only considers the job applicant’s
  • the matching degree of the job application resume with the task also considers the matching degree of the job applicant’s job application resume with the job applicant’s resume of the job applicant who has completed the same or similar tasks as the task in a certain period of time. This makes the tasks calculated by this scheme better
  • the matching score is more accurate, which improves the accuracy of task push.
  • Figure 4 is a schematic diagram of a part of the process of another embodiment of the task pushing method of the present disclosure.
  • the job application resume to be processed is calculated and Before the matching degree between the tasks is used as the third matching degree, it can also include:
  • the matching degree between the job application resume to be processed and the task is calculated as the third matching degree
  • the at least one second job application resume corresponding to the task is calculated.
  • the degree of matching between is regarded as the fourth degree of matching, which can include:
  • the clustering of the at least one task may be implemented by clustering the vector corresponding to the at least one task.
  • the above-mentioned clustering calculation process can be realized with the help of mature clustering algorithms in the field of data mining, for example, k-means clustering algorithm, Canopy algorithm, hierarchical clustering algorithm, LDA algorithm, etc.
  • the vector corresponding to the task can be obtained by extracting keywords from the content of the task, vectorizing the keywords, and combining the vectorized results.
  • the so-called vectorization refers to the conversion of text into meaningful digital vectors (or arrays).
  • NLP natural language processing
  • word2vec words Vector
  • NNLM neural network language model
  • context and target words context&word, C&W
  • other technologies such as using words Vector (word2vec), neural network language model (Neural Network Language Model, NNLM), context and target words (context&word, C&W) and other technologies.
  • the matching degree between the job application resume and the center of the task category is used as at least one second job application resume corresponding to each task in the task category to which the center of the task category belongs.
  • the calculating the matching degree between the job application resume and the center of the task category as the matching degree between the job application resume and each task in the task category may include:
  • the calculation of the degree of matching between the job application resume and the at least one second job application resume corresponding to the center of the task category is taken as one of the at least one second job application resume corresponding to each task in the task category.
  • the degree of match between can include:
  • determining the matching score of the task according to the third matching degree and the fourth matching degree may include:
  • the skill label of the job application resume comes from at least one of the following three aspects: the first aspect is extracted from the job applicant’s work experience; the second aspect is added by the job applicant; The aspect is that the job seeker is added by the system, the recruitment website or the job publisher after completing the task.
  • the score value of each skill label of the job application resume can be set as needed. For a certain skill label T4 in the center of the task category, if the skill label T4 matches a certain skill label T5 of the job resume, then the skill label The T4 score is the score of the skill label T5. If the skill tag T4 matches certain two skill tags T50 and T51 of the job application resume, the score of the skill tag T4 is the larger of the score of the skill tag T50 and the score of the skill tag T51. If the skill label T4 does not match any skill label of the job application resume, the score of the skill label T4 is 0.
  • T6 is the ratio of the score of the type to which T5 belongs, and the ratio of the number of skill tags under the type to which T5 belongs.
  • a certain skill tag of the second job application resume is Oracle database management
  • the skill tags related to the job content of the job application resume include Oracle database management and Oracle database maintenance, and are skills related to the job content in advance. If the label is assigned a value of 80, the Oracle database management skill label of the second job resume has a score of 40.
  • the score of T6 is the ratio of the score of the type to which T50 belongs to the number of skill tags in the type to which T50 belongs, and The larger the ratio of the score of the category to which T51 belongs to the number of skill tags in the category to which T51 belongs. If any skill tag of the job application resume does not match the skill tag T6, the score of T6 is 0.
  • the weight of the third matching degree and the weight of the fourth matching degree are values in the range of 0 to 1, and the sum of the two is 1.
  • the method may further include:
  • the fourth matching degree corresponding to the task for obtaining the maximum matching score if it is determined that the fourth matching degree is less than the second value, the publisher of the job application resume is prompted to improve the job application resume information.
  • the second value can be set as required, and will not be repeated here.
  • the publisher of the job application resume is prompted to improve the job application resume information.
  • the matching degree between the task and each second job resume is greater than the first value.
  • the first value can be set as required, and will not be repeated here. Limiting the matching degree between the task and each second job application resume to be greater than the first value can ensure that the second job application resume is a job application resume that matches the task, so that the calculated fourth matching degree is more accurate. Compared with the foregoing embodiments, this embodiment improves the accuracy of task pushing.
  • the present disclosure also provides a device for pushing job application resumes, including:
  • the first calculation unit 50 is used for calculating the matching degree between the task to be processed and the job application resume as the first matching degree for each job application resume in the at least one job application resume, and calculating at least one first job application resume and the job application resume.
  • the matching degree between the resumes is used as the second matching degree, and the matching score of the job application resume is determined according to the first matching degree and the second matching degree, wherein each first job application resume is completed and matched within a certain period of history. Resume of job applicants who describe the same or similar tasks; and
  • the first pushing unit 51 is configured to select a target job application resume from the at least one job application resume according to the matching score of the at least one job application resume, and push the target job application resume to the publisher of the task.
  • At least one job application resume may be obtained by obtaining the resume posted by the user on the recruitment website.
  • the number of historical period of time and the number of first job resumes can be set as needed, and will not be repeated here.
  • the same or similar task content can be measured by the similarity between the task content (the value can be a percentage). When the similarity is greater than a certain value (such as 80%), it means that the content of the corresponding two tasks are similar. ; When the similarity is greater than a larger value (such as 99%), it means that the content of the corresponding two tasks is the same.
  • the selecting a target job application resume from the at least one job application resume according to the matching score of the at least one job application resume may include:
  • the matching scores of the at least one job application resume are sorted in descending order, and the job application resume corresponding to at least one of the matching scores of the previous preset values is pushed to the publisher of the task.
  • the preset value can be set as required, and will not be repeated here.
  • the job application resume pushing device is based on the matching degree between the job applicant’s job application resume and the task, as well as the job applicant’s job application resume and the job application resume of the job applicant who has completed the same or similar tasks as the task content over a period of time.
  • the matching degree calculates the matching score of the job applicant’s job application resume, and pushes the resume based on the matching score of the job applicant’s job resume.
  • a task pushing device including:
  • the second calculation unit 60 is configured to, for each task in at least one task, calculate the matching degree between the job application resume to be processed and the task as a third matching degree, and calculate at least one of the job application resume and the task
  • the matching degree between the second job application resumes is taken as the fourth matching degree, and the matching score of the task is determined according to the third matching degree and the fourth matching degree, where each second job application resume is completed within a certain period of history Resume of job applicants for tasks with the same or similar content as the task;
  • the second pushing unit 61 is configured to select a target task from the at least one task according to the matching score of the at least one task, and push the target task to the publisher of the job application resume.
  • At least one task can be obtained by obtaining a task posted by a user or an enterprise on a recruitment website.
  • the number and historical period of the second job application resume can be set as needed, so I won’t repeat them here.
  • the same or similar task content can be measured by the similarity between the task content (the value can be a percentage). When the similarity is greater than a certain value (such as 80%), it means that the content of the corresponding two tasks are similar. ; When the similarity is greater than a larger value (such as 99%), it means that the content of the corresponding two tasks is the same.
  • the selecting a target task from the at least one task according to the matching score of the at least one task may include:
  • the matching scores of the at least one task are sorted in descending order, and the task corresponding to at least one matching score among the matching scores of the previous preset value is taken as the target task.
  • the preset value can be set as required, and will not be repeated here.
  • each device described above is merely illustrative.
  • the division of the units may be a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or integrated. To another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of units or modules, and may be in electrical or other forms.
  • the functions of the above units are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium.
  • the technical solution of the present disclosure essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present disclosure.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code .
  • the task pushing device is based on the matching degree of the job applicant’s job application resume with the task, and the matching of the job applicant’s job resume with the job applicant’s resume of the job applicant who has completed the same or similar tasks as the task in a period of time Calculate the matching score of the task and push the task based on the matching score of the task.
  • the embodiments of the present disclosure provide a method and device for pushing a job application resume and a method and device for pushing a task, which can improve the accuracy of resume and task pushing, so as to effectively alleviate the problem that the existing task pushing and resume pushing functions cannot achieve accurate recommendation functions. technical problem.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本公开提供了一种求职简历推送方法与装置以及任务推送方法与装置,能提高简历和任务推送的准确度。所述求职简历推送方法,包括:对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。

Description

求职简历推送方法与装置以及任务推送方法与装置
相关申请交叉引用
本申请要求于2019年03月15日提交中国专利局的申请号为201910200072.3、名称为“求职简历推送方法与装置以及任务推送方法与装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本公开涉及数据处理技术领域,具体而言,涉及一种求职简历推送方法与装置以及任务推送方法与装置。
背景技术
目前,用户或企业在发布月嫂、保姆、司机等任务时,招聘网站往往计算任务和求职简历之间的匹配度,根据匹配度进行相应的任务推送和简历推送。但这种方法的匹配度仅依据任务信息和求职者的简历信息进行计算,而当求职者的简历信息不太丰满时,会使计算出的匹配度较低,造成求职简历与任务的匹配效果较差,使得现有的任务推送和简历推送功能无法实现精准推荐的功能。
发明内容
有鉴于此,本公开的目的包括提供一种求职简历推送方法与装置以及任务推送方法与装置,能提高简历和任务推送的准确度。
本公开实施例提供了一种求职简历推送方法,包括:
对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
可选地,在所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度之前,所述求职简历推送方法还包括:
对所述至少一个求职简历进行聚类得到至少一个求职简历类;
其中,所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,包括:
对于每一个求职简历类,计算所述任务与该求职简历类的中心之间的匹配度,将所述任务与该求职简历类的中心之间的匹配度作为所述任务与该求职简历类中每一个求职简历之间的匹配度,并计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,将所述至少一个第一求职简历与该求职简历类的中心之间的匹配度作为所述至少一个第一求职简历与该求职简历类中每一个求职简历之间的匹配度。
可选地,对所述至少一个求职简历进行聚类得到至少一个求职简历类包括:
对所述至少一个求职简历对应的向量进行聚类,其中所述至少一个求职简历对应的向量通过对求职简历的内容提取关键词,并对关键词进行向量化,对向量化的结果进行组合得到。
可选地,所述计算所述任务与该求职简历类的中心之间的匹配度,包括:
获取该求职简历类的中心的技能标签以及所述任务的技能标签,对该求职简历类的中心的技能标签中与所述任务的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述任务与该求职简历类的中心之间的匹配度;
其中,所述计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,包括:
对于每一个第一求职简历,获取该第一求职简历的技能标签,对该求职简历类的中心的技能标签中与该第一求职简历的任一技能标签相匹配的技能标签的分值进行求和;
对所述至少一个第一求职简历对应的求和结果求均值,将求均值的结果作为所述至少一个第一求职简历与该求职简历类的中心之间的匹配度;
其中,所述根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,包括:
对所述第一匹配度和第二匹配度进行加权求和,并将加权求和的结果作为该求职简历的匹配得分。
可选地,所述获取该求职简历类的中心的技能标签以及所述任务的技能标签的方式包括:
从求职者的工作经历中提取;
所述求职者自己添加;以及
所述求职者完成任务后由系统、招聘网站或任务发布方添加。
可选地,所述从求职者的工作经历中提取的过程包括:
从所述求职者的工作经历中提取关键词,将提取出的所以所述关键词与预设的技能标签库中的技能标签进行语义匹配,将语义匹配上的技能标签作为从所述求职者的工作经历中提取的所述技能标签;
其中所述技能标签库中的技能标签为与能否胜任工作相关的标签,包括性别标签、年龄标签、专业标签、学历标签、工作年限标签以及与工作内容相关的标签。
可选地,所述求职者自己添加的过程包括:
所述求职者可以首先选择任务,然后,由系统根据所述求职者选择的所述任务向所述求职者推送相关的技能标签供求职者选择。
可选地,所述所述求职者完成任务后由系统、招聘网站或任务发布方添加的技能标签为求职者完成的任务的技能标签,所述任务的技能标签由所述任务发布方自行添加;
其中所述任务的各个技能标签可以根据其重要程度设置不同的分值。
可选地,所述任务与每一个第一求职简历之间的匹配度均大于第一数值。
本公开实施例还提供了一种任务推送方法,包括:
对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
可选地,在所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度之前,所述任务推送方法还包括:
对所述至少一个任务进行聚类得到至少一个任务类;
其中,所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,包括:
对于每一个任务类,计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,并计算所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度。
可选地,对所述至少一个任务进行聚类得到至少一个任务类包括:
对所述至少一个任务对应的向量进行聚类,其中所以所述任务对应的向量可以通过对所述任务的内容提取关键词,并对所述关键词进行向量化,对所述向量化的结果进行组合得到。
可选地,所述计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,包括:
获取所述求职简历的技能标签以及该任务类的中心的技能标签,对该任务类的中心的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述求职简历与该任务类的中心之间的匹配度;
其中,所述计算所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度,包括:
对于每一个第二求职简历,获取该第二求职简历的技能标签,对该第二求职简历的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和;
对所述至少一个第二求职简历对应的求和结果求均值,将求均值的结果作为所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度;
其中,所述根据所述第三匹配度和第四匹配度确定该任务的匹配得分,包括:
对所述第三匹配度和第四匹配度进行加权求和,并将加权求和的结果作为该任务的匹配得分。
可选地,所述获取所述求职简历的技能标签以及该任务类的中心的技能标签的方式包括:
从求职者的工作经历中提取的;
求职者自己添加;以及
求职者完成任务后由系统、招聘网站或任务发布方添加。
可选地,所述任务推送方法还包括:
获取最大匹配得分的任务对应的第四匹配度,若判断获知所述第四匹配度小于第二数值,则提示所述求职简历的发布者完善求职简历信息。
可选地,所述任务与每一个第二求职简历之间的匹配度均大于第一数值。
本公开实施例还提供了一种求职简历推送装置,包括:
第一计算单元,用于对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
第一推送单元,用于根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
本公开实施例还提供了一种任务推送装置,包括:
第二计算单元,用于对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
第二推送单元,用于根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
附图说明
本公开可以通过参考下文中结合附图所给出的描述而得到更好的理解,附图连同下面的详细说明一起包含在本说明书中并且形成本说明书的一部分。在附图中:
图1为本公开求职简历推送方法一实施例的流程示意图;
图2为本公开求职简历推送方法另一实施例的部分流程示意图;
图3为本公开任务推送方法一实施例的流程示意图;
图4为本公开任务推送方法另一实施例的部分流程示意图;
图5为本公开求职简历推送装置一实施例的结构示意图;
图6为本公开任务推送装置一实施例的结构示意图。
附图标号:
50-第一计算单元;51-第一推送单元;60-第二计算单元;61-第二推送单元。
具体实施方式
在下文中将结合附图对本公开内容的示例性实施方式进行描述。为了清楚和简明起见,在说明书中并未描述实际实施方式的所有特征。然而,应该了解,在开发任何这种实际实施方式的过程中可以做出很多特定于实施方式的决定,以便实现开发人员的具体目标,并且这些决定可能会随着实施方式的不同而有所改变。
因此,以下对在附图中提供的本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围,而是仅仅表示本公开的选定实施例。基于本公开中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
需要说明的是,本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对 于这些过程、方法、产品或设备固有的其它步骤或单元。
在此,还需要说明的一点是,为了避免因不必要的细节而模糊了本公开,在附图中仅仅示出了与根据本公开内容的方案密切相关的装置结构,而省略了与本公开关系不大的其他细节。
应理解的是,本公开内容并不会由于如下参照附图的描述而只限于所描述的实施形式。在本文中,在可行的情况下,实施方式可以相互组合、不同实施方式之间的特征替换或借用、在一个实施方式中省略一个或多个特征。
参看图1,本公开提供了一种求职简历推送方法,包括:
S10、对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
S11、根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
本实施例中,上述至少一个求职简历可以通过获取用户在招聘网站上投递的简历得到。第一求职简历的数量可以根据需要设置,比如设置为5个,此处不再赘述。历史一段时间可以根据需要设置,比如设置为过去1年,此处不再赘述。任务内容之间的相同或相似程度可以以任务内容之间的相似度(其取值可以为百分比)来衡量,当相似度大于某一个数值(比如80%),说明相应的两个任务的内容相似;当相似度大于某一个较大的数值(比如99%),说明相应的两个任务的内容相同。其中,该相似度的计算可以采用现有技术中的成熟算法计算得到,例如,通过比较两个工作任务描述中的关键字,从诸如工作任务的行业领域,地点,时间安排,资质要求等多方面进行比较并综合比较结果而得到。
所述根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,可以包括:
对所述至少一个求职简历的匹配得分按照从大到小的顺序进行排序,并将前预设数值个匹配得分中的至少一个匹配得分对应的求职简历作为所述目的求职简历。所述预设数值可以根据需要设置,比如设置为10,此处不再赘述。
本公开实施例提供的求职简历推送方法,基于求职者的求职简历与任务的匹配度,以及求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度计算求职者的求职简历的匹配得分,并基于求职者的求职简历的匹配得分进行简历推送,相较于在计算匹配得分时仅考虑求职者的求职简历与任务的匹配度的现有技术,整个方案不仅考虑了求职者的求职简历与任务的匹配度,还考虑了求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度,这就使得基于本方案计算出的求职者的求职简历的匹配得分较为准确,因而提高了简历推送的准确度。
图2为本公开提供的另一求职简历推送方法的部分流程示意图,参看图2,在前述方法的基础上,在所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度之前,还可以包括:
S20、对所述至少一个求职简历进行聚类得到至少一个求职简历类;
其中,所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,可以包括:
S21、对于每一个求职简历类,计算所述任务与该求职简历类的中心之间的匹配度,将所述任务与该求职简历类的中心之间的匹配度作为所述任务与该求职简历类中每一个求职简历之间的匹配度,并计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,将所述至少一个第一求职简历与该求职简历类的中心之间的匹配度作为所述至少一个第一求职简历与该求职简历类中每一个求职简历之间的匹配度。
本实施例中,对所述至少一个求职简历进行聚类可以通过对所述至少一个求职简历对应的向量进行聚类实现。求职简历对应的向量可以通过对求职简历的内容提取关键词,并对关键词进行向量化,对向量化的结果进行组合得到。上述聚类的计算过程可以借助于数据挖掘领域中的成熟聚类算法实现,例如使用k-means聚类算法、Canopy算法、层次聚类算法、LDA算法等。通过对所述至少一个求职简历进行聚类,并计算所述 任务与求职简历类的中心之间的匹配度作为所述任务与求职简历类的中心所属的求职简历类中每一个求职简历之间的匹配度,以及所述至少一个第一求职简历与求职简历类的中心之间的匹配度作为所述至少一个第一求职简历与求职简历类的中心所属的求职简历类中每一个求职简历之间的匹配度,从而在计算所述任务与所述至少一个求职简历之间的匹配度以及所述至少一个第一求职简历与所述至少一个求职简历之间的匹配度时不需要计算所述任务与所述至少一个求职简历中每一个求职简历之间的匹配度,以及所述至少一个第一求职简历与所述至少一个求职简历中每一个求职简历之间的匹配度,大大减少了第一匹配度和第二匹配度的计算量,这就使得本实施例提供的开求职简历推送方法相较于前述实施例提供的方法,提高了简历推送效率。
在前述方法的基础上,所述计算所述任务与该求职简历类的中心之间的匹配度,可以包括:
获取该求职简历类的中心的技能标签以及所述任务的技能标签,对该求职简历类的中心的技能标签中与所述任务的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述任务与该求职简历类的中心之间的匹配度;
其中,所述计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,可以包括:
对于每一个第一求职简历,获取该第一求职简历的技能标签,对该求职简历类的中心的技能标签中与该第一求职简历的任一技能标签相匹配的技能标签的分值进行求和;
对所述至少一个第一求职简历对应的求和结果求均值,将求均值的结果作为所述至少一个第一求职简历与该求职简历类的中心之间的匹配度;
其中,所述根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,可以包括:
对所述第一匹配度和第二匹配度进行加权求和,并将加权求和的结果作为该求职简历的匹配得分。
本实施例中,求职简历的技能标签来自于以下三个方面中的至少一个方面:第一方面是从求职者的工作经历中提取的;第二方面是求职者自己添加的;第三方面是求职者完成任务后由系统、招聘网站或任务发布方添加。
对于第一方面,需要说明的是,从求职者的工作经历中提取技能标签的过程包括:从求职者的工作经历中提取关键词,将提取出的关键词与预设的技能标签库中的技能标签进行语义匹配,将语义匹配上的技能标签作为从求职者的工作经历中提取的技能标签。技能标签库中的技能标签为与能否胜任工作相关的标签,包括性别标签、年龄标签、专业标签、学历标签、工作年限标签以及与工作内容相关的标签。比如假设一任务的工作内容包括管理与维护Oracle数据库,则该任务的技能标签可以包括Oracle数据库管理和Oracle数据库维护。
对于第二方面,需要说明的是,对于一些如月嫂、保姆、司机等学历要求不高的任务,为了使求职者能顺利添加技能标签,可设置任务的技能标签库。在求职者进行技能标签添加时,求职者可以首先选择任务,然后,由系统根据求职者选择的任务向求职者推送相关的技能标签供求职者选择。比如求职者选择的任务是月嫂,那么网站向该求职者推送健康证、性格等与月嫂相关的技能标签供该求职者选择。
第三方面添加的技能标签为求职者完成的任务的技能标签,而任务的技能标签可以由任务发布者自行添加。
所述任务的各个技能标签可以根据其重要程度设置不同的分值,对于该求职简历的某一个技能标签T1,如果技能标签T1与所述任务的某个技能标签T2相匹配,则技能标签T1的分值即为技能标签T2的分值。如果技能标签T1与所述任务的某两个技能标签T20和T21均相匹配,则技能标签T1的分值为技能标签T20的分值和技能标签T21的分值中的较大值。而如果技能标签T1与所述任务的任一个技能标签均不相匹配,则技能标签T1的分值为0。
需要说明的是,在计算所述第二匹配度之前,可以对每一类型技能标签分别赋予不同的分值,则对于某一个求职简历类的中心的某一个技能标签T1,以及第一求职简历的某一个技能标签T3,如果T1和T3相匹配,则T1的分值为T3所属的类型的分值,与T3所属的类型下的技能标签的数量的比值。举例来说,假设某一个求职简历类的中心的某一个技能标签为Oracle数据库管理,第一求职简历的与工作内容相关的技能标签包括Oracle数据库管理和Oracle数据库维护,且预先为与工作内容相关的技能标签赋值为80,则该求职简历类的中心的Oracle数据库管理这一技能标签的分值为40。如果技能标签T1和第一求职简历的某两个技能标签T30和T31相匹配,则T1的分值为T30所 属的类型的分值与T30所属的类型下的技能标签的数量的比值,与T31所属的类型的分值与T31所属的类型下的技能标签的数量的比值中的较大值。而如果技能标签T1与第一求职简历的任一个技能标签均不相匹配,则技能标签T1的分值为0。
需要注意的是,出于方便说明的目的,仅仅选取了较为简单的一种情况用于举例说明以上的对各技能标签赋予分值的过程。在实际中实施本公开提供的种求职简历推送方法时,还可以根据进行招聘的用户或企业的需求以及待处理简历的实际情况设置各种对技能标签赋予分值的规则。可选地,如果对于某一个求职简历类的中心的某一个技能标签T1部分地对应于第一求职简历的某一个技能标签T3,则可以设定T1得到的分值为T3所属的类型的分值的一预设部分。假设某一个求职简历类的中心的某一个技能标签为MySQL数据库管理,则也可以设定该求职简历可以得到原技能标签对应的分数40的一部分,例如30%。以上仅仅是实例性的,在实际应用中,还可以使用各种现有的算法对该赋值过程进行定制优化,从而为不同的用户或招聘情况,定制出针对性的赋值方案。
可以理解的是,所述第一匹配度的权重和第二匹配度的权重为0至1内的数值,且二者的和为1。
在前述方法实施例的基础上,所述任务与每一个第一求职简历之间的匹配度均大于第一数值。
本实施例中,第一数值可以根据需要设置,此处不再赘述。限定所述任务与每一个第一求职简历之间的匹配度均大于第一数值,能够保证第一求职简历为与所述任务比较匹配的求职简历,使得计算出的第二匹配度较为准确,这就使得本求职简历推送方法相较于前述方法,提高了简历推送的准确度。
参看图3,本公开还提供了一种任务推送方法,包括:
S30、对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
S31、根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
本实施例中,至少一个任务可以通过获取用户或企业在招聘网站上发布的任务得到。第二求职简历的数量可以根据需要设置,比如设置为5个,此处不再赘述。历史一段时间可以根据需要设置,比如设置为过去1年,此处不再赘述。任务内容之间的相同或相似可以以任务内容之间的相似度(其取值可以为百分比)来衡量,当相似度大于某一个数值(比如80%),说明相应的两个任务的内容相似;当相似度大于某一个较大的数值(比如99%),说明相应的两个任务的内容相同。其中,此处任务之间相似度的计算可以参见前述实施方式中的描述,在此不再赘述。
所述根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,可以包括:
对所述至少一个任务的匹配得分按照从大到小的顺序进行排序,并将前预设数值个匹配得分中的至少一个匹配得分对应的任务作为所述目的任务。所述预设数值可以根据需要设置,比如设置为10,此处不再赘述。
本公开实施例提供的任务推送方法,基于求职者的求职简历与任务的匹配度,以及求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度计算任务的匹配得分,并基于任务的匹配得分进行任务推送,相较于在计算匹配得分时仅考虑求职者的求职简历与任务的匹配度的现有技术,整个方案不仅考虑了求职者的求职简历与任务的匹配度,还考虑了求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度,这就使得本方案计算出的任务的匹配得分较为准确,因而提高了任务推送的准确度。
图4为本公开任务推送方法另一实施例的部分流程示意图,参看图4,在前述方法实施例的基础上,在所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度之前,还可以包括:
S40、对所述至少一个任务进行聚类得到至少一个任务类;
其中,所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,可以包括:
S41、对于每一个任务类,计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,并计算所述求职简历与该任务类 的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度。
本实施例中,对所述至少一个任务进行聚类可以通过对所述至少一个任务对应的向量进行聚类实现。上述聚类的计算过程可以借助于数据挖掘领域中的成熟聚类算法实现,例如使用k-means聚类算法、Canopy算法、层次聚类算法、LDA算法等。任务对应的向量可以通过对任务的内容提取关键词,并对关键词进行向量化,对向量化的结果进行组合得到。其中,所谓向量化指的是就是将文本转换为有意义的数字向量(或数组),上述向量化的运算过程可以借助于自然语言处理(NLP)领域中的常用技术手段来实现,例如使用词向量(word2vec)、神经网络语言模型(Neural Network Language Model,NNLM)、上下文和目标词(context&word,C&W)等技术。通过对所述至少一个任务进行聚类,并计算所述求职简历与任务类的中心之间的匹配度作为所述求职简历与任务类的中心所属的任务类中每一个任务之间的匹配度,以及所述求职简历与任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与任务类的中心所属的任务类中每一个任务对应的至少一个第二求职简历之间的匹配度,从而在计算所述求职简历与所述至少一个任务之间的匹配度以及所述求职简历与所述至少一个任务对应的至少一个第二求职简历之间的匹配度时不需要计算所述求职简历与所述至少一个任务中每一个任务之间的匹配度,以及所述求职简历与所述至少一个任务中每一个任务对应的至少一个第二求职简历之间的匹配度,大大减少了第三匹配度和第四匹配度的计算量,这就使得本实施例相较于前述实施例,提高了任务推送效率。
在前述方法实施例的基础上,所述计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,可以包括:
获取所述求职简历的技能标签以及该任务类的中心的技能标签,对该任务类的中心的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述求职简历与该任务类的中心之间的匹配度;
其中,所述计算所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度,可以包括:
对于每一个第二求职简历,获取该第二求职简历的技能标签,对该第二求职简历的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和;
对所述至少一个第二求职简历对应的求和结果求均值,将求均值的结果作为所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度;
其中,所述根据所述第三匹配度和第四匹配度确定该任务的匹配得分,可以包括:
对所述第三匹配度和第四匹配度进行加权求和,并将加权求和的结果作为该任务的匹配得分。
本实施例中,所述求职简历的技能标签来自于以下三个方面中的至少一个方面:第一方面是从求职者的工作经历中提取的;第二方面是求职者自己添加的;第三方面是求职者完成任务后由系统、招聘网站或任务发布方添加。
第一方面、第二方面和第三方面生成技能标签的过程与前述求职简历推送方法实施例一致,此处不再赘述。
所述求职简历的各个技能标签的分值可以根据需要设置,对于该任务类的中心的某一个技能标签T4,如果技能标签T4与所述求职简历的某个技能标签T5相匹配,则技能标签T4的分值即为技能标签T5的分值。如果技能标签T4与所述求职简历的某两个技能标签T50和T51均相匹配,则技能标签T4的分值为技能标签T50的分值和技能标签T51的分值中的较大值。而如果技能标签T4与所述求职简历的任一个技能标签均不相匹配,则技能标签T4的分值为0。
需要说明的是,在计算所述第四匹配度之前,可以对每一类型技能标签分别赋予不同的分值,则对于所述求职简历的某一个技能标签T5,以及该第二求职简历的某一个技能标签T6,如果T5和T6相匹配,则T6的分值为T5所属的类型的分值,与T5所属的类型下的技能标签的数量的比值。举例来说,假设该第二求职简历的某一个技能标签为Oracle数据库管理,所述求职简历的与工作内容相关的技能标签包括Oracle数据库管理和Oracle数据库维护,且预先为与工作内容相关的技能标签赋值为80,则该第二求职简历的Oracle数据库管理这一技能标签的分值为40。如果所述求职简历的某两个技能标签T50和T51均与技能标签T6相匹配,则T6的分值为T50所属的类型的分值与T50所属的类型下的技能标签的数量的比值,与T51所属的类型的分值与T51所属的类型下 的技能标签的数量的比值中的较大值。如果所述求职简历的任一个技能标签均与技能标签T6不相匹配,则T6的分值为0。
以上的对各技能标签赋予分值的过程可以参见前文中的相关描述,故在此不再赘述。
可以理解的是,所述第三匹配度的权重和第四匹配度的权重为0至1内的数值,且二者的和为1。
在前述方法实施例的基础上,所述方法还可以包括:
获取最大匹配得分的任务对应的第四匹配度,若判断获知所述第四匹配度小于第二数值,则提示所述求职简历的发布者完善求职简历信息。
本实施例中,第二数值可以根据需要设置,此处不再赘述。当最大匹配得分的任务对应的第四匹配度小于第二数值,说明没有与所述求职简历比较匹配的任务,则提示所述求职简历的发布者完善求职简历信息。
在前述方法实施例的基础上,该任务与每一个第二求职简历之间的匹配度均大于第一数值。
本实施例中,第一数值可以根据需要设置,此处不再赘述。限定该任务与每一个第二求职简历之间的匹配度均大于第一数值,能够保证第二求职简历为与该任务比较匹配的求职简历,使得计算出的第四匹配度较为准确,这就使得本实施例相较于前述实施例,提高了任务推送的准确度。
参看图5,本公开还提供了一种求职简历推送装置,包括:
第一计算单元50,用于对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
第一推送单元51,用于根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
本实施例中,至少一个求职简历可以通过获取用户在招聘网站上投递的简历得到。历史一段时间和第一求职简历的数量可以根据需要设置,此处不再赘述。任务内容之间 的相同或相似可以以任务内容之间的相似度(其取值可以为百分比)来衡量,当相似度大于某一个数值(比如80%),说明相应的两个任务的内容相似;当相似度大于某一个较大的数值(比如99%),说明相应的两个任务的内容相同。
所述根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,可以包括:
对所述至少一个求职简历的匹配得分按照从大到小的顺序进行排序,并将前预设数值各匹配得分中的最少一个匹配得分对应的求职简历推送给所述任务的发布者。所述预设数值可以根据需要设置,此处不再赘述。
本公开实施例提供的求职简历推送装置,基于求职者的求职简历与任务的匹配度,以及求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度计算求职者的求职简历的匹配得分,并基于求职者的求职简历的匹配得分进行简历推送,相较于在计算匹配得分时仅考虑求职者的求职简历与任务的匹配度的现有技术,整个方案不仅考虑了求职者的求职简历与任务的匹配度,还考虑了求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度,这就使得本方案计算出的求职者的求职简历的匹配得分较为准确,因而提高了简历推送的准确度。
参看图6,本公开还提供了一种任务推送装置,包括:
第二计算单元60,用于对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
第二推送单元61,用于根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
本实施例中,至少一个任务可以通过获取用户或企业在招聘网站上发布的任务得到。第二求职简历的数量和历史一段时间可以根据需要设置,此处不再赘述。任务内容之间的相同或相似可以以任务内容之间的相似度(其取值可以为百分比)来衡量,当相 似度大于某一个数值(比如80%),说明相应的两个任务的内容相似;当相似度大于某一个较大的数值(比如99%),说明相应的两个任务的内容相同。
所述根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,可以包括:
对所述至少一个任务的匹配得分按照从大到小的顺序进行排序,并将前预设数值各匹配得分中的最少一个匹配得分对应的任务作为所述目的任务。所述预设数值可以根据需要设置,此处不再赘述。
在本实施例中,应该理解到,所揭露的技术内容,可通过其它的方式实现。其中,以上所描述的各装置仅仅是示意性的,例如所述单元的划分,可以为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,单元或模块的间接耦合或通信连接,可以是电性或其它的形式。
以上各单元的功能如果以软件功能模块的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
本公开实施例提供的任务推送装置,基于求职者的求职简历与任务的匹配度,以及求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度计算任务的匹配得分,并基于任务的匹配得分进行任务推送,相较于在计算匹配得分时仅考虑求职者的求职简历与任务的匹配度的现有技术,整个方案不仅考虑了求职者的求职简历与任务的匹配度,还考虑了求职者的求职简历与历史一段时间完成与该任务内容相同或相似的任务的求职者的求职简历的匹配度,这就使得本方案计算出的任务的匹配得分较为准确,因而提高了任务推送的准确度。上文已经参考附图描述 了本公开的优选实施例,当然,本公开并不限于上面的示例。在所附的权利要求的范围内,本领域的技术人员可以进行各种改变和修改,并且应当明白,这些改变和修改自然落入本公开的技术范围内。
工业实用性
本公开实施例提供了一种求职简历推送方法与装置以及任务推送方法与装置,能够提高简历和任务推送的准确度,以有效缓解现有的任务推送和简历推送功能无法实现精准推荐的功能的技术问题。

Claims (18)

  1. 一种求职简历推送方法,其特征在于,包括:
    对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
    根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
  2. 根据权利要求1所述的方法,其特征在于,在所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度之前,还包括:
    对所述至少一个求职简历进行聚类得到至少一个求职简历类;
    其中,所述对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,包括:
    对于每一个求职简历类,计算所述任务与该求职简历类的中心之间的匹配度,将所述任务与该求职简历类的中心之间的匹配度作为所述任务与该求职简历类中每一个求职简历之间的匹配度,并计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,将所述至少一个第一求职简历与该求职简历类的中心之间的匹配度作为所述至少一个第一求职简历与该求职简历类中每一个求职简历之间的匹配度。
  3. 根据权利要求2所述的方法,其特征在于,对所述至少一个求职简历进行聚类得到至少一个求职简历类包括:
    对所述至少一个求职简历对应的向量进行聚类,其中所述至少一个求职简历对应的向量通过对求职简历的内容提取关键词,并对关键词进行向量化,对向量化的结果进行组合得到。
  4. 根据权利要求2或3所述的方法,其特征在于,所述计算所述任务与该求职简历类的中心之间的匹配度,包括:
    获取该求职简历类的中心的技能标签以及所述任务的技能标签,对该求职简历类的中心的技能标签中与所述任务的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述任务与该求职简历类的中心之间的匹配度;
    其中,所述计算所述至少一个第一求职简历与该求职简历类的中心之间的匹配度,包括:
    对于每一个第一求职简历,获取该第一求职简历的技能标签,对该求职简历类的中心的技能标签中与该第一求职简历的任一技能标签相匹配的技能标签的分值进行求和;
    对所述至少一个第一求职简历对应的求和结果求均值,将求均值的结果作为所述至少一个第一求职简历与该求职简历类的中心之间的匹配度;
    其中,所述根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,包括:
    对所述第一匹配度和第二匹配度进行加权求和,并将加权求和的结果作为该求职简历的匹配得分。
  5. 根据权利要求4所述的方法,其特征在于,所述获取该求职简历类的中心的技能标签以及所述任务的技能标签的方式包括:
    从求职者的工作经历中提取;
    所述求职者自己添加;以及
    所述求职者完成任务后由系统、招聘网站或任务发布方添加。
  6. 根据权利要求5所述的方法,其特征在于,所述从求职者的工作经历中提取的过程包括:
    从所述求职者的工作经历中提取关键词,将提取出的所以所述关键词与预设的技能标签库中的技能标签进行语义匹配,将语义匹配上的技能标签作为从所述求职者的工作经历中提取的所述技能标签;
    其中所述技能标签库中的技能标签为与能否胜任工作相关的标签,包括性别标签、年龄标签、专业标签、学历标签、工作年限标签以及与工作内容相关的标签。
  7. 根据权利要求5或6所述的方法,其特征在于,所述求职者自己添加的过程包括:
    所述求职者可以首先选择任务,然后,由系统根据所述求职者选择的所述任务向所述求职者推送相关的技能标签供求职者选择。
  8. 根据权利要求5至7中任一项所述的方法,其特征在于,所述所述求职者完成任务后由系统、招聘网站或任务发布方添加的技能标签为求职者完成的任务的技能标签,所述任务的技能标签由所述任务发布方自行添加;
    其中所述任务的各个技能标签可以根据其重要程度设置不同的分值。
  9. 根据权利要求1至8中任一项所述的方法,其特征在于,所述任务与每一个第一求职简历之间的匹配度均大于第一数值。
  10. 一种任务推送方法,其特征在于,包括:
    对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
    根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
  11. 根据权利要求10所述的方法,其特征在于,在所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度之前,还包括:
    对所述至少一个任务进行聚类得到至少一个任务类;
    其中,所述对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,包括:
    对于每一个任务类,计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,并计算所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度。
  12. 根据权利要求11所述的方法,其特征在于,对所述至少一个任务进行聚类得到至少一个任务类包括:
    对所述至少一个任务对应的向量进行聚类,其中所以所述任务对应的向量可以通过对所述任务的内容提取关键词,并对所述关键词进行向量化,对所述向量化的结果进行组合得到。
  13. 根据权利要求11或12所述的方法,其特征在于,所述计算所述求职简历与该任务类的中心之间的匹配度作为所述求职简历与该任务类中每一个任务之间的匹配度,包括:
    获取所述求职简历的技能标签以及该任务类的中心的技能标签,对该任务类的中心的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和,并将求和结果作为所述求职简历与该任务类的中心之间的匹配度;
    其中,所述计算所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度作为所述求职简历与该任务类中每一个任务对应的至少一个第二求职简历之间的匹配度,包括:
    对于每一个第二求职简历,获取该第二求职简历的技能标签,对该第二求职简历的技能标签中与所述求职简历的任一技能标签相匹配的技能标签的分值进行求和;
    对所述至少一个第二求职简历对应的求和结果求均值,将求均值的结果作为所述求职简历与该任务类的中心对应的至少一个第二求职简历之间的匹配度;
    其中,所述根据所述第三匹配度和第四匹配度确定该任务的匹配得分,包括:
    对所述第三匹配度和第四匹配度进行加权求和,并将加权求和的结果作为该任务的匹配得分。
  14. 根据权利要求13所述的方法,其特征在于,所述获取所述求职简历的技能标签以及该任务类的中心的技能标签的方式包括:
    从求职者的工作经历中提取的;
    求职者自己添加;以及
    求职者完成任务后由系统、招聘网站或任务发布方添加。
  15. 根据权利要求10至14中任一项所述的方法,其特征在于,还包括:
    获取最大匹配得分的任务对应的第四匹配度,若判断获知所述第四匹配度小于第二数值,则提示所述求职简历的发布者完善求职简历信息。
  16. 根据权利要求10至15中任一项所述的方法,其特征在于,所述任务与每一个第二求职简历之间的匹配度均大于第一数值。
  17. 一种求职简历推送装置,其特征在于,包括:
    第一计算单元,用于对于至少一个求职简历中的每一个求职简历,计算待处理的任务与该求职简历之间的匹配度作为第一匹配度,计算至少一个第一求职简历与该求职简历之间的匹配度作为第二匹配度,并根据所述第一匹配度和第二匹配度确定该求职简历的匹配得分,其中,每一个第一求职简历为在历史一段时间内完成与所述任务内容相同或相似的任务的求职者的求职简历;以及
    第一推送单元,用于根据所述至少一个求职简历的匹配得分从所述至少一个求职简历中选取目的求职简历,并将所述目的求职简历推送给所述任务的发布者。
  18. 一种任务推送装置,其特征在于,包括:
    第二计算单元,用于对于至少一个任务中的每一个任务,计算待处理的求职简历与该任务之间的匹配度作为第三匹配度,计算所述求职简历与该任务对应的至少一个第二求职简历之间的匹配度作为第四匹配度,并根据所述第三匹配度和第四匹配度确定该任务的匹配得分,其中,每一个第二求职简历为在历史一段时间内完成与该任务内容相同或相似的任务的求职者的求职简历;以及
    第二推送单元,用于根据所述至少一个任务的匹配得分从所述至少一个任务中选取目的任务,并将所述目的任务推送给所述求职简历的发布者。
PCT/CN2020/079346 2019-03-15 2020-03-13 求职简历推送方法与装置以及任务推送方法与装置 WO2020187168A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910200072.3 2019-03-15
CN201910200072.3A CN109831531B (zh) 2019-03-15 2019-03-15 求职简历推送方法与装置以及任务推送方法与装置

Publications (1)

Publication Number Publication Date
WO2020187168A1 true WO2020187168A1 (zh) 2020-09-24

Family

ID=66870255

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/079346 WO2020187168A1 (zh) 2019-03-15 2020-03-13 求职简历推送方法与装置以及任务推送方法与装置

Country Status (2)

Country Link
CN (1) CN109831531B (zh)
WO (1) WO2020187168A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112948705A (zh) * 2020-12-30 2021-06-11 广州博士信息技术研究院有限公司 一种基于政策大数据的智能匹配方法、装置及介质
CN113657496A (zh) * 2021-08-17 2021-11-16 深圳平安智汇企业信息管理有限公司 基于相似度匹配模型的信息匹配方法、装置、设备及介质
CN117910994A (zh) * 2024-03-19 2024-04-19 浙江之科智慧科技有限公司 一种基于深度学习的课程推荐方法、系统与存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109831531B (zh) * 2019-03-15 2020-05-05 河北冀联人力资源服务集团有限公司 求职简历推送方法与装置以及任务推送方法与装置
CN110900623A (zh) * 2019-11-25 2020-03-24 广西秀牛科技有限公司 一种自助面试机器人
CN111144723A (zh) * 2019-12-17 2020-05-12 埃摩森网络科技(上海)有限公司 人岗匹配推荐方法及系统、存储介质
CN111861361B (zh) * 2020-04-09 2021-07-27 河北利至人力资源服务有限公司 一种智能简历推送系统及方法
CN112182383B (zh) * 2020-09-28 2023-11-14 深圳平安智汇企业信息管理有限公司 第二岗位的推荐方法、装置以及计算机设备
CN113392291B (zh) * 2021-07-06 2024-01-23 山东电力工程咨询院有限公司 一种基于数据中心的服务推荐方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102117323A (zh) * 2011-02-21 2011-07-06 深圳埃斯欧纳信息咨询有限公司 一种推荐求职简历的处理方法和系统
US8595149B1 (en) * 2003-04-18 2013-11-26 Jobdiva, Inc. Resume management and recruitment workflow system and method
CN105117863A (zh) * 2015-09-28 2015-12-02 北京橙鑫数据科技有限公司 简历职位匹配方法及装置
CN107590133A (zh) * 2017-10-24 2018-01-16 武汉理工大学 基于语义的招聘职位与求职简历匹配的方法及系统
CN109831531A (zh) * 2019-03-15 2019-05-31 河北冀联人力资源服务集团有限公司 求职简历推送方法与装置以及任务推送方法与装置

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104517166A (zh) * 2013-09-29 2015-04-15 顾山林 应聘者技能评估方法、装置及系统
CN104573998B (zh) * 2015-01-30 2018-01-09 薛晔伟 求职数据的处理方法和装置
CN107291715A (zh) * 2016-03-30 2017-10-24 阿里巴巴集团控股有限公司 简历评估方法和装置
CN106022728A (zh) * 2016-05-26 2016-10-12 深圳峰创智诚科技有限公司 用户测评方法和装置
CN106408249A (zh) * 2016-08-31 2017-02-15 五八同城信息技术有限公司 简历与职位匹配方法及装置
WO2019046528A1 (en) * 2017-08-30 2019-03-07 4D Technologies, LLC METHOD AND SYSTEM FOR LEARNING CHARACTERISTICS AND RESPONSE TO INTERACTIVE QUERY WITH AUTOMATIC UPDATE
CN108038622B (zh) * 2017-12-26 2022-01-28 北京理工大学 一种群智感知系统推荐用户方法
CN108763427A (zh) * 2018-05-24 2018-11-06 佛山市轻遣网络有限公司 招聘系统推荐方法及系统
CN109214772B (zh) * 2018-08-07 2024-01-16 平安科技(深圳)有限公司 项目推荐方法、装置、计算机设备及存储介质
CN109242411A (zh) * 2018-08-14 2019-01-18 安徽网才信息技术股份有限公司 一种网络招聘系统中面向企业的人才推荐方法
CN109241446B (zh) * 2018-10-17 2021-11-26 重庆汇博信息科技有限公司 一种职位推荐方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8595149B1 (en) * 2003-04-18 2013-11-26 Jobdiva, Inc. Resume management and recruitment workflow system and method
CN102117323A (zh) * 2011-02-21 2011-07-06 深圳埃斯欧纳信息咨询有限公司 一种推荐求职简历的处理方法和系统
CN105117863A (zh) * 2015-09-28 2015-12-02 北京橙鑫数据科技有限公司 简历职位匹配方法及装置
CN107590133A (zh) * 2017-10-24 2018-01-16 武汉理工大学 基于语义的招聘职位与求职简历匹配的方法及系统
CN109831531A (zh) * 2019-03-15 2019-05-31 河北冀联人力资源服务集团有限公司 求职简历推送方法与装置以及任务推送方法与装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHEN XIAO ,WANG JIAN-MIN: "Algorithm Research for Personalization Resume Recommendation of E-Recruitment", COMPUTER SCIENCE, vol. 35, no. 10A, 24 October 2008 (2008-10-24), pages 79 - 82, XP055734259 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112948705A (zh) * 2020-12-30 2021-06-11 广州博士信息技术研究院有限公司 一种基于政策大数据的智能匹配方法、装置及介质
CN113657496A (zh) * 2021-08-17 2021-11-16 深圳平安智汇企业信息管理有限公司 基于相似度匹配模型的信息匹配方法、装置、设备及介质
CN113657496B (zh) * 2021-08-17 2023-08-18 深圳平安智汇企业信息管理有限公司 基于相似度匹配模型的信息匹配方法、装置、设备及介质
CN117910994A (zh) * 2024-03-19 2024-04-19 浙江之科智慧科技有限公司 一种基于深度学习的课程推荐方法、系统与存储介质

Also Published As

Publication number Publication date
CN109831531B (zh) 2020-05-05
CN109831531A (zh) 2019-05-31

Similar Documents

Publication Publication Date Title
WO2020187168A1 (zh) 求职简历推送方法与装置以及任务推送方法与装置
CN110147551B (zh) 多类别实体识别模型训练、实体识别方法、服务器及终端
CN110866181B (zh) 资源推荐的方法、装置及存储介质
WO2019169756A1 (zh) 产品推荐方法、装置及存储介质
US11080608B2 (en) Agent aptitude prediction
CN109033101B (zh) 标签推荐方法及装置
CN103164463A (zh) 推荐标签的方法和装置
CN111966886A (zh) 对象推荐方法、对象推荐装置、电子设备及存储介质
TW201814556A (zh) 資訊匹配方法及相關裝置
WO2017159402A1 (ja) 共クラスタリングシステム、方法およびプログラム
JP7350206B1 (ja) 求人データ管理システム及び求人データ管理方法
JP6062384B2 (ja) タスク割り当てサーバ、タスク割り当て方法およびプログラム
CN113111924A (zh) 电力客户分类方法及装置
JP6308339B1 (ja) クラスタリングシステム、方法およびプログラム、並びに、レコメンドシステム
CN110019563B (zh) 一种基于多维数据的肖像建模方法和装置
Al Hammadi et al. Data mining in education-an experimental study
CN110929169A (zh) 基于改进Canopy聚类协同过滤算法的职位推荐方法
Do et al. A fuzzy approach to detect spammer groups
CN109213873B (zh) 一种为待售专利自动匹配潜在买方的专利匹配方法和匹配系统
Kaya et al. Maintenance of Data Richness in Business Communication Data.
Wang et al. Recommendation based on mining product reviewers preference similarity network
US9336249B2 (en) Decision tree with just-in-time nodal computations
CN115169637A (zh) 社交关系预测方法、装置、设备和介质
CN114969371A (zh) 一种联合知识图谱的热度排序方法及装置
CN112925913A (zh) 用于匹配数据的方法、装置、设备和计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20773671

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20773671

Country of ref document: EP

Kind code of ref document: A1