Recruitment system capable of performing analysis through activeness

Abstract

The invention provides a recruitment system capable of performing analysis through activeness, and relates to the field of recruitment servers, the system comprises a server, a recruitment terminal and a personal terminal; the recruitment terminal is endowed with recruitment authority, and the employer can publish recruitment information through the recruitment terminal; the job hunting terminal is endowed with a job hunting permission, and job hunters can publish job hunting information through the job hunting terminal; and the server is endowed with contact with the recruitment end and the job hunting end, the server screens the job hunting information published by the job hunting end through the screening module and then transmits the job hunting information to the recruitment end, and the server screens the recruitment information published by the recruitment end through the screening module and then transmits the recruitment information to the job hunting end. According to the invention, the recruitment information of the recruiters is synchronized to the job hunting end with the same post, so that the information obtained by the recruitment end is ensured to be the job hunters seeking the post, and the information obtained by the job hunting end is ensured to be the recruiters who want to seek the post, thereby ensuring that the recruiters and the job hunters in the recruitment system are quickly matched, and improving the recruitment efficiency.

Description

Technical Field

[0001] The present invention relates to the field of recruitment servers, and in particular to a recruitment system capable of performing analysis based on activity. Background Art

[0002] With the promotion of network information technology, employers and job seekers usually choose to conduct preliminary communication and understanding online, and then conduct offline interviews when both parties are satisfied. The recruitment website server summarizes the resume data of job seekers and the job data of recruiters, and then pushes them to recruiters and job seekers respectively. Both parties can communicate through their respective ports.

[0003] Nowadays, when employers need to recruit, they will ask their own human resources specialists to screen resumes. After the resumes are screened by the human resources specialists, they will communicate with the remaining personnel. Since there is a large amount of resume information on recruitment websites, and since resumes are made by job seekers themselves and have beautified elements, the information on the resumes may not be the job seekers' true thoughts. Secondly, resumes made by job seekers themselves are prone to diversification of job titles, which greatly increases the time for recruiters to screen resumes and indirectly increases communication costs. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a recruitment system that can be analyzed by activity.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions: a recruitment system capable of performing analysis by activity, the system comprising a server, a recruitment terminal and a personal terminal;

[0006] The recruitment end is granted recruitment authority, where employers can publish recruitment information through the recruitment end;

[0007] The job-seeking terminal is granted with job-seeking authority, where job seekers can publish job-seeking information through the job-seeking terminal;

[0008] The server is configured to connect the recruiting end and the job-seeking end, wherein the server screens the job-seeking information released by the job-seeking end through a screening module and then transmits it to the recruiting end, and the server screens the job-seeking information released by the recruiting end through a screening module and then transmits it to the job-seeking end.

[0009] The effects achieved by the above components are as follows: the server records the names of various positions, recruiters select position information according to the content of the recruitment they need to recruit and fill in the recruitment information on the recruitment side, the filled-in recruitment information is uploaded to the server, and the server backend personnel manually review the information posted by the recruiter (to determine whether there is some prohibited information), job seekers select the job title they need to apply for on the job search side according to the job they want to seek and fill in their resumes and then upload them to the server for manual review (to determine whether there is some prohibited information), after the review is completed, the job seeker's resume information is synchronized to the recruitment side with the same position, and the recruiter's recruitment information is synchronized to the job search side with the same position, so as to ensure that the information obtained by the recruitment side is all job seekers seeking the position, and the information obtained by the job search side is all recruiters who want to seek the position, thereby ensuring rapid matching between recruiters and job seekers in the recruitment system to improve recruitment efficiency.

[0010] Preferably, the server includes a job search module, a recruitment module and a networking module;

[0011] A job search module, wherein the job search module is used to store and process job search information;

[0012] Recruitment module, where the recruitment module is used to store and process recruitment information;

[0013] The networking module is used to connect the Internet with the job search module and the recruitment module. The job titles in the recruitment module are synchronized to the job search module through the networking module. Job seekers select the synchronized job titles in the job search module for entry.

[0014] The effects achieved by the above components are: by pre-entering the job titles on the server, first of all, the uniformity of the job titles can be ensured to avoid the occurrence of multiple names for the same job, which causes information mismatch between recruiters and job seekers; secondly, pre-entering the job titles can adopt a charging system for popular jobs to avoid circumventing the charging model by customizing jobs; finally, pre-entering the job titles can simplify the recruitment jobs and remove some misleading qualified names of jobs, which may lead to mutual misleading between job seekers and recruiters and cause invalid communication; pre-entering the job titles can avoid the occurrence of custom jobs, where a recruiter posts multiple identical jobs to increase server pressure, resulting in more identical jobs appearing on the job search side and affecting job seekers' choices.

[0015] Preferably, the job seeker information entered by each of the job seeker terminals is transmitted to the job search module through the networking module, wherein the job search module includes a job search storage unit and a job search processing unit, and the job seeker information is packaged as an information subset and stored in the storage unit, wherein the information subset is marked with the title of the job sought by the job seeker as the subset name, and the information subset name mark is compared with each job title in the networking module through a comparison program, and the information subset name marks with the same name are classified into the same information set and stored.

[0016] The effect achieved by the above components is: the resume of each job seeker is stored in a subset and then the job title is stored as the subset name. Multiple subsets with the same name are classified and stored in an information set. The job title is used as the information set marker, that is, the same information set stores the same job seeker information for the same position. The information set can be directly obtained based on the job title, thereby improving the recruiter's ability to quickly screen the corresponding job seekers to shorten the recruitment process.

[0017] Preferably, the server further includes an activity determination module;

[0018] Activity judgment module, through Calculate, where act is the activity, P is the online time, P i The access records of users during the online period are extracted for the time period without operation. Each access record is screened to remove the time when no operation is performed for a long time to obtain the real online time. The real online time of one day is accumulated to obtain the effective online time of one day. The activity data calculated on the day comes from the data of the previous day. If there is no online data on the previous day, it is automatically marked as a new user, and the number of new users is accumulated and counted.

[0019] The effect achieved by the above components is: since the software of smart terminals (such as mobile phones or computers) now has the function of running in the background, the server will automatically collect its time as online time when it is running in the background. If the job seeker does not perform any operation for a long time, it is deemed that there is no job search activity. In this state, the job seeker is deemed to be in an idle state. The time obtained by removing the time of this state is the real job search time of the job seeker. The activity percentage is obtained by dividing the real job search time by the total time of the day. The higher the activity percentage, the more eager the job seeker is for the job demand. The probability of the job seeker joining the job is proportional to the activity percentage. The more eager the job seeker is for the job demand, the greater the probability of joining the job. Recruiters can screen job seekers by the activity percentage when urgently hiring. Since the above data are all derived from the previous day, the data of the previous day as the most recent data has a relatively high calculation accuracy. The degree of job search desire can be deduced by the activity data of the previous day. If there is no online data on the previous day, it is counted as a new account, and the number of new accounts is used as an identifier. When the number of new accounts is large, it can be deduced that the job seeker is not eager to find a job.

[0020] Preferably, various behavioral data of job seekers during their active period the previous day are extracted, wherein the behavioral data include the number of job post visits and active time periods, and a rectangular coordinate system with two Y axes is established with one day as the X-axis, the job post visit status and the active status as the Y-axis, wherein online activity is recorded as 1, and the visited job is recorded as 1 in the coordinate system, online activity exists in the form of a line segment in the coordinate system, and the visit status exists in the form of a point in the coordinate system, and a time axis is set with one day as the interval, and the time periods from 9 a.m. to 12 noon and from 2 p.m. to 5 p.m. are marked on the time axis, and the time points where online activity is recorded as 1 are marked on the time axis, and the number of time periods falling into the marked area is counted and then divided by the total number, and the time points where the job post is recorded as 1 are marked on the time axis, and the number of time periods falling into the marked area is counted and then divided by the total number, and the job seeker's job search expectations are judged by the size of the quotient.

[0021] The effect achieved by the above components is: a rectangular coordinate system with a double Y-axis is established with one day as the X-axis and online activity and visited positions as the Y-axis, where the time with online activity recorded as 1 is the line segment, and the activity level is obtained by dividing the length of the online active line segment by the length of the day. The activity level is used to further judge the job-seeking desire. The higher the activity level, the higher the job-seeking desire. The length of the active line segment for marking and the total length of the active line segment are extracted to find the quotient. When the quotient value is less than 0.7, it means that the job seeker is in daily working hours for a small part of the time, and a temporary label (i.e., employed status) can be set for the job seeker. Label), the smaller the quotient, the higher the accuracy of judging the job seeker's employment status. Conversely, if the quotient is greater than or equal to 0.7, it means that the job seeker is in daily working hours most of the time, and a temporary label (i.e., a resigned status label) can be set for the job seeker. The larger the quotient, the higher the accuracy of judging the job seeker's employment status. By judging the job seeker's true employment status, the recruitment efficiency can be improved. If the job seeker is in the resigned status, the probability of successful recruitment is higher and the time to join the job is shorter. When the job seeker is in the employed status, the time to successfully join the job is longer, which can provide more accurate push for urgent recruitment.

[0022] Preferably, various behavioral data of job seekers during their active period of the previous day are extracted, wherein the behavioral data also includes the content of reply messages, and multiple category libraries are established in the server. Finally, keywords are manually entered into the category libraries, and by comparing the extracted reply message content with the keywords in each category library, if there is any overlap with the keywords in the category library, the name of the category library is used as a label to mark the job seeker information, thereby marking the job seeker's real needs on the job seeker information in real time.

[0023] The effect achieved by the above components is: keywords can be entered in the server category library according to needs (keywords can be salary, working hours, rest time, holidays, benefits and other related information), job seekers and the same recruiter communication information are grouped together, and the keyword repetition value of each group of communication information is compared. When the repetition value occupies a communication information group quantity quotient value greater than or equal to 0.5, the repetition value keyword is marked on the job seeker information, and then the job seeker information marked with keyword information is synchronized to the recruitment end, so that the recruiter using the recruitment end can intuitively and quickly screen out the desired job seekers, thereby improving recruitment efficiency.

[0024] Preferably, the server extracts the message content sent by the job seeker the day before by extracting various behavioral data during the job seeker's active period the day before, and compares the extracted chat content with the quick message edited by the job seeker's job search portal, wherein: the positions with a quick message ratio higher than 50% are low-expectation positions, and the rest are high-expectation positions, and the job seeker's resume information is sent to the remaining recruiting companies with the same open high-expectation positions.

[0025] The effect achieved by the above components is: job seekers can manually enter quick messages on the job search side, and compare the extracted reply messages with the quick messages. The job seekers and the same recruiter's communication information are grouped together, and the quick messages of each group of communication information are compared for repetition values. If the repetition value of the quick message is higher than 50%, it can be identified as a low-expectation position. Otherwise, if it is less than or equal to 50%, it is a high-expectation position, indicating that the job seeker is very interested in this position. The server synchronizes the job seeker information to the recruitment side with the same position, thereby improving recruitment efficiency.

[0026] Preferably, in the server, k=α1×label1+α2×label2+......+α n ×label n Assign weights to each label of each job seeker, where k is the ranking coefficient, α1…α n The assignment coefficient, label1, and label2 are manually entered by the recruitment end. n A constant 1 is calculated for the value of each tag. If the job seeker does not have the tag, a constant 0 is calculated. The job seekers are sorted by the size of the sorting coefficient k, and the sorting table required by the recruitment end for the job seekers is sent from the server to the recruitment end.

[0027] The effect achieved by the above components is: the recruiter inputs α1...α n , where α1…α n For the cumulative sum to be 1, α1…α nThe size is the importance that the recruiter attaches to each label (for example, label 1 is salary, label 2 is benefits, and label 3 is working hours. The recruiter needs to recruit people with the highest requirements for benefits, the second highest requirements for salary, and the lowest requirements for working hours. Then, according to their own requirements, they can input α1=0.5, α2=0.3, and α3=0.2. Then the recruiter will pass α1=0.5, α2=0.3, and α3=0.2 to the server. The server will calculate k based on the data and the representative value of the job seeker's own label. The k value is sorted. The higher the ranking, the more it meets the recruiter's needs, thereby improving the recruitment rate.

[0028] The label can also define an age label (for example, job seekers aged between 20 and 30 are defined as label4, job seekers aged between 30 and 40 are defined as label5, and job seekers aged between 40 and 50 are defined as label6. When the recruiter tends to recruit mainly those aged 20-30, the age requirement can be relaxed to those aged 30-40 with excellent conditions, and the other conditions are still those with the highest welfare requirements, the second highest salary requirements, and the lowest working hours requirements, you can choose to input α1=0.4, α2=0.2, α3=0.1, α4=0.2, α5=0.1, and α6=0 according to the degree of your own requirements to calculate the k value for sorting).

[0029] Preferably, various behavioral data of the recruiter during the active period of the previous day are extracted, where the data is divided into reply time period and the proportion of unread messages, and sorted from low to high by the proportion of unread messages, where messages that have been read but not replied are recorded as unread messages. This sorting is the main recruitment sorting, and the reply time period is marked as a filtering label on the recruiter's job information. The server pushes the main recruitment sorting of each job information to the job seeker with corresponding job requirements.

[0030] The effect achieved by the above components is: if the proportion of unread messages is greater than 30%, it can be inferred that too many people have submitted resumes to the recruiter or the recruitment is not urgent. The main recruitment ranking push list is ranked from low to high according to the proportion of unread messages. Job seekers can choose companies that are ranked higher on the main recruitment ranking list for communication, thereby improving the recruitment communication effect.

[0031] Preferably, the server refreshes and pushes the main recruitment information according to real-time time, and the job seeker can choose to filter and prioritize quick reply messages. At this time, recruiters who do not reply to messages at the current time are hidden in the main recruitment information sorting, and the new main recruitment information sorting is sent to the job seeker.

[0032] The effect achieved by the above components is: since most recruiters are HR specialists, by recording the time period of the recruiter's reply on the previous day, the HR specialist's working hours can be deduced on the next day. The current time can be compared with the recruitment specialist's activity time on the previous day to determine whether the recruiter is in a working state. By eliminating positions that are not in working hours, the communication efficiency between job seekers and recruiters can be improved.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are:

[0034] 1. In the present invention, the recruitment information of the recruiter is synchronized to the job seeker terminal with the same position, so as to ensure that the information obtained by the recruiter terminal is all the job seekers seeking the position, and the information obtained by the job seeker terminal is all the recruiters who want to seek the position, thereby ensuring the rapid matching between the recruiter and the job seeker in the recruitment system and improving the recruitment efficiency;

[0035] 2. In the present invention, the job titles are pre-entered on the server to ensure the uniformity of the job titles, avoid the occurrence of multiple titles for the same job title, and cause information mismatch between the recruiter and the job seeker, thereby improving communication efficiency;

[0036] 3. In the present invention, pre-entering job titles can simplify recruitment positions and remove some misleading qualified titles of positions, which may lead to mutual misleading between job seekers and recruiters and cause ineffective communication.

[0037] 4. In the present invention, pre-entering the job title can avoid the situation where a recruiter posts multiple identical jobs due to customized jobs to increase server pressure, resulting in a large number of identical jobs appearing on the job-seeking end, affecting job seekers' choices.

[0038] 5. In the present invention, information sets can be directly obtained according to job titles, thereby improving recruiters to quickly screen corresponding job seekers to shorten recruitment effects.

[0039] 6. In the present invention, the job-seeking desire can be deduced through the activity data of the previous day. If there is no online data on the previous day, it is counted as a new user, and the number of new users is used as an indicator. When the number of new users is large, it can be deduced that the job seeker's job-seeking desire is not high.

[0040] 7. In the present invention, the recruitment efficiency can be improved by judging the actual employment status of the job seeker. If the job seeker is in the resigned state, the probability of successful recruitment is higher and the time to join the job is shorter. When the job seeker is in the employed state, the time to successfully join the job is longer, thereby improving and accurately pushing urgent recruitments. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a recruitment system that can be analyzed by activity is proposed for the present invention;

[0042] Figure 2 The present invention proposes a flow chart of a recruitment system that can be analyzed by activity. DETAILED DESCRIPTION

[0043] Embodiment 1, as Figure 1-2 As shown, a recruitment system that can be analyzed by activity level includes a server, a recruitment terminal and a personal terminal;

[0044] The recruitment end is granted recruitment authority, where employers can publish recruitment information through the recruitment end;

[0045] The job-seeking terminal is granted with job-seeking authority, where job seekers can publish job-seeking information through the job-seeking terminal;

[0046] The server is configured to connect the recruiting end and the job-seeking end, wherein the server screens the job-seeking information published by the job-seeking end through the screening module and transmits it to the recruiting end, and the server screens the job-seeking information published by the recruiting end through the screening module and transmits it to the job-seeking end;

[0047] The server records the names of various positions. Recruiters select the position information according to the content they want to recruit and fill in the recruitment information on the recruitment end. The filled recruitment information is uploaded to the server. The server backend personnel manually review the information posted by the recruiter (to determine whether there is some prohibited information). Job seekers select the job title they want to apply for on the job search end according to the job they want to seek and fill in their resumes and then upload them to the server for manual review (to determine whether there is some prohibited information). After the review is completed, the job seeker's resume information is synchronized to the recruitment end with the same position, and the recruiter's recruitment information is synchronized to the job search end with the same position, so as to ensure that the information obtained by the recruitment end is all job seekers seeking the position, and the information obtained by the job search end is all recruiters who want to seek the position, thereby ensuring the rapid matching between recruiters and job seekers in the recruitment system and improving recruitment efficiency;

[0048] The server includes a job search module, a recruitment module and a networking module;

[0049] A job search module, wherein the job search module is used to store and process job search information;

[0050] Recruitment module, where the recruitment module is used to store and process recruitment information;

[0051] A networking module is used to connect the Internet with the job search module and the recruitment module. The job titles in the recruitment module are synchronized to the job search module through the networking module. Job seekers select the synchronized job titles in the job search module for entry. By pre-entering job titles on the server, firstly, the uniformity of job titles can be ensured to avoid the occurrence of multiple names for the same job, which causes information mismatch between recruiters and job seekers. Secondly, pre-entering job titles can adopt a charging system for popular jobs to avoid circumventing the charging model through customized jobs. Finally, pre-entering job titles can simplify recruitment jobs and remove misleading qualified names of some jobs, which may lead to mutual misunderstanding between job seekers and recruiters and cause invalid communication. Pre-entering job titles can avoid the occurrence of customized jobs, where a recruiter posts multiple identical jobs to increase server pressure, resulting in a large number of identical jobs on the job search side, affecting job seekers' choices.

[0052] Embodiment 2, on the basis of embodiment 1, the job seeker information entered by each job seeker terminal is transmitted to the job seeking module through the networking module, wherein the job seeking module includes a job seeking storage unit and a job seeking processing unit, the job seeker information is packaged as an information subset and stored in the storage unit, wherein the information subset and the job title sought by the job seeker are marked as the subset name, and the information subset name mark and each job title in the networking module are compared through a comparison program, and the information subset name marks with the same name are classified into the same information set and stored, the resume of each job seeker is stored in the subset and then the job title is stored as the subset name, and multiple subsets with the same name are classified and stored in an information set, and the job title is used as the information set mark, that is, the same information set stores the same job seeker information, and the information set can be directly obtained according to the job title, thereby improving the recruiter to quickly screen the corresponding job seekers to shorten the recruitment effect.

[0053] Embodiment 3, based on embodiment 1 or 2, the server further includes an activity determination module;

[0054] Activity judgment module, through Calculate, where act is the activity, P is the online time, P iThe time period is a time period without operation. The access records of users during their online period are extracted. Each access record is screened to remove the time when no operation is performed for a long time to obtain the real online time. The effective online time of a day is obtained by accumulating the real online time. The activity data calculated on the day comes from the data of the previous day. If there is no online data on the previous day, it is automatically marked as a new user, and the number of new users is accumulated and counted. Since the software of smart terminals (such as mobile phones or computers) now has the function of running in the background, the server will automatically collect its time as online time when it is running in the background. If a job seeker has not performed any operation for a long time, it is regarded as not having any job search activities. This state is regarded as the job seeker being in an idle state. By removing this state, The time obtained by the time of the status is the real job-hunting time of the job seeker. The activity percentage is obtained by dividing the real job-hunting time by the total time of the day. The higher the activity percentage, the more eager the job seeker is for the job demand. The probability of the job seeker joining the job is proportional to his / her activity percentage. The more eager the job seeker is for the job demand, the greater the probability of joining the job. Recruiters can use the activity percentage to screen job seekers when they are in a hurry. Since the above data are all from the previous day, the data of the previous day as the most recent data has a relatively high calculation accuracy. The degree of job-hunting desire can be deduced from the activity data of the previous day. If there is no online data on the previous day, it is counted as a new user, and the number of new users is used as an indicator. When the number of new users is large, it can be deduced that the job seeker is not eager to find a job.

[0055] Embodiment 4, based on any one of embodiments 1-3, extract various behavioral data of the job seeker during the active period of the previous day, wherein the behavioral data includes the number of times the job was visited and the active time period, and a rectangular coordinate system with two Y axes is established with one day as the X-axis, the job visit status and the active status as the Y-axis, wherein online activity is recorded as 1, and the job visit is recorded as 1 in the coordinate system, and the online activity exists in the form of a line segment in the coordinate system, and the visit status exists in the form of a point in the coordinate system, and a time axis is set with one day as the interval, and the time from 9 am to 12 noon and from 2 pm to 5 pm are marked on the time axis, and the time point where the online activity is recorded as 1 is marked on the time axis, and the number of time periods falling into the marked area is counted, and then the quotient is calculated with the total number, and the time point where the job is recorded as 1 is marked on the time axis, and the number of time periods falling into the marked area is counted, and then the quotient is calculated with the total number, and the job seeker's job search expectations are judged by the size of the quotient, and a time axis is established with one day as the X-axis and the online activity and the job visit as the Y-axis. A rectangular coordinate system with two Y axes is established, in which the time when online activity is recorded as 1 is a line segment. The activity level is obtained by dividing the length of the online active line segment by the length of the day. The activity level is used to further judge the job-seeking desire. The higher the activity level, the higher the job-seeking desire. The length of the active line segment for marking and the total length of the active line segment are extracted to find the quotient. When the quotient value is less than 0.7, it means that the job seeker is in daily working hours for a small part of the time. A temporary label (i.e., an employed status label) can be set for the job seeker. The smaller the quotient value, the higher the accuracy of judging the job seeker's employed status. On the contrary, if the quotient value is greater than or equal to 0.7, it means that the job seeker is in daily working hours for most of the time. A temporary label (i.e., a resigned status label) can be set for the job seeker. The larger the quotient value, the higher the accuracy of judging the job seeker's employed status. By judging the job seeker's true employed status, the recruitment efficiency can be improved. If the job seeker is in the resigned state, the probability of successful recruitment is higher and the time to join the job is shorter. When the job seeker is in the employed state, the time to successfully join the job is longer, which can improve and accurately push urgent recruitment.

[0056] The label can also define an age label (for example, job seekers aged between 20 and 30 are defined as label4, job seekers aged between 30 and 40 are defined as label5, and job seekers aged between 40 and 50 are defined as label6. When the recruiter tends to recruit mainly those aged 20-30, the age requirement can be relaxed to those aged 30-40 with excellent conditions, and the other conditions are still those with the highest welfare requirements, the second highest salary requirements, and the lowest working hours requirements, you can choose to input α1=0.4, α2=0.2, α3=0.1, α4=0.2, α5=0.1, and α6=0 according to the degree of your own requirements to calculate the k value for sorting).

[0057] Embodiment 5, based on any one of embodiments 1-4, extracts various behavioral data of job seekers during their active period of the previous day, wherein the behavioral data also includes the content of reply information, establishes multiple category libraries in the server, and finally manually enters keywords into the category libraries, and compares the extracted reply information content with the keywords in each category library. If there is overlap with the keywords in the category library, the name of the category library is used as a label to mark the job seeker information, and the job seeker's real needs are marked on the job seeker information in real time. Keywords can be entered into the server category library according to needs (the keywords can be salary, working hours, rest time, holidays, benefits and other related information), and the job seekers and the same recruiter communicate information as a group, and compare the repetition value of the keyword in each group of communication information. When the repetition value occupies a communication information group quantity quotient value greater than or equal to 0.5, the repetition value keyword is marked on the job seeker information, and then the job seeker information marked with keyword information is synchronized to the recruitment end, so that the recruiter using the recruitment end can intuitively and quickly screen out the desired job seekers, thereby improving recruitment efficiency.

[0058] Embodiment 6, on the basis of any one of embodiments 1-5, the server extracts the behavioral data of the job seeker during the active period of the previous day, extracts the message content sent by the job seeker the previous day, and compares the extracted chat content with the quick message edited by the job seeker at the job search port, wherein: the positions with a quick message ratio higher than 50% are low-expectation positions, and the rest are high-expectation positions, and the job seeker's resume information is sent to the remaining recruiting companies with the same open high-expectation positions. The job seeker can manually enter the quick message at the job search end, and compare the extracted reply message with the quick message. The job seeker and the same recruiter's communication information are grouped together, and the quick message repetition value of each group of communication information is compared. If the repetition value with the quick message is higher than 50%, it can be identified as a low-expectation position, otherwise it is less than or equal to 50% for a high-expectation position, indicating that the job seeker is very interested in this position. The server synchronizes the job seeker information to the recruitment end with the same position, thereby improving recruitment efficiency.

[0059] Embodiment 7, based on any one of embodiments 1-6, in the server, k=α1×label1+α2×label2+......+α n ×label n Assign weights to each label of each job seeker, where k is the ranking coefficient, α1…α n The assignment coefficient, label1, and label2 are manually entered by the recruitment end. nThe value of each tag is a constant of 1. If the job seeker does not have the tag, the constant is 0. The job seekers are sorted by the sorting coefficient k. The sorting table required by the recruitment end is sent from the server to the recruitment end. The recruiter enters α1...α n , where α1…α n For the cumulative sum to be 1, α1…α n The size is the degree of importance that the recruiter attaches to each label (for example, label1 is salary, label2 is benefits, and label3 is working hours. The recruiter needs to recruit people with the highest requirements for benefits, the second highest requirements for salary, and the lowest requirements for working hours. Then, α1=0.5, α2=0.3, and α3=0.2 can be input according to their own requirements. Then, the recruiting end transmits α1=0.5, α2=0.3, and α3=0.2 to the server. The server calculates k based on the data and the representative value of the job seeker's own label. The k value is sorted. The higher the ranking, the more it meets the recruiter's needs, thereby improving the recruitment rate.

[0060] Example 8, based on any one of Examples 1-7, extract various behavioral data of the recruiter during the active period of the previous day, where the data is divided into reply time period and unread message ratio, and sorted from low to high by the unread message ratio, where the messages that have been read but not replied are recorded as unread messages. This sorting is the main recruitment sorting, and the reply time period is marked as a filtering tag on the recruiter's job information. The server pushes the main recruitment sorting of each job information to the job seeker with corresponding job requirements. If the unread message ratio is greater than 30%, it can be deduced that too many people have submitted resumes to the recruiter or the recruitment urgency is not high. The main recruitment sorting push list based on the unread message ratio from low to high can then be pushed to the job seeker. Select the companies that are ranked higher on the main recruitment ranking list for communication, thereby improving the recruitment communication effect. The server refreshes and pushes the main recruitment information according to the real-time time. The job seeker can choose to filter and prioritize quick reply messages. At this time, the recruiters who do not reply to the message at the current time are hidden in the main recruitment information ranking, and the new main recruitment information ranking is sent to the job seeker. Since most recruiters are HR specialists, by recording the time period of the recruiter's reply on the previous day, the working hours of the HR specialist can be deduced on the next day. Whether the current time is compared with the activity time of the recruiter on the previous day can be used to judge whether it is in working state. By eliminating the positions that are not in working time, the communication efficiency between job seekers and recruiters can be improved.

[0061] Working principle: the server records the names of various positions. Recruiters select the position information according to the content they want to recruit and fill in the recruitment information on the recruitment end. The filled recruitment information is uploaded to the server. The server backend staff manually reviews the information posted by the recruiter (to determine whether there is some prohibited information). Job seekers select the job title they want to apply for on the job search end according to the job they want to seek and fill in their resumes and then upload them to the server for manual review (to determine whether there is some prohibited information). After the review is completed, the job seeker's resume information is synchronized to the recruitment end with the same position, and the recruiter's recruitment information is synchronized to the job search end with the same position to ensure that the information obtained by the recruitment end is all job seekers seeking the position, and the information obtained by the job search end is The information obtained is all about the recruiters who want to seek positions, thereby ensuring a quick match between recruiters and job seekers in the recruitment system and improving recruitment efficiency. By pre-entering job titles on the server, first of all, the uniformity of job titles can be ensured to avoid the occurrence of multiple names for the same job, which causes information mismatch between recruiters and job seekers. Secondly, pre-entering job titles can adopt a charging system for popular positions to avoid circumventing the charging model through custom positions. Finally, pre-entering job titles can simplify recruitment positions and remove some misleading qualified names of positions, which may lead to mutual misunderstanding between job seekers and recruiters and cause invalid communication. Pre-entering job titles can avoid the occurrence of one recruiter posting multiple positions due to custom positions. Similar positions increase the pressure on the server and cause more similar positions to appear on the job search side, affecting job seekers' choices. The resume of each job seeker is stored in a subset and then the job title is stored as the subset name. Multiple subsets with the same name are classified and stored in an information set. The job title is used as the information set marker, that is, the same information set stores the same job seeker information. The information set can be directly obtained according to the job title, thereby improving recruiters to quickly screen corresponding job seekers to shorten the recruitment effect. Since the software of smart terminals (such as mobile phones or computers) now has the background running function, the server will automatically collect its time as online time when it is in the background running state, and the job seeker who has not performed any operation for a long time is regarded as not in the state. Job-seeking activity, then this state is regarded as the job seeker being idle, and the time obtained by removing the time of this state is the job seeker's real job-seeking time. The activity percentage is obtained by dividing the real job-seeking time by the total time of the day. The higher the activity percentage, the more eager the job seeker is for the job demand. The probability of the job seeker joining the job is proportional to the activity percentage. The more eager the job seeker is for the job demand, the greater the probability of joining the job. Recruiters can screen job seekers by activity percentage when urgently hiring. Since the above data are all from the previous day, the data of the previous day as the most recent data has a relatively high calculation accuracy. The degree of job-seeking desire can be deduced by the activity data of the previous day. If there is no online data on the previous day, it is counted as a new user, and the number of new users is used as an indicator.When the number of times new users are marked is high, it can be deduced that the job seeker is not eager to find a job. A rectangular coordinate system with two Y axes is established with one day as the X-axis and online activity and job visits as the Y-axis. The time when online activity is recorded as 1 is the line segment. The activity level is obtained by dividing the length of the online active line segment by the length of the day. The activity level is used to further judge the job seeker's desire to find a job. The length of the active line segment marked and the total length of the active line segment are extracted to find the quotient. When the quotient value is less than 0.7, it means that the job seeker spends a small part of his time in daily working hours, and the job seeker can be given a high level of desire to find a job. A temporary label (i.e., an employed status label) is set for the applicant. The smaller the quotient value, the higher the accuracy of judging the applicant's employed status. On the contrary, if the quotient value is greater than or equal to 0.7, it means that the applicant is in daily working hours most of the time. A temporary label (i.e., an off-duty status label) can be set for the applicant. The larger the quotient value, the higher the accuracy of judging the applicant's employed status. By judging the applicant's true employed status, the recruitment efficiency can be improved. If the applicant is in the off-duty status, the probability of successful recruitment is higher and the time to join the job is shorter. If the applicant is in the employed status, the time to successfully join the job is longer, which can To provide more accurate push notifications for urgent jobs, you can enter keywords in the server category library as needed (keywords can be salary, working hours, rest time, holidays, benefits and other related information). The job seekers and the same recruiter communicate as a group, and compare the keyword repetition values ​​of each group of communication information. When the repetition value occupies a communication information group quantity quotient value greater than or equal to 0.5, the repetition value keyword is marked on the job seeker information, and then the job seeker information marked with keyword information is synchronized to the recruitment end, so that the recruiters using the recruitment end can intuitively and quickly screen out the job seekers they like, thereby improving the recruitment efficiency. Job seekers can manually enter quick messages on the job search end, and compare the extracted reply messages with quick messages. Job seekers and the same recruiter communicate as a group, and compare the repetition values ​​of each group of communication information quick messages. If the repetition value of the quick message is higher than 50%, it can be identified as a low-expectation position, otherwise it is a high-expectation position, indicating that the job seeker is very interested in this position. The server synchronizes the job seeker information to the recruitment end with the same position, thereby improving the recruitment efficiency. The recruiter enters α1...α through the recruitment end, n , where α1…α n For the cumulative sum to be 1, α1…α nThe size is the degree of importance that the recruiter attaches to each label (for example, label1 is salary, label1 is benefits, label1 is working hours, and the recruiter needs to recruit people with the highest requirements for benefits, the second highest requirements for salary, and the lowest requirements for working hours. Then, according to their own requirements, α1=0.5, α2=0.3, α3=0.2 can be input, and then the recruiting end will pass α1=0.5, α2=0.3, α3=0.2 to the server, and the server will calculate k based on the data and the representative value of the job seeker's own label). The k size value is sorted. The higher the ranking, the more it meets the recruiter's needs, thereby improving the recruitment rate. If the proportion of read messages is greater than 30%, it can be deduced that too many people have submitted resumes to the recruiter or the recruitment is not urgent. The job seekers can choose the companies ranked higher on the main recruitment ranking list to communicate, thereby improving the recruitment communication effect. Since most recruiters are HR specialists, the working hours of HR specialists can be deduced the next day by recording the time period of recruiters' replies on the previous day. The current time can be compared with the activity time of the recruiters on the previous day to determine whether they are in working status. By eliminating positions that are not in working hours, the communication efficiency between job seekers and recruiters can be improved.

[0062] The above are only preferred embodiments of the present invention, and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention. In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

Claims

1. A recruitment system that can be analyzed by activity, characterized in that: The system includes a server, a recruitment terminal and a personal terminal; The recruitment end is granted recruitment authority, where employers can publish recruitment information through the recruitment end; The job-seeking terminal is granted with job-seeking authority, where job seekers can publish job-seeking information through the job-seeking terminal; The server is configured to connect the recruiting end and the job-seeking end, wherein the server screens the job-seeking information released by the job-seeking end through a screening module and then transmits it to the recruiting end, and the server screens the job-seeking information released by the recruiting end through a screening module and then transmits it to the job-seeking end.

2. The recruitment system capable of analyzing by activity according to claim 1, characterized in that: The server includes a job search module, a recruitment module and a networking module; A job search module, wherein the job search module is used to store and process job search information; Recruitment module, where the recruitment module is used to store and process recruitment information; The networking module is used to connect the Internet with the job search module and the recruitment module. The job titles in the recruitment module are synchronized to the job search module through the networking module. Job seekers select the synchronized job titles in the job search module for entry.

3. The recruitment system capable of analyzing by activity according to claim 2, characterized in that: The job seeker information entered by each of the job seeker terminals is transmitted to the job search module through the networking module, wherein the job search module includes a job search storage unit and a job search processing unit, and the job seeker information is packaged as an information subset and stored in the storage unit, wherein the information subset is marked with the title of the job sought by the job seeker as the subset name, and the information subset name mark is compared with each job title in the networking module through a comparison program, and the information subset name marks with the same name are classified into the same information set and stored.

4. The recruitment system capable of analyzing by activity according to claim 3, characterized in that: The server also includes an activity determination module; Activity judgment module, through Calculate, where act is the activity, P is the online time, P i The access records of users during the online period are extracted for the time period without operation. Each access record is screened to remove the time when no operation is performed for a long time to obtain the real online time. The real online time of one day is accumulated to obtain the effective online time of one day. The activity data calculated on the day comes from the data of the previous day. If there is no online data on the previous day, it is automatically marked as a new user, and the number of new users is accumulated and counted.

5. The recruitment system capable of analyzing by activity according to claim 4, characterized in that: Extract various behavioral data of job seekers during their active period the previous day, where the behavioral data includes the number of job post visits and active time periods. A rectangular coordinate system with a double Y-axis is established with one day as the X-axis, the job post visit status and the active status as the Y-axis, where online activity is recorded as 1 and the job post visit is recorded as 1 in the coordinate system. Online activity exists in the form of line segments in the coordinate system, and the visit status exists in the form of points in the coordinate system. Set a time axis with one day as the interval, and mark the time periods from 9 a.m. to 12 noon and from 2 p.m. to 5 p.m. on the time axis. Mark the time points when online activity is recorded as 1 on the time axis, count the number of time periods falling into the marked area, and then find the quotient with the total number, mark the time points when the job post is recorded as 1 on the time axis, count the number of time periods falling into the marked area, and then find the quotient with the total number, and judge the job seeker's job search expectations by the size of the quotient.

6. The recruitment system capable of analyzing by activity according to claim 5, characterized in that: The behavioral data of job seekers during their active period the day before is extracted, including the content of reply messages. Multiple category libraries are established in the server, and finally keywords are manually entered into the category libraries. By comparing the extracted reply message content with the keywords in each category library, if there is any overlap with the keywords in the category library, the name of the category library is used as a label to the job seeker information, and the job seeker’s real needs are marked on the job seeker information in real time.

7. The recruitment system capable of analyzing by activity according to claim 6, characterized in that: The server extracts the behavioral data of job seekers during their active period the day before, extracts the message content sent by job seekers the day before, and compares the extracted chat content with the quick messages edited by the job seekers in the job search portal. Among them, positions with a quick message ratio higher than 50% are low-expectation positions, and the rest are high-expectation positions. The job seekers' resume information is then sent to the remaining recruiting companies with the same open high-expectation positions.

8. The recruitment system capable of analyzing by activity according to claim 7, characterized in that: In the server, k = α1×label1+α2×label2+......+α n ×label n Assign weights to each label of each job seeker, where k is the ranking coefficient, α1…α n The assignment coefficient, label1, and label2 are manually entered by the recruitment end. n A constant 1 is calculated for the value of each tag. If the job seeker does not have the tag, a constant 0 is calculated. The job seekers are sorted by the size of the sorting coefficient k, and the sorting table required by the recruitment end for the job seekers is sent from the server to the recruitment end.

9. The recruitment system capable of analyzing by activity according to claim 5, characterized in that: The behavioral data of the recruiter during the active period of the previous day are extracted, and the data are divided into the reply time period and the proportion of unread messages. The data are sorted from low to high by the proportion of unread messages, and the messages that have been read but not replied are recorded as unread messages. This sorting is the main recruitment sorting, and the reply time period is marked as a filtering label on the recruiter's job information. The server pushes the main recruitment sorting of each job information to the job seeker with corresponding job requirements.

10. The recruitment system capable of analyzing by activity according to claim 9, characterized in that: The server refreshes and pushes the main recruitment information according to the real-time time. The job seeker can choose to filter and prioritize quick reply messages. At this time, the recruiters who do not reply to the message at the current time are hidden in the main recruitment information sorting, and the new main recruitment information sorting is sent to the job seeker.

Images