Position information recommendation method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a position information recommendation method and device, electronic equipment and storage medium, the method comprises: in the network recruitment process, the user data of different users can be obtained, wherein the user data can include user behavior data, user attribute information corresponding to the user behavior data, and target positions corresponding to the user behavior data, then the job-seeking association data corresponding to the user attribute information and the position attribute information of the target position can be generated according to the user attribute information and the target position, and then the group relationship data corresponding to the user attribute information and the position attribute information can be generated according to the job-seeking association data, the job-seeking tendency of the same type of user can be determined by constructing the group relationship data associated with the user, so as to expand the diversity and relevance of the position recommendation, and when receiving a position request instruction, the position recommendation information for the position request instruction can be obtained according to the group relationship data.

Description

Technical Field

[0001] This invention relates to the field of job information recommendation technology, and in particular to a job information recommendation method, a job information recommendation device, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the rise and development of the internet and artificial intelligence technologies, more and more people are searching for satisfactory jobs on various recruitment platforms. At the same time, recruitment platforms are also utilizing internet technologies to provide users with jobs that match their job search intentions, such as artificial intelligence, search recommendations, user profiling, group profiling, and knowledge graphs. During the job search process, users generate a massive amount of interaction with recruitment platforms, including browsing, clicking, communicating, and submitting applications for desired positions. This behavioral data reflects certain job search intentions. While recruitment platforms can recommend positions based on user behavior, this approach, while improving the match between positions and users, cannot provide users with richer job information. Furthermore, for new users, since their behavioral data is not yet available, recruitment platforms cannot recommend suitable job information. Summary of the Invention

[0003] This invention provides a method, apparatus, electronic device, and computer-readable storage medium for recommending job information, in order to solve or partially solve the problems of insufficient diversity and relevance of job information in the job information recommendation process.

[0004] This invention discloses a method for recommending job information, comprising:

[0005] Acquire user data from different users, the user data including user behavior data, user attribute information corresponding to the user behavior data, and target job titles corresponding to the user behavior data;

[0006] Based on the user attribute information and the target job, generate job search association data corresponding to the user attribute information and the job attribute information of the target job;

[0007] Based on the job search association data, generate group relationship data corresponding to the user attribute information and the job attribute information;

[0008] In response to a job request instruction, the system obtains job information relevant to the job request instruction based on the group relationship data and recommends the job information.

[0009] Optionally, generating job-seeking association data corresponding to the user attribute information and the job attribute information of the target job based on the user attribute information and the target job includes:

[0010] Obtain the job attribute information of the target job;

[0011] Using the user attribute information and the job attribute information, job search association data corresponding to the target job is generated.

[0012] Optionally, the user attribute information includes different user attribute items, the job attribute information includes different job attribute items, and the step of generating group relationship data corresponding to the user attribute information and the job attribute information based on the job search association data includes:

[0013] Obtain the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data;

[0014] Based on the first support of the user attribute item and the second support of the job attribute item, an attribute item tree corresponding to the job search related data is constructed, and the attribute item tree includes several attribute paths;

[0015] Using the attribute path, several attribute item sets corresponding to the attribute item tree are determined, and the attribute item sets include user attribute combinations and job attribute combinations;

[0016] Obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence level corresponding to each attribute item set based on the third support and the fourth support.

[0017] The set of attribute items with a confidence level greater than or equal to a preset confidence threshold is used as the group relationship data in the job search association data corresponding to the user attribute information and the job attribute information.

[0018] Optionally, constructing an attribute tree corresponding to the job search data based on the first support of the user attribute item and the second support of the job attribute item includes:

[0019] User attribute items with a first support level less than a preset support threshold and job attribute items with a second support level less than the preset support threshold are used as the first attribute items;

[0020] User attribute items with a first support level greater than or equal to the preset support threshold and job attribute items with a second support level greater than or equal to the preset support threshold are used as second attribute items;

[0021] By linking the first attribute item with the second attribute item, an attribute item tree corresponding to the job application data is constructed.

[0022] Optionally, the step of linking the first attribute item and the second attribute item to construct an attribute item tree corresponding to the job application data includes:

[0023] Based on the support level of each of the second attribute items, construct an attribute item form corresponding to the second attribute item;

[0024] Remove each of the first attribute items from the job-related data to which they belong, and sort the second attribute items in the same job-related data according to the size of their support to generate target data corresponding to the job-related data.

[0025] Using the attribute item form and the target data, construct an attribute item tree for the job application related data.

[0026] Optionally, constructing an attribute tree for the job-related data using the attribute item form and the target data includes:

[0027] According to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes in sequence to construct the attribute path corresponding to the target data;

[0028] Using the attribute paths corresponding to the target data, construct an attribute item tree for the job-related data.

[0029] Optionally, the step of constructing an attribute tree for the job-related data using the attribute item form and the target data further includes:

[0030] If there are duplicate attribute nodes, increment the count of the corresponding attribute node by one;

[0031] If no identical attribute node exists, a new attribute node is created.

[0032] Optionally, the attribute path includes attribute nodes, and the step of using the attribute path to determine a plurality of attribute item sets corresponding to the attribute item tree includes:

[0033] Obtain the condition information corresponding to the attribute node in each attribute path;

[0034] The conditional information is used to perform data mining on the attribute paths to generate several attribute item sets corresponding to each attribute path.

[0035] Optionally, calculating the confidence level corresponding to each of the attribute itemsets based on the third support and the fourth support includes:

[0036] The target support corresponding to the attribute item set is calculated by using the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination.

[0037] The confidence level corresponding to each attribute item set is calculated using the third support corresponding to the combination of the target support and the user attribute.

[0038] Optionally, in response to a job request instruction, obtaining job information for the job request instruction based on the group relationship data and recommending the job information includes:

[0039] In response to a job request instruction, obtain the user category corresponding to the job request instruction;

[0040] If the user category is represented as a historical user, then based on the group relationship data and the user behavior data, the first job information for the job request instruction is obtained, and the first job information is recommended;

[0041] If the user category table is a new user, then based on the group relationship data and the user attribute information corresponding to the new user, the second job information for the job request instruction is obtained, and the second job information is recommended.

[0042] This invention also discloses a job information recommendation device, comprising:

[0043] The user data acquisition module is used to acquire user data from different users. The user data includes user behavior data, user attribute information corresponding to the user behavior data, and target job corresponding to the user behavior data.

[0044] The job search association data generation module is used to generate job search association data corresponding to the user attribute information and the job attribute information of the target job, based on the user attribute information and the target job.

[0045] The group relationship data generation module is used to generate group relationship data corresponding to the user attribute information and the job attribute information based on the job search association data;

[0046] The job information recommendation module is used to respond to a job request instruction, obtain job information related to the job request instruction based on the group relationship data, and recommend the job information.

[0047] Optionally, the job-seeking related data generation module includes:

[0048] The job attribute information acquisition submodule is used to acquire the job attribute information of the target job.

[0049] The job search association data generation submodule is used to generate job search association data corresponding to the target job by using the user attribute information and the job attribute information.

[0050] Optionally, the user attribute information includes different user attribute items, the job attribute information includes different job attribute items, and the group relationship data generation module includes:

[0051] The support acquisition submodule is used to acquire the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data.

[0052] The attribute item tree generation submodule is used to construct an attribute item tree corresponding to the job application data based on the first support of the user attribute item and the second support of the job attribute item. The attribute item tree includes several attribute paths.

[0053] The attribute item set determination submodule is used to determine several attribute item sets corresponding to the attribute item tree using the attribute path. The attribute item sets include user attribute combinations and job attribute combinations.

[0054] The confidence calculation submodule is used to obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence of each attribute item set based on the third support and the fourth support.

[0055] The group relationship data generation submodule is used to take the set of attribute items with a confidence level greater than or equal to a preset confidence threshold as the group relationship data in the job application association data corresponding to the user attribute information and the job attribute information.

[0056] Optionally, the attribute item tree generation submodule is specifically used for:

[0057] User attribute items with a first support level less than a preset support threshold and job attribute items with a second support level less than the preset support threshold are used as the first attribute items;

[0058] User attribute items with a first support level greater than or equal to the preset support threshold and job attribute items with a second support level greater than or equal to the preset support threshold are used as second attribute items;

[0059] By linking the first attribute item with the second attribute item, an attribute item tree corresponding to the job application data is constructed.

[0060] Optionally, the attribute item tree generation submodule is specifically used for:

[0061] Based on the support level of each of the second attribute items, construct an attribute item form corresponding to the second attribute item;

[0062] Remove each of the first attribute items from the job-related data to which they belong, and sort the second attribute items in the same job-related data according to the size of their support to generate target data corresponding to the job-related data.

[0063] Using the attribute item form and the target data, construct an attribute item tree for the job application related data.

[0064] Optionally, the attribute item tree generation submodule is specifically used for:

[0065] According to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes in sequence to construct the attribute path corresponding to the target data;

[0066] Using the attribute paths corresponding to the target data, construct an attribute item tree for the job-related data.

[0067] Optionally, the attribute item tree generation submodule is further used for:

[0068] If there are duplicate attribute nodes, increment the count of the corresponding attribute node by one;

[0069] If no identical attribute node exists, a new attribute node is created.

[0070] Optionally, the attribute path includes attribute nodes, and the attribute itemset determination submodule is specifically used for:

[0071] Obtain the condition information corresponding to the attribute node in each attribute path;

[0072] The conditional information is used to perform data mining on the attribute paths to generate several attribute item sets corresponding to each attribute path.

[0073] Optionally, the confidence calculation submodule is specifically used for:

[0074] The target support corresponding to the attribute item set is calculated by using the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination.

[0075] The confidence level corresponding to each attribute item set is calculated using the third support corresponding to the combination of the target support and the user attribute.

[0076] Optionally, the job information recommendation module includes:

[0077] The user category acquisition submodule is used to acquire the user category corresponding to the job request instruction in response to the job request instruction;

[0078] The first recommendation submodule is used to obtain first job information for the job request instruction based on the group relationship data and the user behavior data if the user category is characterized as a historical user, and then recommend the first job information.

[0079] The second recommendation submodule is used to, if the user category table is a new user, obtain second job information for the job request instruction based on the group relationship data and the user attribute information corresponding to the new user, and recommend the second job information.

[0080] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0081] The memory is used to store computer programs;

[0082] When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.

[0083] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.

[0084] The embodiments of the present invention have the following advantages:

[0085] In this embodiment of the invention, during the online recruitment process, user data from different users can be acquired. This user data may include user behavior data, corresponding user attribute information, and target positions. Then, job-seeking association data corresponding to the user attribute information and the target position's job attribute information can be generated based on the user attribute information and the target position. Next, group relationship data corresponding to the user attribute information and the job attribute information can be generated based on the job-seeking association data. By constructing group relationship data associated with users, the job-seeking tendencies of users of the same type can be determined, thereby expanding the diversity and relevance of job recommendations. Simultaneously, when a job request instruction is received, job recommendation information for the job request instruction can be obtained based on the group relationship data. Thus, after determining the job-seeking tendencies of users of the same type, for each user's job request, the group circle matching the user can be determined through the group relationship data, and relevant job information can be recommended to the user. This not only enriches the diversity of job information but also improves the relevance between job information and users. Attached Figure Description

[0086] Figure 1 This is a flowchart illustrating the steps of a job information recommendation method provided in an embodiment of the present invention;

[0087] Figure 2 This is a schematic diagram of attribute item processing provided in an embodiment of the present invention;

[0088] Figure 3 This is a schematic diagram of the FP-Tree construction provided in the embodiments of the present invention;

[0089] Figure 4 This is a schematic diagram of the FP-Tree construction provided in the embodiments of the present invention;

[0090] Figure 5 This is a schematic diagram of the FP-Tree construction provided in the embodiments of the present invention;

[0091] Figure 6 This is a schematic diagram illustrating the construction of the conditional mode base provided in an embodiment of the present invention;

[0092] Figure 7 This is a structural block diagram of a job information recommendation device provided in an embodiment of the present invention;

[0093] Figure 8 This is a block diagram of an electronic device provided in an embodiment of the present invention;

[0094] Figure 9 This is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. Detailed Implementation

[0095] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0096] As an example, with the rise and development of internet and artificial intelligence technologies, more and more people are searching for satisfactory jobs on various recruitment platforms. At the same time, these platforms are also leveraging internet technologies to provide users with jobs that match their job search preferences, such as artificial intelligence, search recommendations, user profiling, group profiling, and knowledge graphs. During the job search process, users generate a massive amount of interaction with recruitment platforms, including browsing, clicking, communicating, and submitting applications for desired positions. This behavioral data reflects certain job search intentions of job seekers.

[0097] Different users exhibit varying behavioral patterns due to factors such as their geographical location, preferred job positions, expected salary, gender, age, work experience, and education level. Consequently, their interactions with recruitment platforms differ, and different job-seeking user groups inevitably possess their own group behavioral characteristics. Therefore, by acquiring these group behavioral characteristics, recruitment platforms can effectively segment users based on their different attributes. Then, by considering the behavioral habits of these different user groups, recruitment platforms can better recommend jobs that match users' job preferences, improving the accuracy and diversity of job information recommendations, thus achieving the goal of personalized recruitment experiences.

[0098] However, in related technologies, recruitment platforms recommend jobs that match user preferences based on user profiles, tags, and similar positions. While this method can ensure the accuracy of job recommendations, it falls short in terms of job information diversity. Furthermore, for new users of recruitment platforms, since there is no corresponding user behavior data, the platform can only recommend jobs based on the basic information provided. This may lead to situations where recommended jobs do not match user needs, and the limited job information available from the basic information provided by the user cannot provide diverse job opportunities.

[0099] In this regard, one of the core inventive points of this invention is to construct group relationship data based on the user behavior habits generated by different users and using data mining technology. This allows for the segmentation of users from different regions, with different interests, different experiences, and different genders and ages, thus building group circles. When a user initiates a job request, the system can recommend corresponding job information to the user by finding the job-seeking tendencies of people in the same group circles, thereby expanding the diversity and relevance of job recommendations on the recruitment platform.

[0100] To enable those skilled in the art to better understand the technical solutions of the embodiments of the present invention, some technical names involved in the embodiments of the present invention will be explained and described below.

[0101] Support can be the number of times an attribute item appears in a dataset, or the percentage of the number of times an attribute item appears in a dataset relative to the total number of all attribute items in the dataset.

[0102] Confidence level can be considered as the probability that another label will appear after the first label has appeared; it is a conditional probability.

[0103] An attribute item tree, which can be an FP-Tree (frequent pattern tree), can consist of a root node (with a value of null), an item prefix subtree (as children), and a frequent item table.

[0104] The attribute item can be a frequent item corresponding to a node in the FP-Tree. In this embodiment of the invention, the frequent items constituting the FP-Tree can include user attribute items and job attribute items. For user attribute items, it can be the basic information corresponding to the user (such as gender, age, education, experience, city, position, salary, etc.); for job attribute items, it can be the basic information corresponding to the job (such as position, benefits, salary, industry, etc.).

[0105] An attribute path can be a node path corresponding to a data point in an FP-Tree. Correspondingly, the condition information can be the Conditional Pattern Base corresponding to the attribute path. Specifically, this can be achieved by identifying the attribute nodes for each attribute path, then determining the Conditional Pattern Base based on the prefix paths of the attribute nodes, and finally determining the attribute itemset corresponding to each attribute node using the Conditional Pattern Base. For the attribute itemset, it can be a frequent itemset. During data mining of the FP-Tree, the Conditional Pattern Base uses the attribute nodes being mined as the leaf nodes of the FP subtree. After obtaining this FP subtree, the count of each node in the subtree can be set to the count of the leaf node, and nodes with counts lower than the support can be deleted, thus obtaining the frequent itemset corresponding to that attribute node.

[0106] Reference Figure 1 The diagram illustrates a flowchart of a job information recommendation method provided in an embodiment of the present invention, which may specifically include the following steps:

[0107] Step 101: Obtain user data for different users, including user behavior data, user attribute information corresponding to the user behavior data, and target job corresponding to the user behavior data;

[0108] It should be noted that the acquisition of user data can be data generated by the recruitment platform during the user's use of the recruitment platform after obtaining the user's authorization. Thus, user data is only collected with the user's authorization, which can effectively ensure the privacy and security of user data.

[0109] In this embodiment of the invention, during the online recruitment process, users can search for jobs online through the application corresponding to the recruitment platform. During the online job search process, users can first fill in their own user attribute information in the application corresponding to the recruitment platform, and then enter the corresponding job search request. Then, the application can recommend job information based on the user's job search request and user attribute information. After the user uses the application for a certain period of time, corresponding user behavior data can be generated. The application can also recommend job information based on the user behavior data, etc. This invention does not limit this.

[0110] In practical implementation, before recommending job information, big data mining can be used to extract user group characteristics. These characteristics can group users with similar attributes into clusters, where similar job-seeking needs may exist. This user group characteristic can then assist recruitment platforms in recommending diverse job information. Specifically, massive amounts of user behavior data generated during user interactions with the recruitment platform can be used to mine group relationships. This user behavior data can include browsing behavior data, communication behavior data between users and recruiting companies, click behavior data related to clicks on interested positions, and application behavior data, with application behavior revealing the user's true intentions. Browsing behavior data can be represented as a user's browsing behavior of a specific job posting; communication behavior data can be represented as a user's communication behavior with the corresponding recruiter regarding a specific job posting; click behavior data can be represented as a user's clicking behavior on the job posting information corresponding to a specific job posting; and application behavior data can be represented as a user's resume submission behavior for a specific job posting. Therefore, big data mining can be used to process the relationship between user behavior data and target job postings to obtain correlation data between user attribute information and job attribute information, and then group characteristics can be mined based on this correlation data.

[0111] Step 102: Based on the user attribute information and the target job, generate job search association data corresponding to the user attribute information and the job attribute information of the target job;

[0112] In practical implementation, different users correspond to different user behavior data and different user attribute information. User attribute information can be determined through user behavior data, or it can be obtained through user identifiers, including gender, age, education, experience, city, job title, and salary. Correspondingly, for a target job, the corresponding job attribute information can be obtained, including position, benefits, salary, and industry. Then, based on the user attribute information and job attribute information, job search association data corresponding to the target job can be generated.

[0113] For example, for users A, B, C, and D, where:

[0114] User A's user attribute information includes: gender - male, age - 25 years old, job intention - auto mechanic, and education level - junior college.

[0115] User B's corresponding user attribute information ② includes: gender - male, age - 30 years old, job intention - human resources specialist, and education - bachelor's degree;

[0116] User C's corresponding user attribute information ③ includes: gender - male, age - 27 years old, job intention - patent agent, and education - bachelor's degree;

[0117] User D's corresponding user attribute information ④ includes: gender - female, age - 27 years old, job intention - waiter / waitress.

[0118] During the interaction between the aforementioned users and the recruitment platform, corresponding user behavior data can be generated. Assuming that user A's user behavior data includes browsing behavior data, specifically, if user A browsed target job a, then during data mining, user A's browsing behavior corresponds to target job a, and user attribute information ① corresponding to user A and job attribute information Ⅰ corresponding to target job a can be obtained, including: position - auto mechanic, industry - driver / transportation, salary - 8000, etc. Then, user attribute information ① and job attribute information Ⅰ can be concatenated to obtain the job-seeking association data "gender - male, age - 25 years old, job intention - auto mechanic, education - junior college, position - auto mechanic, industry - driver / transportation, salary - 8000", which represents that users with user attribute information ① prefer jobs with job attribute information Ⅰ.

[0119] Assuming user B's user behavior data includes communication behavior data, specifically, user B communicating online with relevant recruiters regarding target job b, then during data mining, user B's communication behavior corresponds to target job b. This allows us to obtain user attribute information ② for user B and job attribute information Ⅱ for target job b, including: position - HR specialist, bachelor's degree, salary - 6500, etc. Then, user attribute information ② and job attribute information Ⅱ can be concatenated to obtain job-related data such as "gender - male, age - 30 years old, job intention - HR specialist, bachelor's degree, position - HR specialist, bachelor's degree, salary - 6500". This data represents that users with user attribute information ② prefer jobs with job attribute information Ⅱ.

[0120] Assuming user C's user behavior data includes click behavior data, specifically, if user C clicks on target job c while searching for a job online, then during data mining, user C's communication behavior corresponds to target job c. This allows us to obtain user attribute information ③ for user C and job attribute information Ⅲ for target job c, including: position - patent attorney, bachelor's degree, salary - 10000, etc. Then, user attribute information ③ and job attribute information Ⅲ can be concatenated to obtain the job-related data: "gender - male, age - 27, job intention - patent attorney, bachelor's degree, bachelor's degree, position - patent attorney, bachelor's degree, salary - 10000". This data represents that users with user attribute information ③ prefer jobs with job attribute information Ⅲ.

[0121] Assuming user D's user behavior data includes application behavior data, specifically, user D submitting a resume for target position d while searching for a job online, then during data mining, user D's communication behavior corresponds to target position d, allowing us to obtain user attribute information ④ for user D and job attribute information IV for target position d, including: position - restaurant server, salary - 5500, benefits - room and board included, industry - catering, etc. Then, user attribute information ④ and job attribute information IV can be concatenated to obtain job-related data such as "gender - female, age - 27 years old, job intention - server, position - restaurant server, salary - 5500, benefits - room and board included, industry - catering", which indicates that users with user attribute information ④ prefer positions with job attribute information IV.

[0122] Through the above process, user behavior data of different users on recruitment platforms can be obtained, and big data analysis and mining can be performed on the user behavior data to obtain job search correlation data between user attribute information and job attribute information. This allows for further data mining based on the job search correlation data to obtain group characteristic data.

[0123] Step 103: Based on the job search association data, generate group relationship data corresponding to the user attribute information and the job attribute information;

[0124] Each job-related data point can include user attribute information and job attribute information, used to represent the relationship between user attributes and job attributes. User attribute information can include different user attribute items, such as gender, age, education, experience, city, position, and salary, each corresponding to different user attributes; job attribute information can include different job attribute items, such as position, benefits, salary, and industry, each corresponding to different job attributes.

[0125] For group relationship data, we can obtain the first support of each user attribute item in all job-related data and the second support of each job attribute item in all job-related data. Then, based on the first support of user attribute items and the second support of job attribute items, we construct an attribute item tree corresponding to the job-related data. The attribute item tree includes several attribute paths. Then, using the attribute paths, we determine several attribute item sets corresponding to the attribute item tree. Each attribute item set includes user attribute combinations and job attribute combinations. Next, we obtain the third support corresponding to the user attribute combinations and the fourth support corresponding to the job attribute combinations. Based on the third and fourth support, we calculate the confidence level corresponding to each attribute item set. The attribute item sets with confidence levels greater than or equal to a preset confidence threshold are taken as the group relationship data in the job-related data corresponding to user attribute information and job attribute information.

[0126] For user and job attribute items, a preset support threshold can be set to filter out user and / or job attribute items that do not meet the support requirement, thus achieving preliminary data screening. Specifically, by scanning all job-related data, the first support of all user attribute items and the second support of all job attribute items can be obtained. Then, user attribute items with a first support less than the preset support threshold and job attribute items with a second support less than the preset support threshold can be designated as first attribute items, and user attribute items with a first support greater than or equal to the preset support threshold and job attribute items with a second support greater than or equal to the preset support threshold can be designated as second attribute items. Finally, attribute items are linked using the first and second attribute items to construct an attribute item tree corresponding to the job-related data. For example, if support can be the number of times an attribute item appears in all job-related data, then a support threshold of 2 can be set. By scanning all job-related data, the number of times each user attribute item appears in the job-related data and the number of times each job attribute item appears in the job-related data can be obtained. Then, the user attribute item and / or job attribute item that appears 1 time in each job-related data are filtered. The remaining user attribute item after filtering is taken as the first attribute item and the remaining job attribute item is taken as the second attribute item. The attribute items are then linked to construct an attribute tree corresponding to the job-related data, that is, an FP-Tree is constructed through attribute items.

[0127] Before constructing the FP-Tree, we can first construct an attribute item form corresponding to each second attribute item according to the support of each second attribute item, remove each first attribute item from the job-related data, sort the second attribute items in the same job-related data according to the support of each second attribute item, generate target data corresponding to the job-related data, and then construct the attribute item tree for the job-related data using the attribute item form and the target data.

[0128] Specifically, the attribute item form can be a header table, and the target data can be the data obtained after filtering and sorting each job-related data. Then, according to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes to construct an attribute path corresponding to the target data. Finally, using the attribute path corresponding to the target data, an attribute item tree is constructed for the job-related data. During the construction of the attribute path, if duplicate attribute nodes exist, their count is incremented; otherwise, a new attribute node is created.

[0129] In one example, refer to Figure 2This diagram illustrates the attribute item processing provided in this embodiment of the invention. It assumes that the job-related data may include 10 data entries: ABCEFO, ACG, EI, ACDEG, ACEGL, EJ, ABCEFP, ACD, ACEGM, and ACEGN. ​​Different letters can represent user attribute items or job attribute items. Each job-related data entry includes at least one user attribute item and at least one job attribute item. When generating the attribute item tree (FP-Tree), these 10 job-related data entries are first scanned to obtain the frequency (i.e., support) of attribute items A, B, C, D, E, F, G, I, J, N, M, and P. Then, attribute items with a frequency less than 2 (i.e., support less than 20%) are filtered out, including O, I, L, J, P, M, and N. The remaining attribute items (i.e., support greater than 20%) are then used to construct corresponding item header tables, and the frequency of each attribute item is counted, including A: 8, C: 8, E: 8, G: 5, B: 2, D: 2, and F: 2. While constructing the header table, the attribute items in each job-related data can be sorted according to the frequency of each attribute item in all job-related data. For example, for ABCEFO, since the frequency of O is less than the support threshold, it is filtered out. Sort by frequency to get ACEBF. Other job-related data are sorted in the same way to obtain the target data, including: ACEBF, ACG, E, ACEGD, ACEG, E, ACEBF, ACD, ACEG, ACEG.

[0130] Once the header table and sorted target data are obtained, the FP-Tree can be constructed. Specifically, initially, the FP-Tree contains no data. When building the FP-Tree, the sorted target data is read sequentially, and the corresponding attribute items from the target data are added to the FP-Tree in the sorted order. For example, when adding to the FP-Tree, the nodes that appear earlier in the sorted list can be used as parent nodes, and those that appear later can be used as child nodes. If there is a shared parent node, the count of the shared parent node is incremented by 1. At the same time, during the addition process, if a new node appears, the node corresponding to the header table will be linked to the new node through the node linked list until all the data is added to the FP-Tree, thus completing the construction of the FP-Tree corresponding to the job search data.

[0131] refer to Figure 3 This illustrates a schematic diagram of FP-Tree construction provided in an embodiment of the present invention, with the first sorted target data ACEBF added, as shown below. Figure 3 As shown, the FP-Tree has no nodes at this point, therefore ACEBF is an independent path, with all nodes counted as 1. The item header table is linked to the corresponding newly added node through the node linked list. Next, refer to... Figure 4This diagram illustrates the construction of an FP-Tree provided in this embodiment of the invention. A second target data item, ACG, is added. Since ACG and the existing FP-Tree can share a common parent node sequence AC, only a new node G needs to be added, and the count of the new node G is set to 1. Simultaneously, the counts of A and C are incremented by 1 to become 2. Of course, the node list of the corresponding node G needs to be updated. Correspondingly, other target data can be processed using the above process to ultimately complete the construction of the FP-Tree. Figure 5 The diagram illustrates the construction of an FP-Tree provided in this embodiment of the invention. After all the target data is added, an FP-Tree corresponding to all job-related data can be obtained.

[0132] After constructing an FP-Tree for job-related data, we can obtain the conditional information corresponding to the attribute nodes in each attribute path of the FP-Tree. We then use this conditional information to perform data mining on the attribute paths, generating several attribute item sets corresponding to each attribute path. Each attribute item set can include user attribute combinations and job attribute sets. We can then obtain the third support corresponding to user attribute combinations and the fourth support corresponding to job attribute combinations. Next, we use the third support and the fourth support to calculate the target support for each attribute item set. Finally, we use the target support and the third support to calculate the confidence level for each attribute item set. Thus, attribute item sets with confidence levels greater than or equal to a pre-set confidence threshold can be used as group relationship data corresponding to user attribute information and job attribute information in the job-related data. Through this process, we can mine group relationship data from massive amounts of user behavior data. This allows us to locate the user's associated group circles when the user initiates a job request, and recommend corresponding job information to the user based on these group circles, enriching the diversity of job information and improving the relevance between job information and users.

[0133] In the specific implementation, for a set of attribute items, a conditional pattern base can be obtained from the FP-Tree. Then, a conditional FP-Tree can be constructed using the conditional pattern base, and the above steps can be repeated iteratively until the FP-Tree contains an attribute item. The conditional pattern base can be a set of paths ending with the queried attribute item. Each path is actually a prefix path, meaning a prefix path contains all content between the queried attribute item and the root node. During the query process, the corresponding prefix path can be obtained through the created attribute form. The attribute form can contain the starting attribute items of the same type of attribute chain table. When querying the corresponding attribute item, the prefix nodes involved can be traced up the FP-Tree until the root node of the FP-Tree.

[0134] In one example, for FP-Tree data mining, after obtaining the FP-Tree and the attribute item list (item header table), we can start mining from the bottom of the item header table upwards. For each attribute item in the attribute item list relative to the FP-Tree, we need to find its conditional pattern base. Specifically, we can use the attribute node to be mined as the leaf node to find the corresponding FP-sub-Tree. Once we obtain the FP-sub-Tree, we can set the count of each attribute node in the subtree to the count of the leaf node, and delete nodes with counts lower than the support. Then, we recursively mine from the obtained conditional pattern base to obtain frequent itemsets. (See reference...) Figure 6 This diagram illustrates the construction of the conditional pattern base provided in this embodiment of the invention. Starting from the bottom F node of the attribute item form, the conditional pattern base of node F is searched within the FP-Tree. Since F has only one node in the FP-Tree, the only candidates are the following... Figure 6 The path shown in ① corresponds to {A:8, C:8, E:6, B:2, F:2}. Next, the counts of all parent nodes can be set to the counts of the leaf nodes, meaning the FP-Tree becomes {A:2, C:2, E:2, B:2, F:2}. Leaf nodes can be omitted from the conditional pattern base, resulting in the final conditional pattern base for F as follows: Figure 6 As shown in ②, the conditional pattern base corresponding to attribute item F is {A:2, C:2, E:2, B:2}.

[0135] After obtaining the conditional pattern base, it is recursively applied to the corresponding attribute itemsets. For example, by obtaining the frequent 2-itemsets of F as {A:2, F:2}, {C:2, F:2}, {E:2, F:2}, {B:2, F:2}, and then recursively merging the 2-itemsets, we obtain the frequent 3-itemsets as {A:2, C:2, F:2}, {A:2, E:2, F:2}, and so on. At the same time, we can continue recursively to obtain the largest frequent itemset, the frequent 5-itemset, which is {A:2, C:2, E:2, B:2, F:2}. Thus, we can obtain the number of attribute itemsets corresponding to attribute item F.

[0136] For each attribute set, it can be split into different types of attribute items, namely, user attribute combinations and job attribute combinations. Each user attribute combination or job attribute combination includes at least one attribute item. For example, in the above process, {A:2, F:2} can correspond to one user attribute item and one job attribute item; for {A:2, C:2, F:2}, it can correspond to one user attribute item and two job attribute items, or two user attribute items and one job attribute item, and so on. Thus, by splitting the attribute set, various combination relationships between user attribute items and job attribute items can be obtained. Furthermore, during the splitting process, the attribute set can be split into the form A => B, where A represents user attribute item combinations, including gender, age, education, job intention, etc., and B represents job information attributes, including position, salary, benefits, industry, etc. Then, the confidence score of A => B is calculated, and those that meet the confidence score threshold are considered as group relationship data. Optionally, confidence represents a conditional probability, i.e., the probability of B occurring given that A has occurred. The probability of B occurring given A can be calculated as follows: Confidence(A=>B) = P(B|A) = Support(AUB) / Support(A), where Support(AUB) is the probability of both A and B occurring simultaneously, and Support(A) is the probability of A occurring. Corresponding to the FP-Tree, the probability can be the support level. For example, in the filtered and sorted target data ACEBF, ACG, E, ACEGD, ACEG, E, ACEBF, ACD, ACEG, ACEG, for F, assuming ACEB is a user attribute and F is a job attribute, then the probability of F occurring given ACEB can be: Confidence(ACBE=>F) = P(F|ACBE) = Support(ACBE) / Support(AUB ... The formula is: UF) / Support(ACBE), where Support(ACBE) has a support of 2 and an occurrence probability of 2 / 10, and Support(F) also has a support of 2 and an occurrence probability of 2 / 10. Using this method, we can obtain that in the case of ACBE, the probability of F occurring is 1 / 5. This probability can then be used as the corresponding confidence level. When the confidence level is greater than or equal to a preset confidence threshold, ACBEF is used as the group relationship data corresponding to the job search association data. Through this iterative process, group relationship data associated with users can be mined from massive amounts of user behavior data. This group relationship data can then be used to assist recruitment platforms in improving the richness and relevance of recommended job information when recommending job information to users.

[0137] In one example, suppose user attribute information and job attribute information are obtained based on user behavior data, as shown in Table 1 below:

[0138]

[0139] Table 1

[0140] Next, by constructing an FP-Tree for data mining, the group relationship data shown in Table 2 below can be obtained:

[0141]

[0142] Table 2

[0143] After obtaining the corresponding group relationship data through the above process, job information can be recommended based on this group relationship data. At the same time, by combining the user's city and basic user information, the "cold start" recommendation problem of the recommendation system for new users can be solved, enabling job recommendations for different types of users and improving the flexibility of job information recommendation.

[0144] It should be noted that the embodiments of the present invention include, but are not limited to, the examples described above. It is understood that, under the guidance of the ideas in the embodiments of the present invention, those skilled in the art can also process user attribute information, job attribute information, etc., according to actual needs, and the present invention does not limit this.

[0145] Step 104: In response to the job request instruction, obtain job information for the job request instruction based on the group relationship data, and recommend the job information.

[0146] Once group relationship data is generated, the recruitment platform can recommend relevant job information based on the job request instructions initiated by users. Specifically, after receiving a job request instruction from a user, the terminal running the recruitment platform can obtain the user category corresponding to the instruction. If the user category is a historical user, the platform obtains the first job recommendation information based on the group relationship data and user behavior data. If the user category is a new user, the platform obtains the second job recommendation information based on the group relationship data and the user attribute information corresponding to the new user. Thus, when a user is a new user, the platform can locate the user's social circles using the established group relationship data and the user attribute information provided. This allows the platform to determine which groups the new user belongs to and recommend matching job information based on the job information corresponding to those groups. For existing users, the platform can increase the diversity of job recommendations by providing more job information with a wider range of employment opportunities, effectively improving the user experience of online job hunting.

[0147] In one example, for user A, assuming they are a returning user, based on user A's job request instruction, the recruitment platform, having already mined group relationship data based on group characteristics, can recommend job information solely based on user A's user behavior data, solely based on user A's user attribute information, or based on group relationship data, or a combination of at least two of the above. By constructing corresponding group relationship data, when recommending job information to returning users, the diversity of job recommendations can be effectively expanded, and the relevance between the recommended jobs and the user can be improved. For user E, assuming they are a new user, since user E's user behavior data is not yet available on the recruitment platform, and the platform only stores the user attribute information they filled in, the recruitment platform can recommend job information to user E based on user attribute information, or it can recommend job information based on group relationship data. By constructing group relationship data, when recommending job information to new users, the problem of insufficient job information can be avoided when recommending job information based on user attribute information. This not only solves the "cold start" problem of job information recommendation, but also allows for group circle positioning through established group relationship data and the user attribute information filled in by the new user. This determines which group circles the new user belongs to, and then recommends matching job information to the new user based on the job information corresponding to that group circle, improving the diversity and relevance of job information recommendations for new users.

[0148] Optionally, in the above process, regarding the process of recommending job information based on group relationship data, since group relationship data can be used to characterize the preference of users with certain user attributes for jobs with relevant job attributes, that is, the correlation between user attribute information and job attribute information, for both new and old users, after obtaining the user's corresponding user attribute information, group circle positioning can be performed through the user attribute information, that is, finding the group relationship data corresponding to the user attribute information, and then job information recommendation can be performed based on the job attribute information corresponding to the group relationship data. For example, Table 3 below shows some group relationship data:

[0149]

[0150] Table 3

[0151] If a new user F's user attributes include: gender - male, age - 25, education - associate degree, etc., then based on these attributes, the user can be placed into a specific group with the following characteristics: gender - male, age - 25, job intention - auto mechanic, education - associate degree. Next, the job attribute information corresponding to this group can be obtained, including position - auto mechanic, industry - driver / transportation, salary - 8000, etc. Job recommendations can then be made based on these job attribute information. It's understandable that for existing users, the same process can be used for job recommendations. Furthermore, combining user behavior data can lead to various methods of job recommendations, expanding the diversity of recommendations and improving the relevance between recommended jobs and users.

[0152] In this embodiment of the invention, during the online recruitment process, user data from different users can be acquired. This user data may include user behavior data, corresponding user attribute information, and target positions. Then, job-seeking association data corresponding to the user attribute information and the target position's job attribute information can be generated based on the user attribute information and the target position. Next, group relationship data corresponding to the user attribute information and the job attribute information can be generated based on the job-seeking association data. By constructing group relationship data associated with users, the job-seeking tendencies of users of the same type can be determined, thereby expanding the diversity and relevance of job recommendations. Simultaneously, when a job request instruction is received, job recommendation information for the job request instruction can be obtained based on the group relationship data. Thus, after determining the job-seeking tendencies of users of the same type, for each user's job request, the group circle matching the user can be determined through the group relationship data, and relevant job information can be recommended to the user. This not only enriches the diversity of job information but also improves the relevance between job information and users.

[0153] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0154] Reference Figure 7 The diagram illustrates a structural block diagram of a job information recommendation device provided in an embodiment of the present invention, which may specifically include the following modules:

[0155] User data acquisition module 701 is used to acquire user data of different users. The user data includes user behavior data, user attribute information corresponding to the user behavior data, and target job corresponding to the user behavior data.

[0156] The job search association data generation module 702 is used to generate job search association data corresponding to the user attribute information and the job attribute information of the target job based on the user attribute information and the target job.

[0157] The group relationship data generation module 703 is used to generate group relationship data corresponding to the user attribute information and the job attribute information based on the job search association data;

[0158] The job information recommendation module 704 is used to respond to a job request instruction, obtain job information for the job request instruction based on the group relationship data, and recommend the job information.

[0159] In one optional embodiment, the job-seeking related data generation module 702 includes:

[0160] The job attribute information acquisition submodule is used to acquire the job attribute information of the target job.

[0161] The job search association data generation submodule is used to generate job search association data corresponding to the target job by using the user attribute information and the job attribute information.

[0162] In one optional embodiment, the user attribute information includes different user attribute items, the job attribute information includes different job attribute items, and the group relationship data generation module 703 includes:

[0163] The support acquisition submodule is used to acquire the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data.

[0164] The attribute item tree generation submodule is used to construct an attribute item tree corresponding to the job application related data based on the first support of the user attribute item and the second support of the job attribute item. The attribute item tree includes several attribute paths.

[0165] The attribute item set determination submodule is used to determine several attribute item sets corresponding to the attribute item tree using the attribute path. The attribute item sets include user attribute combinations and job attribute combinations.

[0166] The confidence calculation submodule is used to obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence of each attribute item set based on the third support and the fourth support.

[0167] The group relationship data generation submodule is used to take the set of attribute items with a confidence level greater than or equal to a preset confidence threshold as the group relationship data in the job application association data corresponding to the user attribute information and the job attribute information.

[0168] In one optional embodiment, the attribute item tree generation submodule is specifically used for:

[0169] User attribute items with a first support level less than a preset support threshold and job attribute items with a second support level less than the preset support threshold are used as the first attribute items;

[0170] User attribute items with a first support level greater than or equal to the preset support threshold and job attribute items with a second support level greater than or equal to the preset support threshold are used as second attribute items;

[0171] By linking the first attribute item with the second attribute item, an attribute item tree corresponding to the job application data is constructed.

[0172] In one optional embodiment, the attribute item tree generation submodule is specifically used for:

[0173] Based on the support level of each of the second attribute items, construct an attribute item form corresponding to the second attribute item;

[0174] Remove each of the first attribute items from the job-related data to which they belong, and sort the second attribute items in the same job-related data according to the size of their support to generate target data corresponding to the job-related data.

[0175] Using the attribute item form and the target data, construct an attribute item tree for the job application related data.

[0176] In one optional embodiment, the attribute item tree generation submodule is specifically used for:

[0177] According to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes in sequence to construct the attribute path corresponding to the target data;

[0178] Using the attribute paths corresponding to the target data, construct an attribute item tree for the job-related data.

[0179] In one optional embodiment, the attribute item tree generation submodule is further configured to:

[0180] If there are duplicate attribute nodes, increment the count of the corresponding attribute node by one;

[0181] If no identical attribute node exists, a new attribute node is created.

[0182] In one alternative embodiment, the attribute path includes attribute nodes, and the attribute itemset determination submodule is specifically used for:

[0183] Obtain the condition information corresponding to the attribute node in each attribute path;

[0184] The conditional information is used to perform data mining on the attribute paths to generate several attribute item sets corresponding to each attribute path.

[0185] In one optional embodiment, the confidence calculation submodule is specifically used for:

[0186] The target support corresponding to the attribute item set is calculated by using the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination.

[0187] The confidence level corresponding to each attribute item set is calculated using the third support corresponding to the combination of the target support and the user attribute.

[0188] In one optional embodiment, the job information recommendation module 704 includes:

[0189] The user category acquisition submodule is used to acquire the user category corresponding to the job request instruction in response to the job request instruction;

[0190] The first recommendation submodule is used to obtain first job information for the job request instruction based on the group relationship data and the user behavior data if the user category is characterized as a historical user, and then recommend the first job information.

[0191] The second recommendation submodule is used to, if the user category table is a new user, obtain second job information for the job request instruction based on the group relationship data and the user attribute information corresponding to the new user, and recommend the second job information.

[0192] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0193] In addition, embodiments of the present invention also provide an electronic device, such as... Figure 8As shown, it includes a processor 801, a communication interface 802, a memory 803, and a communication bus 804. The processor 801, communication interface 802, and memory 803 communicate with each other via the communication bus 804.

[0194] Memory 803 is used to store computer programs;

[0195] When processor 801 executes a program stored in memory 803, it performs the following steps:

[0196] Acquire user data from different users, the user data including user behavior data, user attribute information corresponding to the user behavior data, and target job titles corresponding to the user behavior data;

[0197] Based on the user attribute information and the target job, generate job search association data corresponding to the user attribute information and the job attribute information of the target job;

[0198] Based on the job search association data, generate group relationship data corresponding to the user attribute information and the job attribute information;

[0199] In response to a job request instruction, the system obtains job information relevant to the job request instruction based on the group relationship data and recommends the job information.

[0200] In one optional embodiment, generating job-seeking association data corresponding to the user attribute information and the job attribute information of the target job based on the user attribute information and the target job includes:

[0201] Obtain the job attribute information of the target job;

[0202] Using the user attribute information and the job attribute information, job search association data corresponding to the target job is generated.

[0203] In one optional embodiment, the user attribute information includes different user attribute items, the job attribute information includes different job attribute items, and the step of generating group relationship data corresponding to the user attribute information and the job attribute information based on the job search association data includes:

[0204] Obtain the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data;

[0205] Based on the first support of the user attribute item and the second support of the job attribute item, an attribute item tree corresponding to the job search related data is constructed, and the attribute item tree includes several attribute paths;

[0206] Using the attribute path, several attribute item sets corresponding to the attribute item tree are determined, and the attribute item sets include user attribute combinations and job attribute combinations;

[0207] Obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence level corresponding to each attribute item set based on the third support and the fourth support.

[0208] The set of attribute items with a confidence level greater than or equal to a preset confidence threshold is used as the group relationship data in the job search association data corresponding to the user attribute information and the job attribute information.

[0209] In one optional embodiment, constructing an attribute tree corresponding to the job search association data based on the first support of the user attribute item and the second support of the job attribute item includes:

[0210] User attribute items with a first support level less than a preset support threshold and job attribute items with a second support level less than the preset support threshold are used as the first attribute items;

[0211] User attribute items with a first support level greater than or equal to the preset support threshold and job attribute items with a second support level greater than or equal to the preset support threshold are used as second attribute items;

[0212] By linking the first attribute item with the second attribute item, an attribute item tree corresponding to the job application data is constructed.

[0213] In one optional embodiment, the step of linking the first attribute item with the second attribute item to construct an attribute item tree corresponding to the job application related data includes:

[0214] Based on the support level of each of the second attribute items, construct an attribute item form corresponding to the second attribute item;

[0215] Remove each of the first attribute items from the job-related data to which they belong, and sort the second attribute items in the same job-related data according to the size of their support to generate target data corresponding to the job-related data.

[0216] Using the attribute item form and the target data, construct an attribute item tree for the job application related data.

[0217] In one optional embodiment, constructing an attribute tree for the job-related data using the attribute item form and the target data includes:

[0218] According to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes in sequence to construct the attribute path corresponding to the target data;

[0219] Using the attribute paths corresponding to the target data, construct an attribute item tree for the job-related data.

[0220] In one optional embodiment, the step of constructing an attribute tree for the job-related data using the attribute item form and the target data further includes:

[0221] If there are duplicate attribute nodes, increment the count of the corresponding attribute node by one;

[0222] If no identical attribute node exists, a new attribute node is created.

[0223] In one optional embodiment, the attribute path includes attribute nodes, and the step of using the attribute path to determine a plurality of attribute item sets corresponding to the attribute item tree includes:

[0224] Obtain the condition information corresponding to the attribute node in each attribute path;

[0225] The conditional information is used to perform data mining on the attribute paths to generate several attribute item sets corresponding to each attribute path.

[0226] In one optional embodiment, calculating the confidence level corresponding to each of the attribute itemsets based on the third support and the fourth support includes:

[0227] The target support corresponding to the attribute item set is calculated by using the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination.

[0228] The confidence level corresponding to each attribute item set is calculated using the third support corresponding to the combination of the target support and the user attribute.

[0229] In one optional embodiment, the step of responding to a job request instruction, obtaining job information for the job request instruction based on the group relationship data, and recommending the job information includes:

[0230] In response to a job request instruction, obtain the user category corresponding to the job request instruction;

[0231] If the user category is represented as a historical user, then based on the group relationship data and the user behavior data, the first job information for the job request instruction is obtained, and the first job information is recommended;

[0232] If the user category table is a new user, then based on the group relationship data and the user attribute information corresponding to the new user, the second job information for the job request instruction is obtained, and the second job information is recommended.

[0233] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0234] The communication interface is used for communication between the aforementioned terminal and other devices.

[0235] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0236] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0237] like Figure 9 As shown, in another embodiment of the present invention, a computer-readable storage medium 901 is also provided, which stores instructions that, when executed on a computer, cause the computer to perform the job information recommendation method described in the above embodiments.

[0238] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the job information recommendation method described in the above embodiments.

[0239] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0240] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0241] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0242] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for recommending job information, characterized in that, include: Acquire user data from different users, the user data including user behavior data, user attribute information corresponding to the user behavior data, and target job titles corresponding to the user behavior data; Based on the user attribute information and the target job, job search association data corresponding to the user attribute information and the job attribute information of the target job is generated; the user attribute information includes different user attribute items, and the job attribute information includes different job attribute items. Obtain the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data; Based on the first support of the user attribute item and the second support of the job attribute item, an attribute item tree corresponding to the job search related data is constructed, and the attribute item tree includes several attribute paths; Using the attribute path, several attribute item sets corresponding to the attribute item tree are determined, and the attribute item sets include user attribute combinations and job attribute combinations; Obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence level corresponding to each attribute item set based on the third support and the fourth support. The attribute item set with a confidence level greater than or equal to a preset confidence threshold is used as the group relationship data in the job search association data corresponding to the user attribute information and the job attribute information; In response to a job request instruction, the system obtains job information relevant to the job request instruction based on the group relationship data and recommends the job information.

2. The method according to claim 1, characterized in that, The step of generating job-seeking association data corresponding to the user attribute information and the target job attribute information based on the user attribute information and the target job includes: Obtain the job attribute information of the target job; Using the user attribute information and the job attribute information, job search association data corresponding to the target job is generated.

3. The method according to claim 1, characterized in that, The step of constructing an attribute tree corresponding to the job search data based on the first support of the user attribute item and the second support of the job attribute item includes: User attribute items with a first support level less than a preset support threshold and job attribute items with a second support level less than the preset support threshold are used as the first attribute items; User attribute items with a first support level greater than or equal to the preset support threshold and job attribute items with a second support level greater than or equal to the preset support threshold are used as second attribute items; By linking the first attribute item with the second attribute item, an attribute item tree corresponding to the job application data is constructed.

4. The method according to claim 3, characterized in that, The step of linking the first attribute item with the second attribute item to construct an attribute item tree corresponding to the job application data includes: Based on the support level of each of the second attribute items, construct an attribute item form corresponding to the second attribute item; Remove each of the first attribute items from the job-related data to which they belong, and sort the second attribute items in the same job-related data according to the size of their support to generate target data corresponding to the job-related data. Using the attribute item form and the target data, construct an attribute item tree for the job application related data.

5. The method according to claim 4, characterized in that, The step of constructing an attribute tree for the job-related data using the attribute item form and the target data includes: According to the sorting order of each second attribute item in the target data, the corresponding second attribute items are extracted from the attribute item form as attribute nodes in sequence to construct the attribute path corresponding to the target data; Using the attribute paths corresponding to the target data, construct an attribute item tree for the job-related data.

6. The method according to claim 5, characterized in that, The step of constructing an attribute tree for the job-related data using the attribute item form and the target data further includes: If there are duplicate attribute nodes, increment the count of the corresponding attribute node by one; If no identical attribute node exists, a new attribute node is created.

7. The method according to claim 1, characterized in that, The attribute path includes attribute nodes, and the step of using the attribute path to determine several attribute item sets corresponding to the attribute item tree includes: Obtain the condition information corresponding to the attribute node in each attribute path; The conditional information is used to perform data mining on the attribute paths to generate several attribute item sets corresponding to each attribute path.

8. The method according to claim 1, characterized in that, The step of calculating the confidence score corresponding to each attribute itemset based on the third support and the fourth support includes: The target support corresponding to the attribute item set is calculated by using the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination. The confidence level corresponding to each attribute item set is calculated using the third support corresponding to the combination of the target support and the user attribute.

9. The method according to claim 1, characterized in that, In response to a job request instruction, the method of obtaining job information for the job request instruction based on the group relationship data and recommending the job information includes: In response to a job request instruction, obtain the user category corresponding to the job request instruction; If the user category is represented as a historical user, then based on the group relationship data and the user behavior data, the first job information for the job request instruction is obtained, and the first job information is recommended; If the user category table is a new user, then based on the group relationship data and the user attribute information corresponding to the new user, the second job information for the job request instruction is obtained, and the second job information is recommended.

10. A job information recommendation device, characterized in that, include: The user data acquisition module is used to acquire user data from different users. The user data includes user behavior data, user attribute information corresponding to the user behavior data, and target job corresponding to the user behavior data. The job search association data generation module is used to generate job search association data corresponding to the user attribute information and the target job based on the user attribute information and the target job; the user attribute information includes different user attribute items, and the job attribute information includes different job attribute items; The support acquisition submodule is used to acquire the first support of each of the user attribute items in the job search association data, and the second support of each of the job attribute items in the job search association data. The attribute item tree generation submodule is used to construct an attribute item tree corresponding to the job application data based on the first support of the user attribute item and the second support of the job attribute item. The attribute item tree includes several attribute paths. The attribute item set determination submodule is used to determine several attribute item sets corresponding to the attribute item tree using the attribute path. The attribute item sets include user attribute combinations and job attribute combinations. The confidence calculation submodule is used to obtain the third support corresponding to the user attribute combination and the fourth support corresponding to the job attribute combination, and calculate the confidence of each attribute item set based on the third support and the fourth support. The group relationship data generation submodule is used to take the set of attribute items with a confidence level greater than or equal to a preset confidence threshold as the group relationship data in the job application association data that corresponds to the user attribute information and the job attribute information. The job information recommendation module is used to respond to a job request instruction, obtain job information related to the job request instruction based on the group relationship data, and recommend the job information.

11. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-9.

12. A computer-readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-9.

Images