Pushing method, computing device and computer readable storage medium

Abstract

The invention discloses a pushing method, computing equipment and a computer readable storage medium. The method comprises the following steps: acquiring preset information of a target employee; the preset information comprises staff static data and staff dynamic data; analyzing preset information of the target employee to obtain a portrait label of the target employee; and when it is determined that the portrait label of the target employee is matched with the target training information, pushing the target training information to the target employee. Therefore, the training information can be pushed more accurately, the personalized requirements of training information pushing are improved, and the participation degree of employees is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of data recommendation, and in particular to a push method, a computing device and a computer-readable storage medium. Background Art

[0002] In today's highly competitive business environment, the competitiveness of an enterprise depends largely on the quality of its workforce. Therefore, employee training is of great significance. At the same time, with the change in the growth mode of the market economy, corporate training needs to shift from extensive to intensive, which means that limited training resources should be concentrated on the most urgently needed training projects. Since the needs of each enterprise are different at different times, and the backgrounds, experiences and learning styles of employees are different, enterprises must make personalized recommendations for employee training to ensure the effectiveness of training.

[0003] However, existing methods may not be fully adapted to the specific needs of each employee, which may lead to unsatisfactory training results. Summary of the invention

[0004] The purpose of this application is to provide a push method, a computing device and a computer-readable storage medium, which can achieve more accurate training information push, improve the personalized needs of training information push, and enhance employee participation.

[0005] To achieve the above objectives:

[0006] In a first aspect, an embodiment of the present application provides a push method, including:

[0007] Acquire preset information of a target employee; the preset information includes employee static data and employee dynamic data;

[0008] Analyze the preset information of the target employee to obtain a portrait label of the target employee;

[0009] When it is determined that the portrait tag of the target employee matches the target training information, the target training information is pushed to the target employee.

[0010] Optionally, the employee static data includes employee basic data; the employee dynamic data includes at least one of the following: skill-related data, travel data; the skill-related data includes at least one of the following: employee position data, office activity data, network behavior data;

[0011] The analyzing the preset information of the target employee to obtain the portrait label of the target employee includes:

[0012] Analyze the basic data of the target employee and determine the portrait label of the target employee in the basic information dimension;

[0013] Analyze the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension;

[0014] Analyze the travel data of the target employee and determine the portrait label of the target employee in the travel situation dimension.

[0015] Optionally, the basic information dimension includes at least one of the following: name, gender, age, ethnicity, place of origin, degree, education background and technical direction;

[0016] The skill preference dimension includes at least one of the following: position held, years of employment, type of project in charge, technology stack related to the project in charge, standard document browsing records, previous training participation records, and frequently visited websites;

[0017] The travel situation dimension includes at least one of the following: travel destination and travel frequency.

[0018] Optionally, analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes:

[0019] According to the preset correspondence between different feature tags and different data, the employee position data and each data in the office activity data are matched with the feature tags to obtain the cumulative matching times corresponding to each feature tag;

[0020] A target feature tag is determined according to the accumulated matching times corresponding to each of the feature tags, and the target feature tag is determined as a portrait tag of the target employee in the skill preference dimension.

[0021] Optionally, determining the target feature tag according to the accumulated matching times corresponding to each of the feature tags includes:

[0022] Determine the weight corresponding to each of the feature tags according to the accumulated number of matches corresponding to each of the feature tags;

[0023] The feature tag whose weight is greater than a preset threshold is determined as the target feature tag.

[0024] Optionally, determining the weight corresponding to each feature tag according to the accumulated number of matches corresponding to each feature tag includes at least one of the following:

[0025] The ratio of the cumulative number of matches corresponding to each of the feature tags to the sum of the cumulative number of matches corresponding to all feature tags is used as the weight corresponding to the feature tag;

[0026] The cumulative matching times corresponding to each of the feature tags are compared with the average cumulative matching times corresponding to all the feature tags, and the initial weight corresponding to each of the feature tags is adjusted according to the comparison result to obtain the weight corresponding to each of the feature tags.

[0027] Optionally, analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes:

[0028] According to the network behavior data, the proportion of the target employee's browsing volume on different website topics is obtained;

[0029] A portrait label of the target employee regarding the frequently visited website information in the skill preference dimension is determined according to the proportion and the website theme.

[0030] Optionally, determining that the portrait tag of the target employee matches the target training information includes:

[0031] Determining a target portrait label of the target employee according to the weights corresponding to the portrait labels of the target employee in different dimensions;

[0032] Obtaining the similarity between the keywords of the target training information and the target profile label of the target employee;

[0033] When the similarity meets a preset condition, it is determined that the portrait label of the target employee matches the target training information.

[0034] Optionally, the method further comprises:

[0035] The weight corresponding to the target profile label of the target employee is adjusted according to the feedback obtained.

[0036] In a second aspect, an embodiment of the present application provides a computing device, including: a processor and a memory storing a computer program, and when the processor runs the computer program, the above-mentioned push method is implemented.

[0037] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the above-mentioned push method is implemented.

[0038] The push method, computing device and computer-readable storage medium provided by the embodiment of the present application include: obtaining preset information of the target employee; the preset information includes employee static data and employee dynamic data; analyzing the preset information of the target employee to obtain the portrait tag of the target employee; and when it is determined that the portrait tag of the target employee matches the target training information, pushing the target training information to the target employee. In this way, the portrait tag of the target employee is obtained through the preset information of the target employee, and when the portrait tag of the target employee matches the target training information, the target training information is pushed to the target employee, which can realize more accurate training information push, improve the personalized demand for training information push, and enhance employee participation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of a flow chart of a push method provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of the structure of a push system provided by an embodiment of the present invention;

[0041] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0043] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0044] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of..." or "when..." or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising", "including" indicate that there are described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, “A, B, or C” or “A, B and / or C” means “any of the following: A; B; C; A and B; A and C; B and C; A, B, and C.” An exception to this definition will occur only when a combination of elements, functions, steps, or operations are inherently mutually exclusive in some manner.

[0045] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0046] It should be noted that, in this article, step codes such as S101, S102, etc. are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the sequence. When implementing the step, those skilled in the art may execute S102 first and then S101, etc., but these should all be within the scope of protection of this application.

[0047] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0049] See also Figure 1 , is a push method provided in an embodiment of the present application. The push method can be executed by a push device provided in an embodiment of the present application. The push device can be implemented in software and / or hardware, such as a computing device such as a computer or a server. In this embodiment, the execution subject of the push method is a computing device as an example. The push method provided in this embodiment includes:

[0050] Step S101: Acquire preset information of a target employee; the preset information includes employee static data and employee dynamic data.

[0051] The target employee refers to any employee to be determined whether to push training information. Employee static data is used to indicate employee data that changes slowly and is relatively stable, such as employee basic data. Employee dynamic data is used to indicate user data that changes and updates in real time, such as travel data, network behavior data, etc. Different data can be read from different systems. For example, employee static data such as employee basic data can be read from the human resources system, and employee dynamic data such as travel data can be read from the office system.

[0052] Step S102: Analyze the preset information of the target employee to obtain the portrait label of the target employee.

[0053] Among them, the portrait label of the target employee can be used to identify the identity and bias information of the target employee, which can be obtained by analyzing the static data and dynamic data of the target employee.

[0054] In one embodiment, the employee static data includes basic employee data; the employee dynamic data includes at least one of the following: skill-related data, travel data; the skill-related data includes at least one of the following: employee position data, office activity data, network behavior data; analyzing the preset information of the target employee to obtain the target employee's portrait label includes:

[0055] Analyze the basic data of the target employees and determine the portrait labels of the target employees in the basic information dimension;

[0056] Analyze the target employee's skill-related data and determine the target employee's profile label in terms of skill preference;

[0057] Analyze the target employees’ travel data and determine their profile labels in terms of travel status.

[0058] Among them, employee basic data may include at least one of the following: name, gender, age, ethnicity, place of origin, degree, and major; employee position data may include at least one of the following: position and years of employment; office activity data may include at least one of the following: type of project in charge, technology stack related to the project in charge, standard document browsing records, and past training participation records; network behavior data may include network access behavior data; travel data may include at least one of the following: travel destination, travel time or duration.

[0059] Among them, since the employee basic data of the target employee contains the basic information of the target employee, the target employee's portrait label in the basic information dimension can be determined by analyzing the target employee's employee basic data. Here, the basic information dimension includes at least one of the following: name, gender, age, ethnicity, place of origin, degree and technical direction. At the same time, since the employee position data, office activity data and network behavior data of the target employee can reflect the skill preference of the target employee, the target employee's portrait label in the skill preference dimension can be determined by analyzing the employee position data, office activity data and network behavior data. Here, the skill preference dimension includes at least one of the following: position, years of employment, type of project responsible, technology stack related to the project responsible, standard document browsing records, past training participation records and frequently visited website information. In addition, since the target employee's travel data can reflect the countries that the target employee often visits, and thus can reflect the target user's learning needs for language types, the target employee's portrait label in the travel situation dimension can be determined by analyzing the travel data. Here, the travel situation dimension includes at least one of the following: travel destination and travel frequency. In this way, by analyzing the different data of the target employees, the portrait labels of the target employees in different dimensions are determined, which improves the accuracy and comprehensiveness of the portrait labels of the target employees, and further improves the accuracy of the training information push.

[0060] In one embodiment, analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes:

[0061] According to the correspondence between different preset feature tags and different data, each data in the employee position data and office activity data is matched with the feature tags to obtain the cumulative matching times corresponding to each feature tag;

[0062] The target feature label is determined according to the cumulative matching times corresponding to each feature label, and the target feature label is determined as the portrait label of the target employee in the skill preference dimension.

[0063] Among them, the feature tags can be set according to the actual needs, for example, they can include management, economy, software development, AI learning, overseas policies, foreign languages, etc. In this embodiment, the correspondence between different data and different feature tags can be pre-set. For example, if the position is project manager, the corresponding feature tag can be management; if the position is technical manager, the corresponding feature tag can be management, software development; if the project-related technology stack is Java, the corresponding feature tag can be software development, etc. By matching each data in the employee position data and office activity data with each feature tag, the cumulative number of matches corresponding to each feature tag can be obtained, that is, for any feature tag, the number of data matching the feature tag. Exemplarily, if the employee position data of the target user includes the position "technical manager" and the project-related technology stack "Java", the feature tags include management and software development, and when the position is technical manager, the corresponding feature tags are management and software development, and when the project-related technology stack is Java, the corresponding feature tag is software development, then it can be determined that the cumulative number of matches corresponding to the feature tag "management" is 1, and the cumulative number of matches corresponding to the feature tag "software development" is 2. In this way, it is possible to accurately obtain the portrait labels of the target employees in the skill preference dimension, further improving the accuracy of training information push.

[0064] Wherein, the target feature tag is determined according to the cumulative number of matches corresponding to each feature tag, which can be determined by selecting the first N feature tags as the target feature tag in order from high to low according to the cumulative number of matches corresponding to each feature tag, where N is a positive integer, or it can be determined in other ways. In one embodiment, the target feature tag is determined according to the cumulative number of matches corresponding to each feature tag, including: determining the weight corresponding to each feature tag according to the cumulative number of matches corresponding to each feature tag, and determining the feature tag with a weight greater than a preset threshold as the target feature tag. Wherein, for different feature tags, the weight corresponding to the feature tag can be determined according to the cumulative number of matches corresponding to the feature tag, and optionally, the ratio of the cumulative number of matches corresponding to each feature tag to the sum of the cumulative number of matches corresponding to all feature tags is used as the weight corresponding to the feature tag. For example, the ratio of the cumulative number of matches corresponding to a feature tag to the sum of the cumulative number of matches corresponding to all feature tags can be used as the weight corresponding to the feature tag, etc. The preset threshold can be set according to actual needs. For example, it can be set to 0.3, 0.5, etc. The larger the weight corresponding to the feature label, the greater the target user's preference for the feature label, and the higher the target user's demand for training information corresponding to the feature label. The feature label with a weight greater than the preset threshold is determined as the target feature label, and the target feature label is then determined as the portrait label of the target employee in the skill preference dimension.

[0065] In addition, the weight corresponding to each feature tag can be determined according to the cumulative number of matches corresponding to each feature tag, or the cumulative number of matches corresponding to each feature tag can be compared with the average cumulative number of matches corresponding to all feature tags, and the initial weight corresponding to each feature tag can be adjusted according to the comparison result to obtain the weight corresponding to each feature tag, that is, the initial weight corresponding to each feature tag can be adjusted according to the cumulative number of matches corresponding to each feature tag to determine the weight corresponding to each feature tag. For example, assuming that the initial weight corresponding to each feature tag is 0.5, after obtaining the cumulative number of matches corresponding to each feature tag, the cumulative number of matches corresponding to each feature tag can be compared with the average cumulative number of matches. If the cumulative number of matches corresponding to the feature tag is less than the average cumulative number of matches, the initial weight corresponding to the feature tag can be reduced to determine the weight corresponding to the feature tag; if the cumulative number of matches corresponding to the feature tag is greater than the average cumulative number of matches, the initial weight corresponding to the feature tag can be increased to determine the weight corresponding to the feature tag, and the magnitude of the reduction or increase can be determined based on the difference between the cumulative number of matches corresponding to the feature tag and the average cumulative number of matches. In this way, the weight corresponding to each feature tag is determined by the cumulative number of matches corresponding to each feature tag, and then the feature tag with a weight greater than a preset threshold is determined as the target feature tag, which can accurately obtain the portrait tag of the target employee and further improve the accuracy of training information push.

[0066] In one embodiment, analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes:

[0067] Based on network behavior data, obtain the percentage of target employees’ browsing on different website topics;

[0068] Determine the profile labels of target employees regarding frequently visited websites in the skill preference dimension based on the percentage and website topics.

[0069] Among them, website topics include but are not limited to C language, Java, AI models, sensors, intelligent driving, etc. By analyzing the network behavior data of the target employees, it is possible to know the target employees' browsing volume, that is, the number of browsing times, on different website topics, and then according to the target employees' browsing volume on different website topics, the proportion of the target employees' browsing volume on different website topics can be obtained. Determine the portrait label of the target employee's frequently visited website information in the skill preference dimension based on the proportion and the website topic. It can be that the website topic corresponding to the largest proportion is determined as the portrait label of the target employee's frequently visited website information in the skill preference dimension, or it can be that the website topics corresponding to the top N largest proportions are determined as the portrait label of the target employee's frequently visited website information in the skill preference dimension. No specific limitation is made here. In this way, it is possible to accurately obtain the portrait label of the target employee's frequently visited website information in the skill preference dimension, further improving the accuracy of training information push.

[0070] Step S103: When it is determined that the portrait tag of the target employee matches the target training information, the target training information is pushed to the target employee.

[0071] When it is determined that the target employee's portrait tag matches the target training information, it means that the target training information may be what the target employee needs, so the target training information can be pushed to the target employee to facilitate the target employee to know the target training information. The target training information can specifically be management training, overseas trade policy training, website usage training, etc., which is not specifically limited here.

[0072] In one embodiment, determining whether the portrait tag of the target employee matches the target training information includes: determining the target portrait tag of the target employee based on the weights corresponding to the portrait tags of the target employee in different dimensions; obtaining the similarity between the keywords of the target training information and the target portrait tag of the target employee; and determining that the portrait tag of the target employee matches the target training information when the similarity meets a preset condition.

[0073] Among them, the target training information may include one or more keywords, and the keywords may specifically be the name, type, subject word, etc. of the training. It is understandable that since the training needs of the target employees may change from time to time or focus on certain types of training information, the target employee's portrait tags in all different dimensions may not accurately recommend the target employee the training information he needs. Therefore, the target portrait tag of the target employee can be determined from the portrait tags of different dimensions according to the weights corresponding to the portrait tags of the target employee in different dimensions. The target portrait tag of the target employee is the current main or important portrait tag of the target employee. Then, after the keywords of the target training information are extracted, the similarity between each keyword of the target training information and the target portrait tag of the target employee can be calculated, and when the similarity meets the preset conditions, it is determined that the portrait tag of the target employee matches the target training information. Among them, the similarity meeting the preset conditions can be that the similarity between a keyword and the target portrait tag of the target employee is greater than the preset similarity threshold, or the similarity between multiple keywords and the target portrait tags of the target employee is greater than the preset similarity threshold, etc. Here, the initial weights corresponding to the portrait tags of the target employee in different dimensions can all be the same value, and adaptive adjustments are made based on the push opinions fed back by the target employees in the subsequent period. In this way, it is possible to accurately determine whether the target employee's portrait label matches the target training information, further improving the accuracy of training information push.

[0074] In summary, in the push method provided in the above embodiment, the portrait tag of the target employee is obtained through the preset information of the target employee, and when the portrait tag of the target employee matches the target training information, the target training information is pushed to the target employee, which can achieve more accurate training information push, improve the personalized needs of training information push, and enhance employee participation.

[0075] In one embodiment, the method further comprises:

[0076] Adjust the weights corresponding to the target employee’s target portrait labels based on the feedback obtained.

[0077] Wherein, the feedback is the satisfaction of the target employee with respect to the pushed target training information. When the feedback is that the target user is satisfied with the pushed target training information, the weight corresponding to the target portrait tag can be increased so that the training information required by the target employee can be pushed based on the target portrait tag of the target employee in the future. When the feedback is that the target user is dissatisfied with the pushed target training information, the weight corresponding to the target portrait tag can be reduced so that the training information is subsequently reduced or no longer pushed to the target employee based on the target portrait tag of the target employee. Wherein, the feedback may include a satisfaction score. When the feedback includes a satisfaction score, the weight corresponding to the target portrait tag of the target employee can be adjusted according to the satisfaction score. For example, when the satisfaction score is the highest score of 10 points, the weight corresponding to the target portrait tag can be increased by 0.1, when the satisfaction score is the lowest score of 1 point, the weight corresponding to the target portrait tag can be reduced by 0.1, and when the satisfaction score is 5 points, the weight corresponding to the target portrait tag can be reduced by 0.02, etc. In this way, by adjusting the weight corresponding to the target portrait tag of the target employee according to the feedback, the portrait tag of the target employee can be made more accurate, further improving the personalized demand for training information push.

[0078] Based on the same inventive concept as the above-mentioned embodiment, a specific example is given below to illustrate.

[0079] See also Figure 2 , which is a push system provided in this embodiment, can implement the corresponding push method, mainly including a data collection and processing module, a user portrait generation module, a training information accurate push module, and an employee satisfaction feedback module. The functions of each module are as follows:

[0080] 1) Data acquisition and processing module

[0081] Employee data is extracted by the big data platform from various relevant systems of the enterprise (such as human resources system, office system, technical platform, operation and maintenance system, etc.). As shown in Table 1, the data types are mainly divided into two categories: one is static data, that is, data that changes slowly and is relatively stable; the other is dynamic data, that is, information data that changes and updates in real time. Static data includes some basic information of employees, and dynamic data mainly includes office activity data, travel data, and network behavior data. The operation of updating the portrait is also mainly for this type of data.

[0082] Table 1

[0083]

[0084]

[0085] There are three main types of employee data: structured, semi-structured, and unstructured. Structured data is relatively easy to collect, such as basic employee information, office activity data, and travel data. This type of data only needs simple cleaning and mining to form labels. However, semi-structured and unstructured data require relatively more time for early preprocessing, such as network behavior data and standard document access data. Most of their original data have problems such as missing items, small values, data anomalies, redundancy, and noise. In order to avoid these problems from confusing user portraits, the data must be preprocessed in three ways: extraction, cleaning, and standardization to ensure the accuracy and effectiveness of the data.

[0086] 2) User portrait generation module

[0087] Based on the analysis of data collected by the big data platform and the needs of employee user portraits, a user portrait dimension system is set up, which consists of three dimensions: basic information, skill preference, and travel status. The basic information dimension includes name, gender, age, ethnicity, place of origin, degree and technical direction; the skill preference dimension includes position, years of employment, type of project responsible, technology stack related to the project responsible, standard document browsing records, previous training participation records and frequently visited website information; the travel status dimension includes travel destination and travel frequency.

[0088] User tags are divided into two categories: static tags and dynamic tags. Static tags are the most basic tags, which are formed by static data. This type of static tags is directly written into the user portrait, and this type of tag is the basic tag of the user portrait. For the static data of employees, the ontology modeling method is adopted to build a qualitative user portrait, and the portrait output form is the keyword method. Dynamic tags are formed based on dynamic data. Dynamic tags are divided into statistical dynamic tags and mining dynamic tags according to different formation methods. For statistical dynamic data, according to different modeling methods, it is divided into statistics of basic data and statistics of network log data.

[0089] For the statistics of basic data, this type of data mainly includes positions, project types, related technology stacks, skill preferences, travel conditions, etc. A modeling method that integrates the skill preferences required by users is adopted, and a vector space model is used for representation, namely ID{(α1, β1), (α2, β2), …(αn, βn)}, as shown in Table 2. Among them, αn represents the feature label, and βn represents the weight of αn in the entire model, that is, the degree of user preference for αn. The larger βn is, the higher the user's demand for this skill is. When the value exceeds the threshold, it is added to the portrait. Among them, βn is obtained by counting the number of occurrences of the corresponding feature label αn, and the corresponding relationship between the statistical data item and the feature label is obtained through the matching table.

[0090] Table 2

[0091]

[0092] For the statistics of network log classification, the concept of proportion is introduced on the basis of basic data statistics. According to the analysis of network log data, each employee will generate the corresponding skill website visit data. The data can be expressed as: Staff{(t1, p1, n1), (t2, p2, n2), …, (tm, pm, nm)} (m<11), where tm represents the mth website theme, pm represents the proportion of employee visits to the mth website theme in the total website, and nm represents the number of times an employee visits the mth website theme. After that, the category labels with the top N' (threshold N") in terms of proportion can be added to the portrait. For mining dynamic data, the first part is basic data mining. This type of data mainly includes training and lecture information that employees have participated in. For this type of data, the TF-IDF method, i.e., the term frequency-inverse document frequency calculation method, is generally used to mark the obtained dynamic labels into the user portrait.

[0093] After cleaning, integrating and analyzing the data, labels are generated and added to the target employee's portrait model. A set of labels is usually used to describe an employee. These labels describe the overall characteristics of the employee from multiple dimensions and different levels, and finally generate multi-level portrait labels, which mainly include three aspects, namely basic information, skill preference and travel status. After forming a user portrait, the needs of employees can be predicted based on it, and the accurate recommendation service of training lecture information can be realized through similarity calculation.

[0094] 3) Training information push module

[0095] Training material push is the process of matching employee skill needs with training materials and lectures. Therefore, the attribute features of the training need to be processed into vectors for easy calculation. The processing method is similar to the generation of portraits. The feature extraction of training includes the name of the training, category, basic information of the lecturer (if any), the lecturer's professional direction, the hosting college (if any), and keywords (3 to 10). When generating keywords, it is necessary to consider matching with the constructed user portraits. For example, according to the training content, the keywords are set as "management", "science and innovation" (matching with network behavior), "foreign language" (matching with travel data), etc.

[0096] Use the big data platform to build user portraits and extract training features, then use relevant recommendation algorithms to find the target group for training, find this group among the employee group, and send the training information to relevant employees. Specifically, the existing content-based recommendation algorithm can be used.

[0097] The content-based recommendation algorithm can unify the relevant information of the training into a set of standardized descriptive statistics, and then calculate whether the training is the skill direction required by a certain employee. If so, it will be pushed to the employee. The specific implementation process can use semantic similarity based on HowNet. The keywords in the user portrait and the keywords of the upcoming training will be compared according to the HowNet similarity. If there are the same keywords, they will be pushed directly; if there are no identical keywords, but the semantic similarity of multiple keywords reaches or exceeds the threshold, the training information will also be pushed to the employee; if the threshold is not reached, the training information will not be pushed.

[0098] 4) Employee feedback module

[0099] As time goes by, the skill stack required by employees will continue to change. In order to better recommend training materials based on the current status of employees, it is necessary to consider employees' feedback on the push of training information. For static data weights, the questionnaire form can be used for fine-tuning. Based on feedback, dynamic data will make full use of the window sliding mechanism to regularly add or update label weights. Each user portrait dynamic data label has a corresponding weight W (W is used to calculate the matching degree between employee skill requirements and training materials). The interval for portrait updates is set to T. The dynamic label weight changes of employees are calculated, and additions and deletions are made according to the changes, and the updated label data is stored in the database. In specific practice, T can be selected as 1 month, that is, the training needs of employees are calculated once at the beginning of each month.

[0100] In summary, in the push method provided in the above embodiment, by extracting data from the three dimensions of employee basic information, skill preferences, and travel status for analysis, and generating user portraits, more accurate training information push is achieved, which facilitates the careful development and design of training courses that closely match the actual situation of the enterprise, and can also be continuously optimized based on feedback.

[0101] Based on the same inventive concept as the above embodiments, an embodiment of the present invention provides a computing device, such as Figure 3 As shown, the computing device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 3 The processor 310 shown in the figure is not used to indicate that the number of the processor 310 is one, but is only used to indicate the positional relationship of the processor 310 relative to other devices. In actual applications, the number of the processor 310 may be one or more; similarly, Figure 3 The memory 311 shown in the figure has the same meaning, that is, it is only used to refer to the position relationship of the memory 311 relative to other devices. In practical applications, the number of memories 311 can be one or more. When the processor 310 runs the computer program, the above-mentioned push method is implemented.

[0102] The computing device may also include: at least one network interface 312. The various components in the computing device are coupled together via a bus system 313. It is understood that the bus system 313 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the bus system 313 is not used in the example above. Figure 3 Various buses are labeled as bus system 313.

[0103] The memory 311 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. The nonvolatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAM bus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 311 described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable types of memories.

[0104] The memory 311 in the embodiment of the present invention is used to store various types of data to support the operation of the computing device. Examples of these data include: any computer program used to operate on the computing device, such as an operating system and an application; contact data; phone book data; messages; pictures; videos, etc. Among them, the operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application program may include various applications, such as a media player (Media Player), a browser (Browser), etc., which are used to implement various application services. Here, the program that implements the method of the embodiment of the present invention may be included in the application program.

[0105] Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a computer storage medium, in which a computer program is stored. The computer storage medium may be a ferromagnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM) or the like; it may also be various devices including one or any combination of the above-mentioned memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc. When the computer program stored in the computer storage medium is run by the processor, the above-mentioned push method is implemented. For the specific steps implemented when the computer program is executed by the processor, please refer to Figure 1 The description of the illustrated embodiment will not be repeated here.

[0106] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] In this document, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than those listed and may also include additional elements not expressly listed.

[0108] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A push method, characterized in that: The method comprises: Acquire preset information of a target employee; the preset information includes employee static data and employee dynamic data; Analyze the preset information of the target employee to obtain a portrait label of the target employee; When it is determined that the portrait tag of the target employee matches the target training information, the target training information is pushed to the target employee.

2. The method according to claim 1, characterized in that The employee static data includes employee basic data; the employee dynamic data includes at least one of the following: skill-related data, travel data; the skill-related data includes at least one of the following: employee position data, office activity data, network behavior data; The analyzing the preset information of the target employee to obtain the portrait label of the target employee includes: Analyze the basic data of the target employee and determine the portrait label of the target employee in the basic information dimension; Analyze the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension; Analyze the travel data of the target employee and determine the portrait label of the target employee in the travel situation dimension.

3. The method according to claim 2, characterized in that The basic information dimensions include at least one of the following: name, gender, age, ethnicity, place of origin, degree, education background and technical direction; The skill preference dimension includes at least one of the following: position held, years of employment, type of project in charge, technology stack related to the project in charge, standard document browsing records, previous training participation records, and frequently visited websites; The travel situation dimension includes at least one of the following: travel destination and travel frequency.

4. The method according to claim 3, characterized in that The step of analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes: According to the preset correspondence between different feature tags and different data, the employee position data and each data in the office activity data are matched with the feature tags to obtain the cumulative matching times corresponding to each feature tag; A target feature tag is determined according to the accumulated matching times corresponding to each of the feature tags, and the target feature tag is determined as a portrait tag of the target employee in the skill preference dimension.

5. The method according to claim 4, characterized in that The step of determining the target feature tag according to the accumulated matching times corresponding to each of the feature tags includes: Determine the weight corresponding to each of the feature tags according to the accumulated number of matches corresponding to each of the feature tags; The feature tag whose weight is greater than a preset threshold is determined as the target feature tag.

6. The method according to claim 5, characterized in that The step of determining the weight corresponding to each of the feature tags according to the accumulated matching times corresponding to each of the feature tags includes at least one of the following: The ratio of the cumulative number of matches corresponding to each of the feature tags to the sum of the cumulative number of matches corresponding to all feature tags is used as the weight corresponding to the feature tag; The cumulative matching times corresponding to each of the feature tags are compared with the average cumulative matching times corresponding to all the feature tags, and the initial weight corresponding to each of the feature tags is adjusted according to the comparison result to obtain the weight corresponding to each of the feature tags.

7. The method according to claim 3, characterized in that The analyzing the skill-related data of the target employee to determine the portrait label of the target employee in the skill preference dimension includes: According to the network behavior data, obtaining the percentage of the target employee's browsing volume on different website topics; A portrait label of the target employee regarding the frequently visited website information in the skill preference dimension is determined according to the proportion and the website theme.

8. The method according to any one of claims 2 to 7, characterized in that The determining that the portrait label of the target employee matches the target training information includes: Determining a target portrait label of the target employee according to the weights corresponding to the portrait labels of the target employee in different dimensions; Obtaining the similarity between the keywords of the target training information and the target profile label of the target employee; When the similarity meets a preset condition, it is determined that the portrait label of the target employee matches the target training information.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the processor runs the computer program, the push method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the push method according to any one of claims 1 to 8 is implemented.

Images