Enterprise recruitment propaganda map generation method and device, equipment, medium and product

Abstract

The invention discloses an enterprise recruitment propaganda map generation method and device, equipment, a medium and a product. The method comprises the steps of obtaining a target recruitment text of a target enterprise, and obtaining a target propaganda map style description text matched with the target enterprise according to a pre-constructed enterprise propaganda map description database; combining the target recruitment text and the target publicity map style description text to obtain text input data, and inputting the text input data into a pre-trained publicity map format generation model to obtain publicity map style data in a JSON format; and inputting the propaganda graph style data into a pre-constructed rendering component, generating an enterprise recruitment propaganda graph for the target recruitment text, and publishing the enterprise recruitment propaganda graph on a preset information publishing platform. According to the technical scheme of the embodiment of the invention, a new mode for automatically generating the enterprise recruitment propaganda map is provided, the image generation quality is ensured, the labor cost is reduced, and the information release efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, in particular to artificial intelligence technology, and specifically relates to a method, device, equipment, medium and product for generating an enterprise recruitment promotion map. Background Art

[0002] With the continuous development of mobile Internet technology and the emergence of various employment methods, there are job-seeking application programs on the market that specifically provide recruitment information for mobile Internet workers. When a worker registers and becomes a member of the job-seeking application program, an online broker will be assigned to the member.

[0003] In the related art, after learning the recruitment information of an enterprise, in addition to being able to recommend the recruitment information to members individually, the broker can also manually draw a recruitment promotion map based on the recruitment information and publish the recruitment promotion map on the personal information publishing platform for all members served by him / her to view together.

[0004] The implementation method of manually drawing and publishing the recruitment promotion map by the broker requires the broker to have a certain foundation in using image design software and requires a lot of time and energy. In addition, due to the differences in the design level and aesthetic concepts of brokers, the quality of the recruitment promotion maps is uneven. Summary of the Invention

[0005] Embodiments of the present invention provide a method, device, equipment, medium and product for generating an enterprise recruitment promotion map, so as to provide a new way of automatically generating an enterprise recruitment promotion map, while ensuring the quality of image generation, greatly reducing the labor cost, and effectively improving the information publishing efficiency.

[0006] According to one aspect of the embodiments of the present invention, there is provided a method for generating an enterprise recruitment promotion map, which is executed by a broker client, including:

[0007] Obtain the target recruitment text of the target enterprise, and obtain the target promotion map style description text matching the target enterprise according to the pre-constructed enterprise promotion map description database;

[0008] Combine the target recruitment text and the target promotion map style description text to obtain text input data, and input the text input data into the pre-trained promotion map format generation model to obtain the promotion map style data in JSON format;

[0009] Input the promotion map style data into the pre-constructed rendering component to generate an enterprise recruitment promotion map for the target recruitment text, and publish the enterprise recruitment promotion map on the preset information publishing platform.

[0010] According to another aspect of the embodiments of the present invention, there is also provided a generating device for enterprise recruitment promotion pictures, which is configured in a broker client and includes:

[0011] A style description text acquisition module, configured to acquire target recruitment text of a target enterprise, and according to a pre-constructed enterprise promotion picture description database, acquire target promotion picture style description text matching the target enterprise;

[0012] A style data acquisition module, configured to combine the target recruitment text and the target promotion picture style description text to obtain text input data, and input the text input data into a pre-trained promotion picture format generation model to obtain promotion picture style data in JSON format;

[0013] A promotion picture publishing module, configured to input the promotion picture style data into a pre-constructed rendering component to generate an enterprise recruitment promotion picture for the target recruitment text, and publish the enterprise recruitment promotion picture on a preset information publishing platform.

[0014] According to another aspect of the embodiments of the present invention, there is also provided an electronic device, where the electronic device includes:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the method for generating an enterprise recruitment promotion picture according to any embodiment of the present invention.

[0018] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and when the computer instructions are used by a processor, the method for generating an enterprise recruitment promotion picture according to any embodiment of the present invention is implemented.

[0019] According to another aspect of the embodiments of the present invention, there is also provided a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method for generating an enterprise recruitment promotion picture according to any embodiment of the present invention are implemented.

[0020] In the technical solution of the embodiment of the present invention, by automatically executing on the broker side to obtain the target recruitment text of the target enterprise, and according to the pre-constructed enterprise publicity map description database, obtaining the target publicity map style description text matching the target enterprise; combining the target recruitment text and the target publicity map style description text to obtain text input data, and inputting the text input data into the pre-trained publicity map format generation model to obtain publicity map style data in JSON format; inputting the publicity map style data into the pre-constructed rendering component to generate an enterprise recruitment publicity map for the target recruitment text, and publishing the enterprise recruitment publicity map on the preset information release platform and other operations, it is possible to rely on the high-quality image generation advantage of the artificial intelligence model to automatically generate enterprise recruitment publicity maps that meet the enterprise characteristics and styles for each recruitment information for publication on the premise of introducing the minimum labor cost, greatly reducing the labor cost and effectively improving the information release efficiency.

[0021] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0023] Figure 1 is a flowchart of a method for generating an enterprise recruitment publicity map according to an embodiment of the present invention;

[0024] Figure 2 is a flowchart of another method for generating an enterprise recruitment publicity map according to an embodiment of the present invention;

[0025] Figure 3 is a schematic structural diagram of an apparatus for generating an enterprise recruitment publicity map according to an embodiment of the present invention;

[0026] Figure 4 is a schematic structural diagram of an electronic device for implementing the method for generating an enterprise recruitment publicity map according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0029] Figure 1 The figure is a flowchart of a method for generating an enterprise recruitment promotion map provided by an embodiment of the present invention. This embodiment is applicable to the situation where, on the broker side, a matching enterprise recruitment promotion map is automatically generated according to the recruitment text and automatically published on the platform. This method can be executed by a device for generating an enterprise recruitment promotion map, and this device can be implemented in the form of hardware and / or software, and is generally configured in various terminal devices, such as mobile phones, tablets, PCs, or laptops, etc. Among them, the terminal device is pre-installed with a job hunting application program, which is used for users in the broker role to log in.

[0030] Correspondingly, as Figure 1 shown, the method may include:

[0031] S110. Obtain the target recruitment text of the target enterprise, and obtain the target promotion map style description text matching the target enterprise according to the pre-constructed enterprise promotion map description database.

[0032] Among them, the target recruitment text of the target enterprise can be understood as the recruitment information in text form that the target enterprise publishes in real time through the enterprise official account on the entire network or the set network channels. The target recruitment text is generally structured data in the form of "recruitment attribute - specific attribute value", for example: "Work location: XXX, recruitment position: XXX, wage rate under the pay cycle: XXX, ……", etc.

[0033] In the enterprise promotion map description database, description texts of the promotion map styles of different enterprises are pre-stored. The description text of the promotion map style can be understood as description information about the basic requirements and styles of the enterprise recruitment promotion map to be generated for a specific enterprise. Generally, the description text of the promotion map style can be structured data in the form of "style attribute - specific attribute value".

[0034] Among them, the description text of the promotion map style can include information describing various elements related to the display effect of the promotion map, such as: "background map type: XXX, layout method: XXX, font type: XXX", etc. Or, the description text of the promotion map style can also include information describing various elements related to the fixed display content in the promotion map, such as: whether to include the enterprise logo: "yes", warm reminder content: XXXXXX; precautions content: XXXXX, weekly salary standard: XXXXX,..." etc.

[0035] After obtaining the target recruitment text newly released by the target enterprise, according to the enterprise identifier of the target enterprise, it can be searched in the pre-constructed enterprise promotion map description database to obtain the target promotion map style description text matching the target enterprise.

[0036] Among them, the enterprise promotion map description database can be constructed by the method of each enterprise actively reporting the description text of the promotion map style, or, it can also be constructed by the method of performing artificial intelligence automatic analysis and processing on the historical enterprise recruitment promotion maps of each enterprise collected in advance, etc. This embodiment does not limit this.

[0037] In an optional implementation manner of this embodiment, before obtaining the target promotion map style description text matching the target enterprise according to the pre-constructed enterprise promotion map description database, it may further include:

[0038] Collect the historical enterprise recruitment promotion maps released by each enterprise, and perform clustering processing on each historical enterprise recruitment promotion map according to the enterprise to obtain the promotion map clustering clusters corresponding to each enterprise respectively;

[0039] Extract the image and text features of each historical enterprise recruitment promotion map included in each promotion map clustering cluster to obtain the promotion map style feature sets corresponding to each enterprise respectively;

[0040] Input the promotion map style feature sets of each enterprise into the pre-trained style description information generation model respectively to obtain the promotion map style description texts corresponding to each enterprise respectively;

[0041] Construct the enterprise promotion map description database according to the promotion map style description texts of each enterprise.

[0042] In this alternative embodiment, all the enterprise recruitment promotion pictures already released by each enterprise can be collected first. Then, through the method of automatically scoring using a pre-constructed quality assessment model or through manual evaluation, high-quality historical enterprise recruitment promotion pictures can be screened out from all the enterprise recruitment promotion pictures. Furthermore, the historical enterprise recruitment promotion pictures can be clustered by enterprise to obtain a cluster of promotion pictures corresponding to each enterprise respectively. That is to say, each cluster of promotion pictures contains various high-quality enterprise recruitment promotion pictures once released by the same enterprise.

[0043] After obtaining the above-mentioned clusters of promotion pictures, the image and text features of each historical enterprise recruitment promotion picture included in the cluster of promotion pictures for each enterprise can be extracted to obtain a set of promotion picture style features corresponding to each enterprise respectively.

[0044] Specifically, the text regions can be identified respectively in each historical enterprise recruitment promotion picture in the same cluster of promotion pictures, and according to the text content in the text regions, the text type (or style attribute) of each text region can be determined, for example, dynamic recruitment information or fixed display content of a set type, etc. Furthermore, the text features corresponding to different text types can be obtained respectively. Similarly, the image regions can be identified respectively in each historical enterprise recruitment promotion picture, and the image features of each image region under its respective type (or style attribute) can be extracted respectively. Additionally, according to the position coordinates of the text regions of each type in the same historical enterprise recruitment promotion picture, the text layout features of each historical enterprise recruitment promotion picture can be determined, etc.

[0045] After the image and text features are extracted from all the historical enterprise recruitment promotion pictures for each cluster of promotion pictures respectively, all the extracted features can be used to construct a set of promotion picture style features corresponding to each cluster of promotion pictures respectively. Furthermore, by inputting the sets of promotion picture style features of each cluster of promotion pictures into a pre-trained style description information generation model respectively, the promotion picture style description texts corresponding to each enterprise can be obtained.

[0046] Among them, a neural network model with a specific architecture can be custom-built first. Then, using the sets of standard promotion picture style features marked with standard promotion picture style description texts as training samples, this neural network model can be trained to obtain this style description information generation model. Or, existing various open-source AI models can also be directly used as pre-trained models, and the above training samples can be used as prompts (Prompts) to fine-tune this pre-trained model to obtain this style description information generation model.

[0047] Among them, the construction process of the above-mentioned enterprise publicity picture description database can be executed on the server side. Furthermore, the style description information generation model can complete the model training and call execution process on the server side. Considering the scarcity of computing power on the server side, the above-mentioned style description information generation model is a lightweight model. That is to say, if the style description information generation model is obtained by fine-tuning a pre-trained model, it is necessary to significantly reduce the model scale of the pre-trained model in advance to filter out various unnecessary functions and only ensure that the pre-trained model can accurately implement the function of generating publicity picture style description text.

[0048] In a specific example, among the publicity picture style description texts corresponding to an enterprise, the same style attribute may correspond to multiple attribute values. For example, for the style attribute of the background picture style, the corresponding attribute values include multiple ones, such as the technology style, the hand-drawn style, or the illustration style, etc. Or the fixed text content corresponding to the style attribute of the warm reminder can be: the first prompt field or the second prompt field, etc.

[0049] When the above situation occurs, it is possible to first detect whether the style attribute corresponding to the multiple attribute values belongs to a style attribute that can be selected multiple times. For example, style attributes such as the background picture style, the background picture color, the layout method, and the font type belong to style attributes that can be selected multiple times, while style attributes such as the content of precautions and the weekly salary standard belong to style attributes that cannot be selected multiple times.

[0050] For the style attributes that can be selected multiple times, all the corresponding multiple attribute values can be associated and stored in the enterprise publicity picture description database. For the style attributes that cannot be selected multiple times, it is necessary to sequentially obtain the most recently generated historical enterprise recruitment publicity picture according to the order from the most recent generation time to the oldest, and based on the recognition result of the attribute value of the style attribute that cannot be selected multiple times in the historical enterprise recruitment publicity picture, perform screening processing on the attribute values of each style attribute that cannot be selected multiple times, so as to only retain the most recently effective single attribute value for such style attributes that cannot be selected multiple times in the enterprise publicity picture description database.

[0051] Specifically, developers can pre-construct a style attribute description table and record in the style attribute description table whether each style attribute is a style attribute that can be selected multiple times.

[0052] Furthermore, when obtaining the target publicity picture style description text that matches the target enterprise, for the multiple attribute values of the above-mentioned style attributes that can be selected multiple times, one attribute value can be selected from the above multiple attribute values according to a preset screening criterion for use as the target publicity picture style description text. Specifically, the screening method can be random selection or balanced selection according to the selection frequency of each attribute value, etc. This embodiment does not limit this.

[0053] Based on the above embodiments, after obtaining the target promotional map style description text matching the target enterprise, the target promotional map style description text can be first provided to the broker for preview, and text modification options are synchronously provided for the broker to fine-tune the automatically generated target promotional map style description text to meet the personalized map generation needs of different brokers.

[0054] S120. Combine the target recruitment text and the target promotional map style description text to obtain text input data, and input the text input data into a pre-trained promotional map format generation model to obtain promotional map style data in JSON format.

[0055] In this embodiment, the promotional map style data refers to key-value pair data that quantitatively describes the enterprise recruitment promotional map to be generated in the form of key-value pairs. For example, background color, background theme color, display positions of various display texts, font sizes, and layout methods, etc.

[0056] Specifically, the promotional map format generation model is used to automatically generate promotional map style data in JSON format for generating enterprise recruitment promotional maps according to the input real-time recruitment text and promotional map style description text.

[0057] Among them, a neural network model with a specific architecture can be first custom-built, and then, using each standard text input data labeled with standard promotional map style data as training samples, the neural network model is trained to obtain the promotional map format generation model. Or existing various open-source AI models can be directly used as pre-trained models, and the above training samples are used as prompts (Prompts) to fine-tune the pre-trained model to obtain the promotional map format generation model.

[0058] Among them, the promotional map format generation model can be stored in the terminal device where the broker client is located, or can also be pre-stored in an edge device on the same local area network as the broker client. Furthermore, the broker client can immediately call the promotional map format generation model to obtain the required promotional map style data. Or, the promotional map format generation model can also be stored on the remote server side and be called and executed by the broker client through remote communication.

[0059] In this embodiment, in order to ensure the instantaneity of enterprise recruitment promotional map generation, the promotional map format generation model can be deployed in an edge device. Similarly, considering that the computing power of edge devices is more scarce, the above promotional map format generation model also needs to be a lightweight model.

[0060] Based on the above embodiments, after obtaining the promotional graphic style data in JSON format, the promotional graphic style data can be first provided to the broker for preview, and modification options for the style data can be synchronously provided, so that the broker can fine-tune the automatically generated promotional graphic style data to meet the personalized graphic output requirements of different brokers.

[0061] S130. Input the promotional graphic style data into a pre-constructed rendering component to generate an enterprise recruitment promotional graphic for the target recruitment text, and publish the enterprise recruitment promotional graphic on a preset information release platform.

[0062] In this embodiment, the inventor further optimizes and improves the existing rendering component (for example, the Canvas rendering component), so that the optimized and improved rendering component can draw an enterprise recruitment promotional graphic corresponding to the target recruitment text by calling the drawing API (Application Programming Interface) based on the input promotional graphic style data.

[0063] In this embodiment, to ensure the immediacy of the generation of the enterprise recruitment promotional graphic, the rendering component can be configured in the broker client. In the process of implementing the present invention, the inventor found that the calculation amount of the entire process of generating the enterprise recruitment promotional graphic is large and the time consumption is long. To further reduce the image generation time and improve the image generation efficiency, it is considered that the rendering component can be optimized by deleting some parts, so as to improve the rendering efficiency of the rendering component on the premise of effectively meeting the requirements for generating the enterprise recruitment promotional graphic.

[0064] Specifically, by analyzing the existing Canvas renderer, the inventor found that the Canvas renderer contains a variety of drawing styles, such as various types of colors and filling styles, line styles, gradient styles, text styles, shadow and projection degree styles, and various canvas configuration styles. These drawing styles need to be pre-loaded in the Canvas renderer and will occupy a large amount of memory space, which will greatly affect the efficiency of the broker client running the Canvas renderer to draw the enterprise recruitment promotional graphic.

[0065] In view of this, the inventor clusters the promotional graphic style data generated by the promotional graphic format generation model to obtain all the required drawing styles to which the promotional graphic style data belongs. By using the required drawing styles obtained through the above clustering operation, all the drawing styles included in the Canvas renderer are deleted and merged to perform an adaptive reduction process on the Canvas renderer, so that the Canvas renderer can effectively adapt to the application scenario of completing the entire process of generating the enterprise recruitment promotional graphic on the broker client to meet the usage requirements of immediate graphic output.

[0066] Furthermore, the inventors analyzed a large number of corporate recruitment promotion images and found that SVG (Scalable Vector Graphics) is a type of image frequently used in corporate recruitment promotion images. For example, charts, arrow keys, or various symbols, etc. In the existing Canvas renderer, it is generally necessary to pre-store a large number of prefabricated vector images to efficiently respond to real-time image rendering requirements. However, these pre-stored vector images also occupy a large amount of memory space and reduce the image rendering speed. Based on this, the inventors further improved the existing Canvas renderer and considered adding a custom vector image generation plugin to the Canvas renderer to generate the vector images required for each rendering in real time according to png images or other types of images, and then the prefabricated vector images can be deleted to the greatest extent to significantly reduce the scale of the Canvas renderer.

[0067] Specifically, the core algorithm of the above vector image generation plugin can be the coordinate mapping method, that is, map each coordinate point of the input image to the SVG coordinate system according to a preset coordinate mapping formula to efficiently generate vector images. Through the above settings, the image rendering efficiency can be further improved, so that the rendering component added to the broker client can better adapt to the actual use requirements of instantaneously generating corporate recruitment promotion images.

[0068] In this embodiment, after the broker client generates a corporate recruitment promotion image, the official account of the affiliated broker can be directly used to publish the corporate recruitment promotion image on a preset information publishing platform. Among them, when publishing information, it can be set to be visible to all people, or visible to some people (for example, members served by the broker), etc. This embodiment does not limit this.

[0069] Furthermore, after automatically generating the corporate recruitment promotion image, the corporate recruitment promotion image can be first displayed to users on the broker client, and image adjustment options (such as color schemes, text positions, etc.) can be provided to the broker for fine-tuning of the corporate recruitment promotion image to be published.

[0070] The technical solution of the embodiment of the present invention automatically executes on the broker side to obtain the target recruitment text of the target enterprise, and obtains the target promotional map style description text matching the target enterprise according to the pre-constructed enterprise promotional map description database; combines the target recruitment text and the target promotional map style description text to obtain text input data, and inputs the text input data into the pre-trained promotional map format generation model to obtain the promotional map style data in JSON format; inputs the promotional map style data into the pre-constructed rendering component to generate an enterprise recruitment promotional map for the target recruitment text, and publishes the enterprise recruitment promotional map on the preset information release platform. By doing so, on the premise of introducing the minimum labor cost, relying on the high-quality image generation advantage of the artificial intelligence model, it can automatically generate enterprise recruitment promotional maps that conform to the enterprise characteristics and styles for each recruitment information and publish them, greatly reducing the labor cost and effectively improving the information release efficiency.

[0071] Based on the above embodiments, the method may further include:

[0072] Match the target recruitment text with the historical standard recruitment information of the target enterprise to obtain the key content information in the target recruitment text;

[0073] Construct a first personalized description text according to the key content information;

[0074] Correspondingly, combining the target recruitment text and the target promotional map style description text to obtain text input data specifically includes:

[0075] Combine the target recruitment text, the target promotional map style description text, and the first personalized description text to obtain text input data.

[0076] In this alternative embodiment, the key content information can be understood as information that is significantly different from the historical standard recruitment information. For example, by formatting and structuring all the historical recruitment information of the target enterprise, the above historical standard recruitment information can be obtained. If the "wage" information provided in the historical standard recruitment information is within a preset numerical range, and the "wage" information described in the target recruitment text still has a relatively large increase compared to the maximum value in the above numerical range, then the above "wage" information can be used as a key content information. In another specific example, if the "working address" information described in the target recruitment text contains "working address" information that has never appeared in the historical standard recruitment information of the target enterprise, then the above never-appeared "working address" information can be used as a key content information.

[0077] It can be understood that, since the determined key content information has prominent changes compared with the historical standard recruitment information, furthermore, it is necessary to prominently display the above-mentioned key content information in the enterprise recruitment promotion map, so that members who view the enterprise recruitment promotion map can quickly obtain this important information and improve the information interaction efficiency.

[0078] Among them, it can be considered to prominently display the above-mentioned key content information in the enterprise recruitment promotion map. Furthermore, a first personalized description text can be constructed according to the key content information. For example, "Center (or emphasize) the 'XXXXX' information" and so on.

[0079] Specifically, one or more first personalized description text generation templates can be constructed. Furthermore, according to the information type of the key content information, a matching first personalized description text generation template can be selected to automatically generate the first personalized description text.

[0080] Based on the above embodiments, the method may further include:

[0081] Obtain the current system time and obtain time feature information according to the current system time;

[0082] Construct a second personalized description text according to the time feature information;

[0083] Correspondingly, combining the target recruitment text and the target promotion map style description text to obtain text input data, specifically including:

[0084] Combine the target recruitment text, the target promotion map style description text, and the second personalized description text to obtain text input data.

[0085] In this optional implementation manner, the time feature information can be understood as special information with time attributes, such as important events like festivals. After obtaining the time feature information, a second personalized description text can be constructed based on the time feature information. For example, if it is found that the release time of the recruitment information is close to the Spring Festival, a second personalized description text such as "Add patterns with festive elements" can be considered, so that the instantaneously generated enterprise recruitment promotion map better meets the aesthetic needs of the current environment, is more likely to arouse the interest points of members, and can improve the probability that the enterprise recruitment promotion map is comprehensively viewed by members to a certain extent, thereby improving the hit rate of the released information.

[0086] Figure 2The flowchart of another method for generating an enterprise recruitment promotion map provided by an embodiment of the present invention. This embodiment is optimized based on the above embodiments. In this embodiment, the method further includes: obtaining the job hunting tendency information of all currently maintained members, and clustering each member according to the job hunting tendency information to obtain a plurality of member clusters; constructing third personalized style description texts respectively corresponding to the job hunting tendency information of each member cluster;

[0087] In addition, the operation of "combining the target recruitment text and the target promotion map style description text to obtain text input data, and inputting the text input data into a pre-trained promotion map format generation model to obtain promotion map style data in JSON format" is specifically optimized as: "combining the target recruitment text and the target promotion map style description text to obtain original input data, and combining the original input data with the third personalized style description texts corresponding to each member cluster respectively to obtain multiple groups of text input data; inputting each group of text input data into a pre-trained promotion map format generation model respectively to obtain promotion map style data in JSON format corresponding to each member cluster respectively";

[0088] And, the operation of "publishing the enterprise recruitment promotion map on a preset information release platform" is specifically optimized as: "publishing the enterprise recruitment promotion maps corresponding to each member cluster respectively on the information release platforms matching each member cluster".

[0089] Correspondingly, as Figure 2 shown, the method may include:

[0090] S210. Obtain the target recruitment text of the target enterprise, and obtain the target promotion map style description text matching the target enterprise according to the pre-constructed enterprise promotion map description database.

[0091] S220. Obtain the job hunting tendency information of all currently maintained members, and cluster each member according to the job hunting tendency information to obtain a plurality of member clusters.

[0092] Among them, the job hunting tendency information can be understood as the factors that a member with job hunting needs will consider when choosing to join or not join an enterprise. For example, commuting distance, working location, working hours, clock-in method, hourly wage or comprehensive salary, etc.

[0093] Specifically, after a user registers and becomes a member of the job application, the job preference information of the member can be obtained by providing a questionnaire for the member to fill out. Alternatively, the characteristics of each enterprise where each member has successfully joined can be comprehensively analyzed through artificial intelligence technology, and based on the analysis results, the job preference information of each member can be determined.

[0094] After obtaining the job preference information of each of the above members, the members are clustered according to the job preference information to obtain multiple member clusters (or also called member sets). For example, a member set that focuses on the work location, a member set that focuses on the clock-in method, etc.

[0095] S230. Construct third personalized style description texts corresponding to the job preference information of each member cluster respectively.

[0096] As mentioned above, since the job preferences of different members are different, in an enterprise recruitment promotion picture, the types of information they focus on are also different. If the same enterprise recruitment promotion picture is pushed to members of different types, it may happen that because the content that a certain member focuses on is not prominently displayed in the enterprise recruitment promotion picture, the member ignores the recruitment information in the enterprise recruitment promotion picture, thus missing a job that may be very suitable for him / her.

[0097] Based on this, the inventor further proposes to construct different types of third personalized style description texts according to the different job preference information of different member clusters, and then different types of enterprise recruitment promotion pictures can be generated for personalized recommendation for different member clusters, so that each member can quickly learn the information content that they focus on in the pushed enterprise recruitment promotion picture, and then can quickly determine whether to respond to the recruitment information.

[0098] In a specific example, if a certain type of member is very concerned about the specific hourly wage, a third personalized style description text such as "Center the specific hourly wage data for display" can be constructed. Another example, if a certain type of member is very concerned about the clock-in method after joining the company, a third personalized style description text such as "Highlight the specific clock-in method" can be constructed, etc.

[0099] S240. Combine the target recruitment text and the target promotion picture style description text to obtain the original input data, and combine the original input data with the third personalized style description texts corresponding to each member cluster respectively to obtain multiple groups of text input data.

[0100] In this embodiment, the target recruitment text and the text description of the target promotional picture style can be combined first to obtain the original input data. Then, each third personalized style description text corresponding to each member cluster is combined with the same original input data respectively to obtain the text input data corresponding to each member cluster respectively.

[0101] S250. Input each group of text input data into a pre-trained promotional picture format generation model respectively to obtain the promotional picture style data in JSON format corresponding to each member cluster respectively.

[0102] S260. Input the promotional picture style data corresponding to each member cluster into a pre-constructed rendering component respectively to generate the enterprise recruitment promotional pictures corresponding to each member cluster respectively.

[0103] By inputting different text input data into the promotional picture format generation model respectively, different promotional picture style data accurately adapted to members with different job hunting tendencies can be obtained. Furthermore, by inputting the above different promotional picture style data into the pre-constructed rendering component respectively, different enterprise recruitment promotional pictures that meet the needs of members with different job hunting tendencies can be obtained.

[0104] S270. Publish the enterprise recruitment promotional pictures corresponding to each member cluster on the information release platform matched with each member cluster respectively.

[0105] In an optional implementation manner of this embodiment, all members served by the broker in the information release platform can be grouped according to the job hunting tendency information under the official account of the broker. That is, one group corresponds to one member cluster. Furthermore, the matched enterprise recruitment promotional pictures can be published respectively under each member group.

[0106] Through the above settings, according to the job hunting tendency information of different members, the information they focus on can be displayed in a very "conspicuous" way in the enterprise recruitment promotional pictures they can view. Furthermore, the member can quickly learn the specific content of the information they focus on in the enterprise recruitment promotional picture, and then can quickly decide whether to respond to the recruitment information in the enterprise recruitment promotional picture, which greatly improves the information interaction efficiency and efficiently meets the actual needs of members.

[0107] The technical solution of the embodiment of the present invention obtains the job hunting tendency information of all currently maintained members, clusters each member according to the job hunting tendency information to obtain multiple member clusters; constructs third personalized style description texts respectively corresponding to the job hunting tendency information of each member cluster; combines the target recruitment text and the target promotional picture style description text to obtain the original input data, and combines the original input data with the third personalized style description texts corresponding to each member cluster respectively to obtain multiple groups of text input data; inputs each group of text input data into a pre-trained promotional picture format generation model respectively to obtain promotional picture style data in JSON format corresponding to each member cluster respectively; and publishes the enterprise recruitment promotional pictures corresponding to each member cluster respectively on the information release platforms matching each member cluster. This implementation method can maximize the promotion effect of recruitment publicity information, greatly reduce the member information browsing time, and effectively improve the hit rate of recruitment publicity information on the actual needs of users.

[0108] In an alternative implementation manner of this embodiment, after generating the enterprise recruitment promotional picture for the target recruitment text, it may further include:

[0109] Extract the image features of the enterprise recruitment promotional picture, and obtain the color matching scheme matching the enterprise recruitment promotional picture according to the extraction result;

[0110] If it is determined that at least one color matching parameter and / or the combination of multiple color matching parameters in the color matching scheme do not meet the preset color matching restriction conditions, then adjust the color matching scheme of the enterprise recruitment promotional picture according to the preset adjustment strategy to obtain the updated enterprise recruitment promotional picture.

[0111] In this alternative implementation manner, a configuration interval of one or more color matching parameters, or a combined configuration interval of multiple sets of preset color matching parameter combinations can be pre-constructed as the color matching restriction conditions, and these color matching restriction conditions can be some classic configuration methods that conform to the public aesthetic.

[0112] When the enterprise recruitment promotional pictures automatically generated by the methods of the embodiments of the present invention do not meet one or more of the above color matching restriction conditions, it indicates that the enterprise recruitment promotional pictures may deviate from the mainstream aesthetic to a certain extent. Considering that the aesthetic levels of brokers vary, therefore, by adding an implementation method of automatically fine-tuning the color matching scheme of the enterprise recruitment promotional pictures, the finally released enterprise recruitment promotional pictures can be based on the above color matching restriction conditions and basically meet the mainstream aesthetic requirements, which to a certain extent guarantees the "bottom line" of the aesthetic quality of the enterprise recruitment promotional pictures.

[0113] It should be emphasized again that the solutions of the embodiments of the present invention can be different from the prior art in multiple dimensions:

[0114] 1. Construction and Optimization Methods of Intelligent Algorithm Models:

[0115] In the embodiments of the present invention, all AI algorithm models involved, including their construction methods, training processes, and optimization strategies, are important technological innovation points. These technological innovation points ensure that the enterprise recruitment promotion map generation solutions provided by the embodiments of the present invention can accurately understand the recruitment business requirements and generate enterprise recruitment promotion maps that meet the requirements.

[0116] 2. Multi-level Data Processing and Conversion Processes

[0117] The processing process from the business-to-text layer to the text-to-JSON layer, and then to the AI algorithm model and Canvas rendering layer used in the embodiments of the present invention is also an important innovation point. This process ensures the accurate transmission and efficient conversion of data, providing a strong guarantee for the automatic generation of enterprise recruitment promotion maps.

[0118] 3. Innovative Application of Canvas Rendering Technology

[0119] The embodiments of the present invention have made innovations in the application of Canvas rendering technology, achieving precise rendering and efficient output of the content of enterprise recruitment promotion maps. This innovative application not only improves the rendering efficiency but also enhances the visual effect and user experience of the promotion maps.

[0120] Furthermore, the technical solutions of the embodiments of the present invention can also achieve beneficial effects in multiple dimensions:

[0121] I. Improving Production Efficiency and Degree of Automation

[0122] 1. Automated Generation Process:

[0123] By introducing AI algorithm models, an automated process from recruitment business requirements to the generation of enterprise recruitment promotion maps has been achieved. This process significantly reduces manual intervention and improves production efficiency. The automated generation process not only shortens the production cycle of enterprise recruitment promotion maps but also reduces the dependence on professional designers, enabling non-professional personnel to easily complete the generation of enterprise recruitment promotion maps.

[0124] 2. Reducing Computational Load:

[0125] During the training and optimization of AI algorithm models, by continuously learning and adjusting parameters, the ability to understand and process recruitment business requirements has been gradually improved. This enables the model to find the optimal solution more quickly when generating enterprise recruitment promotion maps, reducing the computational load. At the same time, Canvas rendering technology also further reduces the computational load and improves the rendering efficiency by optimizing the rendering algorithm and reducing unnecessary rendering steps.

[0126] II. Improve the Quality and Diversity of Recruitment Promotion Graphics

[0127] 1. High-quality Output:

[0128] The AI algorithm model can intelligently select appropriate design elements and styles according to the input recruitment business requirements, and generate high-quality corporate recruitment promotion graphics. This intelligent selection not only ensures the professionalism and aesthetics of corporate recruitment promotion graphics, but also improves their attractiveness and dissemination effect.

[0129] 2. Diversified Design:

[0130] The technical solutions of the embodiments of the present invention support the combination of multiple design styles and elements, making the generated corporate recruitment promotion graphics highly diverse and flexible. This meets the different needs and preferences of different customers and markets for corporate recruitment promotion graphics.

[0131] III. Reduce Production Costs and Improve Economic Benefits

[0132] 1. Reduce Labor Costs:

[0133] The automated generation process reduces the dependence on professional designers and lowers labor costs. This enables enterprises to produce corporate recruitment promotion graphics at a lower cost and improves economic benefits.

[0134] 2. Improve Resource Utilization Efficiency:

[0135] By optimizing the AI algorithm model and Canvas rendering technology, the technical solutions of the embodiments of the present invention improve the utilization efficiency of computing resources and storage resources. This reduces resource waste and lowers production costs.

[0136] 3. Respond Quickly to Market Changes:

[0137] The automated generation process enables enterprises to quickly respond to market changes and changes in customer needs. Enterprises can adjust the content and style of corporate recruitment promotion graphics in a timely manner according to market trends and customer needs, and improve market competitiveness.

[0138] IV. Enhance User Experience and Satisfaction

[0139] 1. Personalized Customization:

[0140] The technical solutions of the embodiments of the present invention support personalized customization services, and can generate corporate recruitment promotion graphics according to the specific needs and preferences of customers. This personalized customization service enhances the user experience and satisfaction, and improves the trust and loyalty of customers to the enterprise.

[0141] 2. Quick Feedback and Adjustment:

[0142] By collecting users' feedback on the enterprise recruitment promotion pictures, the enterprise can promptly discover and adjust the problems and deficiencies of the enterprise recruitment promotion pictures. This rapid feedback and adjustment mechanism ensures that the quality and effect of the enterprise recruitment promotion pictures always meet the expectations and requirements of customers.

[0143] Figure 3 The following is a schematic structural diagram of a device for generating an enterprise recruitment promotion picture provided by an embodiment of the present invention. This device is configured in the broker client, as Figure 3 shown. The device includes: a style description text acquisition module 310, a style data acquisition module 320, and a promotion picture publishing module 330, where:

[0144] The style description text acquisition module 310 is used to acquire the target recruitment text of the target enterprise, and according to the pre-constructed enterprise promotion picture description database, acquire the target promotion picture style description text that matches the target enterprise;

[0145] The style data acquisition module 320 is used to combine the target recruitment text and the target promotion picture style description text to obtain text input data, and input the text input data into a pre-trained promotion picture format generation model to obtain promotion picture style data in JSON format;

[0146] The promotion picture publishing module 330 is used to input the promotion picture style data into a pre-constructed rendering component to generate an enterprise recruitment promotion picture for the target recruitment text, and publish the enterprise recruitment promotion picture on a preset information publishing platform.

[0147] The technical solution of the embodiment of the present invention, by automatically executing on the broker side to acquire the target recruitment text of the target enterprise, and according to the pre-constructed enterprise promotion picture description database, acquire the target promotion picture style description text that matches the target enterprise; combine the target recruitment text and the target promotion picture style description text to obtain text input data, and input the text input data into a pre-trained promotion picture format generation model to obtain promotion picture style data in JSON format; input the promotion picture style data into a pre-constructed rendering component to generate an enterprise recruitment promotion picture for the target recruitment text, and publish the enterprise recruitment promotion picture on a preset information publishing platform and other operations, can, on the premise of introducing the minimum labor cost, rely on the high-quality picture generation advantage of the artificial intelligence model to automatically generate enterprise recruitment promotion pictures that meet the enterprise characteristics and styles for each recruitment information for publication, greatly reducing the labor cost and effectively improving the information publishing efficiency.

[0148] Based on the above embodiments, it may further include an enterprise promotion picture description database construction module, which is used for:

[0149] Before describing the database of enterprise promotion pictures according to the pre-constructed one, and obtaining the target promotion picture style description text matching the target enterprise, collect the historical enterprise recruitment promotion pictures released by each enterprise, and cluster the historical enterprise recruitment promotion pictures according to the enterprises to obtain the promotion picture clustering clusters corresponding to each enterprise respectively;

[0150] Extract the image and text features of each historical enterprise recruitment promotion picture included in each promotion picture clustering cluster to obtain the promotion picture style feature sets corresponding to each enterprise respectively;

[0151] Input the promotion picture style feature set of each enterprise into the pre-trained style description information generation model respectively to obtain the promotion picture style description text corresponding to each enterprise respectively;

[0152] Construct an enterprise promotion picture description database according to the promotion picture style description text of each enterprise.

[0153] On the basis of the above embodiments, it may further include a first description text construction module for:

[0154] Match the target recruitment text with the historical standard recruitment information of the target enterprise to obtain the key content information in the target recruitment text; construct a first personalized description text according to the key content information;

[0155] Correspondingly, the style data acquisition module 320 can be further used for:

[0156] Combine the target recruitment text, the target promotion picture style description text, and the first personalized description text to obtain text input data.

[0157] On the basis of the above embodiments, it may further include a second description text construction module for:

[0158] Obtain the current system time, and obtain time feature information according to the current system time; construct a second personalized description text according to the time feature information;

[0159] Correspondingly, the style data acquisition module 320 can be further used for:

[0160] Combine the target recruitment text, the target promotion picture style description text, and the second personalized description text to obtain text input data.

[0161] On the basis of the above embodiments, it may further include a third description text construction module for:

[0162] Obtain the job hunting tendency information of all currently maintained members, and cluster each member according to the job hunting tendency information to obtain multiple member clusters; construct third personalized style description texts corresponding to the job hunting tendency information of each member cluster respectively;

[0163] The style data acquisition module 320 can be further used for:

[0164] Combine the target recruitment text and the target promotional picture style description text to obtain the original input data, and combine the original input data with the third personalized style description texts corresponding to each member cluster respectively to obtain multiple groups of text input data;

[0165] Input each group of text input data into the pre-trained promotional picture format generation model respectively to obtain the promotional picture style data in JSON format corresponding to each member cluster respectively;

[0166] Correspondingly, the promotional picture publishing module 330 can be further used for:

[0167] Publish the enterprise recruitment promotional pictures corresponding to each member cluster respectively on the information publishing platforms matching each member cluster.

[0168] Based on the above embodiments, it may further include a color scheme adjustment module for:

[0169] After generating the enterprise recruitment promotional picture for the target recruitment text, extract the image features of the enterprise recruitment promotional picture, and obtain the color scheme matching the enterprise recruitment promotional picture according to the extraction result;

[0170] If it is determined that at least one color matching parameter and / or a combination of multiple color matching parameters in the color scheme do not meet the preset color matching limit conditions, then adjust the color scheme of the enterprise recruitment promotional picture according to the preset adjustment strategy to obtain the updated enterprise recruitment promotional picture.

[0171] The device for generating an enterprise recruitment promotional picture provided by the embodiments of the present invention can execute the method for generating an enterprise recruitment promotional picture provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0172] In the technical solution of the present disclosure, the processing of collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0173] Figure 4The structural schematic diagram of the electronic device 10 for implementing the embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0174] As Figure 4 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0175] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0176] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as executing the method for generating an enterprise recruitment promotion map as described in any embodiment of the present invention.

[0177] That is, obtain the target recruitment text of the target enterprise, and according to the pre-constructed enterprise publicity map description database, obtain the target publicity map style description text that matches the target enterprise;

[0178] Combine the target recruitment text and the target publicity map style description text to obtain text input data, and input the text input data into a pre-trained publicity map format generation model to obtain publicity map style data in JSON format;

[0179] Input the publicity map style data into a pre-constructed rendering component to generate an enterprise recruitment publicity map for the target recruitment text, and publish the enterprise recruitment publicity map on a preset information release platform.

[0180] In some embodiments, the method for generating an enterprise recruitment publicity map as described in any embodiment of the present invention can be implemented as a computer program, which is tangibly included in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by the processor 11, one or more steps of the method for generating an enterprise recruitment publicity map as described in any embodiment of the present invention above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for generating an enterprise recruitment publicity map as described in any embodiment of the present invention by any other suitable means (for example, by means of firmware).

[0181] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0182] A computer program for implementing the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer program may be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0183] In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0184] In order to provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, speech input, or tactile input).

[0185] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected with each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0186] A computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0187] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0188] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for generating an enterprise recruitment promotion diagram, which is executed by a broker client, characterized in that Including: Obtain the target recruitment text of the target enterprise, and according to the pre-constructed enterprise publicity map description database, obtain the target publicity map style description text that matches the target enterprise; Combine the target recruitment text and the target publicity map style description text to obtain text input data, and input the text input data into the pre-trained publicity map format generation model to obtain publicity map style data in JSON format; Input the publicity map style data into the pre-constructed rendering component to generate an enterprise recruitment publicity map for the target recruitment text, and publish the enterprise recruitment publicity map on a preset information release platform.

2. The method according to claim 1, wherein Before obtaining the target publicity map style description text that matches the target enterprise according to the pre-constructed enterprise publicity map description database, it also includes: Collect the historical enterprise recruitment publicity maps released by each enterprise, and cluster each historical enterprise recruitment publicity map according to the enterprise to obtain a publicity map clustering cluster corresponding to each enterprise; Extract the image and text features of each historical enterprise recruitment publicity map included in each publicity map clustering cluster to obtain a publicity map style feature set corresponding to each enterprise; Input the publicity map style feature set of each enterprise into the pre-trained style description information generation model respectively to obtain the publicity map style description text corresponding to each enterprise; Construct an enterprise publicity map description database according to the publicity map style description text of each enterprise.

3. The method according to claim 1, characterized in that The method also includes: Match the target recruitment text with the historical standard recruitment information of the target enterprise to obtain the key content information in the target recruitment text; Construct a first personalized description text according to the key content information; Correspondingly, combining the target recruitment text and the target publicity map style description text to obtain text input data specifically includes: Combine the target recruitment text, the target publicity map style description text, and the first personalized description text to obtain text input data.

4. The method according to claim 1, wherein The method also includes: Obtain the current system time, and obtain time feature information according to the current system time; Construct a second personalized description text according to the time feature information; Correspondingly, combining the target recruitment text and the target publicity map style description text to obtain text input data specifically includes: Combine the target recruitment text, the target publicity map style description text, and the second personalized description text to obtain text input data.

5. The method according to claim 1, wherein The method also includes: Obtain the job seeking tendency information of all currently maintained members, and cluster each member according to the job seeking tendency information to obtain multiple member clusters; Construct a third personalized style description text corresponding to the job seeking tendency information of each member cluster; Combine the target recruitment text and the target publicity map style description text to obtain text input data, and input the text input data into the pre-trained publicity map format generation model to obtain publicity map style data in JSON format, specifically including: The target recruitment text and the target promotional graphic style description text are combined to obtain the original input data, and the original input data is respectively combined with the third personalized style description text corresponding to each member cluster to obtain multiple groups of text input data; Each group of text input data is respectively input into a pre-trained promotional graphic format generation model to obtain promotional graphic style data in JSON format corresponding to each member cluster; Correspondingly, the enterprise recruitment promotional graphic is published on a preset information release platform, specifically including: The enterprise recruitment promotional graphics corresponding to each member cluster are respectively published on the information release platforms matching each member cluster.

6. The method according to any one of claims 1-5, characterized in that, After generating the enterprise recruitment promotional graphic for the target recruitment text, it further includes: Performing image feature extraction on the enterprise recruitment promotional graphic, and based on the extraction result, obtaining a color scheme matching the enterprise recruitment promotional graphic; If it is determined that at least one color parameter and / or a combination of multiple color parameters in the color scheme do not meet the preset color matching restriction conditions, the color scheme of the enterprise recruitment promotional graphic is adjusted according to the preset adjustment strategy to obtain an updated enterprise recruitment promotional graphic.

7. A generating device for an enterprise recruitment promotion map, configured in a broker client, characterized in that It includes: A style description text acquisition module, configured to acquire the target recruitment text of the target enterprise, and according to a pre-constructed enterprise promotional graphic description database, acquire the target promotional graphic style description text matching the target enterprise; A style data acquisition module, configured to combine the target recruitment text and the target promotional graphic style description text to obtain text input data, and input the text input data into a pre-trained promotional graphic format generation model to obtain promotional graphic style data in JSON format; A promotional graphic release module, configured to input the promotional graphic style data into a pre-constructed rendering component to generate an enterprise recruitment promotional graphic for the target recruitment text, and publish the enterprise recruitment promotional graphic on a preset information release platform.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for generating an enterprise recruitment promotional graphic according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a processor to implement the method for generating an enterprise recruitment promotional graphic according to any one of claims 1-6 when executed.

10. A computer program product, characterized in that, The computer program product includes a computer program that implements the method for generating an enterprise recruitment promotional graphic according to any one of claims 1-6 when executed by a processor.

Images