A resume generation method and device for both school and enterprise

Abstract

The present application is applicable to the field of data processing technology, and provides a resume generation method and device for both school and enterprise ends, including: in response to a resume generation request of a first object, obtaining the original resume document of the target user in the resume generation request; determining at least one second object corresponding to the resume to be generated according to the resume generation request; determining the target attention dimension associated with the second object according to the correspondence between the preset attention feature dimension and the cluster; performing model correction on the preset behavior description model based on all attention dimensions of the second object to obtain the target behavior conversion model of the second object; importing the original resume document into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension; generating a calibrated resume document corresponding to the second object based on all description paragraphs. The above method can obtain a description paragraph that meets the content attention of the second object, thereby improving the accuracy of the calibrated resume document.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular, relates to a resume generation method and device applied to both schools and enterprises. Background Art

[0002] With the continuous development of electronic technology, more and more documents can be presented in electronic form. Among them, resumes are one of the commonly used and important electronic documents in the onboarding process. The expressiveness and accuracy of their content directly affect the employer's initial understanding of the student's situation and also affect the student's success rate in joining the company. Therefore, how to effectively generate resume documents that match the student's actual situation and the company's concerns has become the focus of users.

[0003] The existing resume generation method is generally written manually by students. However, when students submit resumes to companies, it is easy for the resume content to not match the company's job requirements, or some students exaggerate the resume content, resulting in the company being unable to understand the authenticity of the student's situation. It can be seen that the existing resume generation method relies on users to write manually, and there are problems such as low matching with the job or low authenticity. Summary of the invention

[0004] The embodiments of the present application provide a resume generation method and device for both schools and enterprises, which can solve the problems of existing resume generation technology, which relies on manual writing by users and has low matching degree with occupations or low authenticity.

[0005] In a first aspect, an embodiment of the present application provides a resume generation method applied to both school and enterprise ends, the method comprising:

[0006] In response to a resume generation request of a first object, obtaining an original resume document of a target user in the resume generation request; the first object is a campus cluster or an enterprise cluster;

[0007] Determine, according to the resume generation request, at least one second object corresponding to the resume to be generated; the second object is another cluster of a different cluster type from that of the first object;

[0008] Determining a target attention dimension associated with the second object according to a correspondence between a preset attention feature dimension and a cluster;

[0009] Performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object;

[0010] Importing the original resume document into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension;

[0011] Based on all the description segments, a calibration resume document corresponding to the second object is generated.

[0012] In a possible implementation manner of the first aspect, the first object is the campus cluster; the second object is the enterprise cluster;

[0013] Before determining the target attention dimension associated with the second object according to the correspondence between the preset attention feature dimension and the cluster, the method further includes:

[0014] Determine a plurality of job keywords according to the job description information of the second object, and convert the job keywords into corresponding job word vectors;

[0015] The vector distance between each position word vector is calculated respectively; the vector distance is:

[0016]

[0017] Among them, Dist (Work i ,Work j ) is the vector distance between the i-th job word vector and the j-th job word vector; Work i (n) is the value of the word vector of the i-th position in the n-th word dimension; j (n) is the value of the j-th position word vector in the n-th word dimension; N is the total number of word dimensions;

[0018] Based on the vector distance, keyword merging is performed on the plurality of job keywords to obtain a plurality of characteristic keywords;

[0019] Based on each of the feature keywords, the attention feature dimension of the second object is determined.

[0020] In a possible implementation manner of the first aspect, the performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object includes:

[0021] According to the characteristic keywords, a plurality of historical description paragraphs corresponding to the characteristic keywords in the historical resume are obtained;

[0022] Performing word frequency statistics on the plurality of historical description segments, and selecting, from the word groups in the plurality of historical description segments, a word group whose word frequency value is greater than a preset word frequency threshold as a high-frequency description word;

[0023] Performing model correction on the behavior description model based on all the high-frequency description words to obtain the target behavior conversion model of the target attention dimension corresponding to the characteristic keyword;

[0024] Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including:

[0025] According to each original description paragraph in the original resume document, calculating the dimensional matching degree between each original description paragraph and each target attention dimension;

[0026] If the dimensional matching degree between all the original description segments and any target attention dimension is less than a preset matching threshold, then obtaining the historical behavior record of the target user;

[0027] Extracting user events associated with any target attention dimension from all the historical behavior records;

[0028] The user event is imported into the target behavior conversion model corresponding to any target attention dimension to obtain the description segment corresponding to any target attention dimension.

[0029] In a possible implementation manner of the first aspect, the first object is the enterprise cluster; the second object is the campus cluster;

[0030] The determining, according to the correspondence between the preset attention feature dimension and the cluster, a target attention dimension associated with the second object includes:

[0031] Determine a professional cluster and at least one custom cluster of the target user in the campus cluster;

[0032] Determine a course list for the target user according to the professional type corresponding to the professional cluster; the course list includes at least one professional course for the target user; the professional course is used to determine the course feature dimension;

[0033] Based on the custom cluster, the extracurricular behavior range of the target user is obtained; the extracurricular behavior range is used to determine the extracurricular feature dimension;

[0034] The target attention dimension is obtained according to the course characteristic dimension and the extracurricular characteristic dimension.

[0035] In a possible implementation manner of the first aspect, the performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object includes:

[0036] determining a degree reference coefficient according to the object level of the second object;

[0037] Based on the course feature dimension, obtaining a course description segment associated with the course feature dimension;

[0038] Extracting history extracurricular language segments containing the extracurricular feature dimensions from the history resume database;

[0039] Calibrate the behavior description model according to the degree benchmark coefficient, the course description paragraph and the history extracurricular paragraph to obtain the target behavior conversion model;

[0040] Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including:

[0041] Obtaining a first description paragraph corresponding to the course feature dimension and a second description paragraph corresponding to the extracurricular feature dimension in the original resume document;

[0042] Calibrate the initial degree adverbs in the first description paragraph based on the degree calibration coefficient to obtain calibrated degree adverbs;

[0043] Importing the first description paragraph and the calibration degree adverb into the course conversion model of the course feature dimension, removing other characters in the first description paragraph whose correlation with the course description paragraph is less than a preset correlation threshold, and replacing the initial degree adverb with the calibration degree adverb to obtain a calibration course paragraph;

[0044] The second description paragraph is introduced into the extracurricular conversion model of the extracurricular feature dimension, and special characters in the second description paragraph whose correlation with the extracurricular description paragraph is less than a preset correlation threshold are marked to obtain the extracurricular feature paragraph.

[0045] In a possible implementation of the first aspect, in response to a resume generation request of the first object, obtaining an original resume document of a target user in the resume generation request includes:

[0046] According to the user ID of the target user, searching the user database to see whether there is an existing resume document corresponding to the user ID;

[0047] If the user database does not contain an existing resume document of the user ID, obtaining multiple historical behavior records associated with the user ID;

[0048] Based on the description dimensions included in the preset resume template, determine the associated behavior records corresponding to each description dimension from multiple historical behavior records;

[0049] According to each of the associated behavior records, an initial description segment corresponding to each of the description dimensions is obtained;

[0050] The original resume document is generated based on the initial description paragraphs of all description dimensions.

[0051] In a possible implementation manner of the first aspect, after generating a calibration resume document corresponding to the second object based on all the description segments, the method further includes:

[0052] receiving evaluation information about the calibration resume document sent by the second object;

[0053] Performing semantic analysis on the evaluation information to determine an evaluation level corresponding to the evaluation information and at least one target attention dimension described by the evaluation information;

[0054] If the evaluation level is lower than a preset evaluation threshold, model training is performed on the target behavior conversion model corresponding to the target attention dimension based on the evaluation information so that the description segment output by the target behavior model matches the evaluation information.

[0055] In a second aspect, an embodiment of the present application provides a resume generation device applied to both school and enterprise ends, the device comprising:

[0056] A request response unit, configured to respond to a resume generation request of a first object and obtain an original resume document of a target user in the resume generation request; the first object is a campus cluster or an enterprise cluster;

[0057] An object determination unit, configured to determine, according to the resume generation request, at least one second object corresponding to the resume to be generated; the second object being another cluster of a different cluster type from that of the first object;

[0058] an attention dimension determination unit, configured to determine a target attention dimension associated with the second object according to a correspondence between a preset attention feature dimension and a cluster;

[0059] A model correction unit, configured to perform model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object;

[0060] A description paragraph correction unit, used for importing the original resume document into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension;

[0061] A resume generating unit is used to generate a calibration resume document corresponding to the second object based on all the description segments.

[0062] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a method as described in any one of the first aspects above when executing the computer program.

[0063] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the first aspects above is implemented.

[0064] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes any one of the methods described in the first aspect.

[0065] Compared with the prior art, the embodiment of the present application has the following beneficial effects: after receiving a resume generation request sent by a certain cluster object, the original resume document of the corresponding target user can be obtained, and the target cluster (i.e., the second object) corresponding to the resume generation request can be determined, and then the corresponding target attention dimension can be determined according to the focus of the second object, so that the behavior description model can be calibrated according to the target attention dimension, so that the target behavior conversion model matching the content of the target attention dimension can be output, and the original document can be poured into the target behavior conversion model corresponding to each target attention dimension respectively, so that the description paragraph matching the focus of the second object can be output, and then all the description paragraphs can be merged to generate the corresponding calibration resume document. Compared with the existing resume generation technology, the resume generation method provided by the embodiment of the present application can perform model correction on the native behavior description model according to the focus of the second object, i.e., the target attention dimension, obtain a target behavior conversion model matching the content description method of the target attention dimension, and calibrate the description of the relevant content in the original resume document to obtain a description paragraph matching the content attention of the second object, thereby improving the accuracy of the calibration resume document and improving the efficiency of resume generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0067] Figure 1 It is a structural diagram of a school-enterprise interaction system provided by an embodiment of the present application;

[0068] Figure 2 This is a schematic diagram of an implementation of a resume generation method applied to both school and enterprise ends provided in an embodiment of the present application;

[0069] Figure 3This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends before S203 provided in the second embodiment of the present application;

[0070] Figure 4 This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends in S204 and S205 provided in the third embodiment of the present application;

[0071] Figure 5 This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends before S203 provided in the fourth embodiment of the present application;

[0072] Figure 6 This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends in S204 and S205 provided in the fifth embodiment of the present application;

[0073] Figure 7 This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends in S201 provided in the sixth embodiment of the present application;

[0074] Figure 8 This is a specific implementation flow chart of a resume generation method applied to both school and enterprise ends after S206 provided in the seventh embodiment of the present application;

[0075] Fig. 9 It is a structural diagram of a resume generation device applied to both school and enterprise ends provided in an embodiment of the present application;

[0076] Fig.10 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0077] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0078] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0079] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0080] The resume generation method for both school and enterprise provided in the embodiment of the present application can be applied to the school-enterprise interaction system. For example, Figure 1 FIG. 1 shows a schematic diagram of the structure of a school-enterprise interaction system provided by an embodiment of the present application. Figure 1 The school-enterprise interaction system includes a first server 11 deployed in a campus cluster, a resume generating device 12, and a second server 13 deployed in an enterprise cluster. The first server 11 deployed in the campus cluster can communicate with multiple student terminals 14 to receive original resume documents sent by the student terminals 14.

[0081] The student terminal 14 may be a computer, a smart phone or a tablet computer used by the student. The student terminal 14 may be installed with a client program corresponding to the campus cluster. The client program may include a resume correction function, through which the original resume document imported by the student may be calibrated to obtain a corrected resume document. The student terminal 14 may send a resume generation request to the first server 11, and the first server 11 may forward it to the resume generation device 12, so that the corrected resume document is generated by the resume generation device.

[0082] In this embodiment, the second server 13 deployed in the enterprise cluster can communicate with multiple employee terminals 15 to receive the original resume document sent by the employee terminal 15. The employee terminal 15 can be a terminal device such as a computer, a smart phone or a tablet computer used by enterprise employees. The employee terminal 15 can also be installed with an office client program within the enterprise. The office client program can include a resume correction function, through which the original resume document imported by the employee can be calibrated to obtain a corrected resume document. The employee terminal 15 can send a resume generation request to the second server 13, and the second server 13 can forward it to the resume generation device 12 to generate a corrected resume document through the resume generation device.

[0083] See also Figure 2 , Figure 2 The embodiment of the present application shows a schematic diagram of a resume generation method for both school and enterprise provided in the embodiment of the present application. The resume generation method is applied to the resume generation device mentioned above. The execution subject of the embodiment of the present application can be the resume generation device mentioned above. The resume generation device can be any electronic device, such as a computer, a laptop, a server, etc. Specifically, the method includes the following steps:

[0084] In S201, in response to a resume generation request of a first object, an original resume document of a target user in the resume generation request is obtained; the first object is a campus cluster or an enterprise cluster.

[0085] In this embodiment, the electronic device may receive a resume generation request initiated by a target user. When the target user initiates the resume generation request, the target user may select a source object type, which includes a campus type and an enterprise type. The source object type may specifically be an object type of the cluster to which the target user belongs. For example, if the person who initiates the resume generation request is a student, the first object is a campus cluster, and the source object type is a campus type; if the person who initiates the resume generation request is an employee in an enterprise, the first object is an enterprise cluster, and the source object type is an enterprise type.

[0086] In this embodiment, the above resume generation scenarios specifically include two scenarios, one of which is that a user in a campus cluster initiates a resume generation request to generate a resume document that is easy for enterprise users to read; the other scenario is that a user in an enterprise cluster initiates a resume generation request to generate a resume document that filters exaggerated or irrelevant content. The specific two scenarios can be described as follows:

[0087] Scenario 1: Student A needs to submit a resume to Company 1. Student A's major is journalism, and the content of his resume is mainly journalism-related; the position that Student A needs to apply for in Company 1 is a program compiler position, and the content of this position is less relevant to the content of his original resume. Student A can use the resume generation method provided in the embodiment of the present application to convert the original resume document into a corrected resume document that has a greater correlation with the skills of the program compiler position, thereby improving the efficiency of resume generation, and also improving the matching degree and accuracy between the corrected resume document and the company position that the user needs to apply for, and reducing unnecessary editing operations of the resume text.

[0088] Scenario 2: Employee A of an enterprise receives resumes of multiple students, including the resume of Student A. Since Student A may have exaggerated some of his experiences when writing his resume, or may have included descriptions of experiences that are irrelevant to the position he applied for, in order to improve the reading speed of Student A's resume document, Employee A can use the resume generation method provided in the embodiment of the present application to convert Student A's original resume document into a corrected resume document related to the position he applied for, thereby improving the reading speed of Employee A for the resume document. It should be noted that Employee A of an enterprise can initiate the generation of resume documents for multiple resume documents that need to be read, thereby achieving the purpose of batch generating corrected resume documents.

[0089] In this embodiment, the electronic device may receive a resume generation request sent by any user in the first object. The resume generation request may carry an original resume document. The original resume document records the user ID of the target user. The target user of the original resume document is specifically the user described in the original resume document, that is, the original resume document is specifically used to describe the learning and work experience corresponding to the target user.

[0090] In a possible implementation, the electronic device may communicate with a resume database. In this case, the electronic device may obtain the resume identifier corresponding to the target user from the resume database according to the resume identifier in the resume generation request.

[0091] In S202, at least one second object corresponding to the resume to be generated is determined according to the resume generation request; the second object is another cluster of a different cluster type from that of the first object.

[0092] In this embodiment, the resume generation request sent by the first object may carry the object identifier of the second object required to adapt the resume document, and then determine the cluster where the corresponding second object is located based on the object identifier of the second object. Exemplarily, if the first object is a campus cluster, the second object may specifically be the enterprise cluster to which the target user is posted, or a subcluster of the corresponding position in the enterprise cluster; if the first object is an enterprise cluster, the second object may specifically be the campus cluster where the target user is located, or a subcluster of the corresponding major in the campus cluster where the target user is located. It should be noted that the clusters in which the first object and the second object are located are different. For example, if the first object is any campus cluster, the second object may be any cluster in the enterprise cluster; and if the first object is any enterprise cluster, the second object may be any cluster in the campus cluster. The embodiment of the present application is specifically used to correct the content of the attention dimension differences between different clusters in a resume document.

[0093] In some implementations, in the above scenario 1, when students need to apply for multiple positions in multiple companies at the same time, the number of the above second objects can be multiple, and each second object can correspond to a position in an enterprise, that is, used to determine the cluster corresponding to the position in the enterprise.

[0094] In S203, a target attention dimension associated with the second object is determined based on a correspondence between preset attention feature dimensions and clusters.

[0095] In this embodiment, the electronic device can store the correspondence between different clusters and attention feature dimensions. Exemplarily, Table 1 shows a correspondence table between clusters and attention feature dimensions provided in an embodiment of the present application. As described in Table 1, the electronic device can determine the target attention dimension of the second object from the above correspondence table based on the cluster identifier corresponding to the second object. Among them, the second object includes an identifier corresponding to the enterprise cluster and an identifier corresponding to the position cluster, that is, it is used to determine the enterprise name and the position name. For example, the above-mentioned second object is a program compilation position in enterprise A, and the corresponding target attention dimensions are: C language, Java language, project development, and programming logical thinking.

[0096]

[0097] Table 1

[0098] In some implementations, some cluster identifiers in the above-mentioned second object can be default. For example, when the identifier of the position cluster is default, it means that all professional positions of enterprise A need to be delivered. If the object identifier of the second object is enterprise A, it means that the corrected resume documents corresponding to each position need to be generated separately, that is, the corrected resume document for the program compiler position and the corrected resume document for the sales position need to be converted. For another example, when the identifier of the enterprise cluster is default, the identifier needs to be delivered to all enterprises with this position. If the object identifier of the second object is the program compiler position, it means that the corrected resume document for the position in enterprise A and the corrected resume document for the position in enterprise B need to be generated separately.

[0099] In S204, the preset behavior description model is corrected based on all the target attention dimensions of the second object to obtain the target behavior conversion model of the second object.

[0100] In this embodiment, after determining the target attention dimension of the second object, the electronic device can calibrate the preset behavior description model according to the description features such as description keywords, description emphasis, and character typesetting corresponding to the target attention dimension, that is, the output of the behavior description model matches the description habits and description methods of the target attention dimension. Among them, a second object can correspond to multiple target attention dimensions, and the electronic device can calibrate the behavior description model according to the description features corresponding to each target attention dimension, so as to obtain the target behavior conversion model corresponding to the multiple target attention dimensions.

[0101] In S205, the original resume document is imported into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension.

[0102] In this embodiment, after the electronic device generates the target behavior conversion model of the target attention dimension corresponding to the second object, it can import the original resume document into each target behavior conversion model respectively, so that it can generate a descriptive paragraph that matches the descriptive features of the target attention dimension through the target behavior conversion model corresponding to the target attention dimension.

[0103] For example, the original resume document only describes that the student has studied programming language courses, but does not specifically explain or describe programming ability. Therefore, when the original resume document is imported into the target behavior conversion model corresponding to the programming language dimension, the target behavior conversion model can extract the types of programming language courses studied by the student and the corresponding assessment scores of each programming language course, and convert them into a description paragraph about the programming language dimension based on the above two dimensional features.

[0104] In some implementations, if the original resume document contains an original description segment corresponding to the target attention dimension, then when the original resume document is imported into the above-mentioned target behavior conversion model, each original description segment can be matched with its corresponding target attention dimension through the content extraction layer. If the matching degree between any original description segment and the target attention dimension is greater than a preset matching threshold, the original description segment is identified as a target description segment, and the target description segment is content-calibrated through the description calibration layer in the target behavior model to output the above-mentioned description segment; if the matching degree between all original description segments and the target attention dimension is less than the above-mentioned matching threshold, the characteristic keywords associated with the target attention dimension in the original resume document are extracted, and multiple characteristic keywords are imported into the content conversion layer in the above-mentioned target behavior conversion model to obtain the initial description content, and then the initial description content is imported into the description calibration layer to output the corresponding description segment.

[0105] In S206, a calibration resume document corresponding to the second object is generated based on all the description segments.

[0106] In this embodiment, the electronic device may store a resume template. After generating the description paragraphs corresponding to each target attention dimension, each description paragraph may be imported into the above resume template, and the basic information related to the target user in the original resume document may be imported into the above resume template, thereby obtaining a calibrated resume document matching the second object.

[0107] In some implementations, each target attention dimension is also configured with a corresponding display priority. The electronic device can determine the display position of the corresponding description segment according to each display priority, and import the description segment into the area in the resume template that matches the display position, thereby obtaining the corresponding calibrated resume document.

[0108] As can be seen from the above, a resume generation method applied to both ends of schools and enterprises provided by the embodiment of the present application can obtain the original resume document of the corresponding target user after receiving the resume generation request sent by a certain cluster object, and determine the target cluster (i.e., the second object) corresponding to the resume generation request, and then determine the corresponding target attention dimension according to the focus of the second object, so as to calibrate the behavior description model according to the target attention dimension, so as to output a target behavior conversion model that matches the content of the target attention dimension, and pour the original document into the target behavior conversion model corresponding to each target attention dimension respectively, so as to output a description paragraph that matches the focus of the second object, and then merge all the description paragraphs, so as to generate a corresponding calibration resume document. Compared with the existing resume generation technology, the resume generation method provided by the embodiment of the present application can perform model correction on the native behavior description model according to the focus of the second object, i.e., the target attention dimension, obtain a target behavior conversion model that matches the content description method of the target attention dimension, and calibrate the relevant content in the original resume document to obtain a description paragraph that meets the content attention of the second object, thereby improving the accuracy of the calibration resume document and improving the efficiency of resume generation.

[0109] Figure 3 The specific implementation flow chart of a resume generation method applied to both school and enterprise provided in the second embodiment of the present application in S203 is shown. Figure 3 , relative to Figure 2 The embodiment provides a resume generation method for both school and enterprise, and further includes: S301 to S304, which are described in detail as follows:

[0110] Further, the first object is the campus cluster; the second object is the enterprise cluster;

[0111] Before determining the target attention dimension associated with the second object according to the correspondence between the preset attention feature dimension and the cluster, the method further includes:

[0112] In S301, a plurality of job keywords are determined according to the job description information of the second object, and the job keywords are converted into corresponding job word vectors.

[0113] In this embodiment, the electronic device can automatically generate a corresponding relationship between the company's position and the target attention dimension. If a company publishes a relevant job position, it will attach the job description information corresponding to the job position. The electronic device can monitor the information published by various companies on the recruitment network. If a company publishes new recruitment information, it will identify the recruitment position / position corresponding to the recruitment information, and create a corresponding second object based on the company name and position identifier, that is, determine the cluster corresponding to the company and the sub-cluster corresponding to the position in the company cluster.

[0114] In this embodiment, after obtaining the job description information published by the second object, the electronic device can import the job description information into a preset word segmentation recognition model, perform word segmentation on the job description information, determine multiple candidate phrases, and calculate the correlation between each candidate phrase and the job, and select the candidate phrase with a correlation greater than a preset correlation threshold as the above-mentioned job keyword.

[0115] In this embodiment, the electronic device can establish a corresponding word coordinate system, which includes multiple word dimensions. The word dimensions include: degree dimension (such as very, extremely corresponding to a larger degree dimension value, and slightly, a little understanding, etc. corresponding to a smaller degree dimension value), skill description dimension, etc. The electronic device can determine the dimension value corresponding to each job keyword in each word dimension, and thus generate a job word vector corresponding to the job keyword based on the dimension value of each word dimension.

[0116] In S302, the vector distance between each position word vector is calculated respectively; the vector distance is:

[0117]

[0118] Among them, Dist (Work i ,Work j ) is the vector distance between the i-th job word vector and the j-th job word vector; Work i (n) is the value of the word vector of the i-th position in the n-th word dimension; j (n) is the value of the j-th position word vector in the n-th word dimension; N is the total number of word dimensions.

[0119] In S303, based on the vector distance, keyword merging is performed on the plurality of job keywords to obtain a plurality of feature keywords.

[0120] In this embodiment, after obtaining the job word vectors of each job keyword in the job description information, the electronic device can merge all the job keywords, so it is necessary to determine the related job keywords, that is, to determine the vector distance between each job keyword. Among them, the vector distance specifically determines the deviation between each job word vector in each word dimension. Therefore, the deviation value between each job word vector in each word dimension can be calculated separately, that is, Work i (n)-Work j (n), and according to the deviation values ​​between all word dimensions, the vector distance between the keywords of each position is obtained.

[0121] In this embodiment, if the vector distance between any two job keywords is farther (i.e., the larger the value), the corresponding correlation is lower, which means that the two job keywords correspond to different contents; conversely, if the vector distance between any two job keywords is closer (i.e., the smaller the value), the corresponding correlation is higher, which means that the two job keywords correspond to the same content. Based on this, the electronic device can be set with a corresponding distance threshold, merge multiple job keywords whose vector distance is less than the above distance threshold, and retain each job keyword whose vector distance is greater than the sum of the distance thresholds, and use all the job keywords after the merge process as the above-mentioned feature keywords.

[0122] In S304, the attention feature dimension of the second object is determined based on each of the feature keywords.

[0123] In this embodiment, each feature keyword may correspond to an attention feature dimension, and then the above-mentioned corresponding relationship may be constructed according to all attention feature dimensions of the second object.

[0124] In an embodiment of the present application, by obtaining job description information, and then determining the job keywords associated with the second object, and then merging all job keywords based on the vector distance between the word vectors between the job keywords, the attention feature dimension of the second object can be determined, and a corresponding relationship between the second object and the attention feature dimension can be constructed, thereby achieving the purpose of automatically constructing a corresponding relationship based on the published job description information, and improving the degree of automation and efficiency of resume generation.

[0125] Figure 4 The specific implementation flow chart of a resume generation method applied to both school and enterprise provided in the third embodiment of the present application in S204 and S205 is shown. Figure 4 , relative to Figure 3 In the embodiment, a resume generation method applied to both school and enterprise provided in this embodiment includes: S401 to S403 in S204, and S205 includes S404 to 7, which are described in detail as follows:

[0126] Furthermore, the preset behavior description model is corrected based on all the target attention dimensions of the second object to obtain the target behavior conversion model of the second object, including:

[0127] In S401, based on the characteristic keywords, a plurality of historical description segments corresponding to the characteristic keywords in the historical resume are obtained.

[0128] In this embodiment, the electronic device can store a corresponding resume database. The resume database contains multiple historical resumes that have been received. Each historical resume contains descriptive paragraphs about each dimension, and the electronic device can identify the obtained characteristic keywords and retrieve historical paragraphs from all historical resumes to see if they contain the above-mentioned characteristic keywords. If a historical resume contains a historical paragraph about the characteristic keyword, the historical paragraph is converted into a corresponding content feature vector through a semantic recognition algorithm, and the vector distance between the content feature vector and the word vector of the characteristic keyword is calculated, so as to determine the degree of match between the historical paragraph and the characteristic keyword in content expression, and then the historical paragraph whose vector distance with the word vector of the characteristic keyword is less than a preset distance threshold is used as the historical description paragraph corresponding to the characteristic keyword.

[0129] In S402, word frequency statistics are performed on the plurality of historical description segments, and word groups whose word frequency values ​​are greater than a preset word frequency threshold are selected from the word groups in the plurality of historical description segments as high-frequency description words.

[0130] In this embodiment, after determining the historical description segment corresponding to the characteristic keyword, the electronic device can perform word segmentation processing on the historical description segment to obtain the description keywords contained in the historical description segment, and respectively count the number of occurrences of each description keyword in all historical description segments, that is, the above-mentioned word frequency. If the word frequency value of a certain description keyword is greater than the preset word frequency threshold, the description keyword is used as a high-frequency keyword, that is, the high-frequency keyword is a common word when describing the target attention dimension, so that the description segment can be converted to be more in line with the description habits of the target attention dimension, so as to improve the accuracy of the description segment content.

[0131] In S403, the behavior description model is corrected based on all the high-frequency description words to obtain the target behavior conversion model of the target attention dimension corresponding to the feature keyword.

[0132] In this embodiment, after the electronic device determines the high-frequency descriptive words corresponding to the target attention dimension (i.e., any feature keyword), it can perform model correction on the behavior description model based on all high-frequency descriptive words to obtain a target behavior conversion model corresponding to the target attention dimension, so that the description segment output by the target conversion model can conform to the description characteristics of the target attention dimension, such as descriptive word habits.

[0133] In some implementations, the electronic device can also determine the conjunctions and degree adverbs corresponding to each high-frequency descriptive word to form a corresponding high-frequency phrase, and then import the high-frequency phrase into the above-mentioned behavior description model for model correction, so that the output description paragraph can conform to the corresponding description habits to improve the accuracy of description paragraph generation.

[0134] Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including:

[0135] In S404, the dimensional matching degree between each original description segment and each target attention dimension is calculated according to each original description segment in the original resume document.

[0136] In this embodiment, when the electronic device imports the original resume document into the target behavior conversion model, it can respectively calculate the dimensional matching degree between each original description segment in each original resume document and the target attention dimension, thereby reducing the content conversion operations of unrelated original description segments and reducing the amount of calculation required for converting large amounts of text.

[0137] In this embodiment, the electronic device can convert each original description segment into a corresponding feature vector through a convolutional network, and perform vector distance calculation between the feature vector and the word vector of the feature keyword corresponding to the target attention dimension, and then determine the dimensional matching degree between each original description segment in the original resume document and the target attention dimension based on the above vector distance. The larger the value of the above vector distance, the lower the corresponding dimensional matching degree; conversely, if the value of the above vector distance is smaller, the corresponding dimensional matching degree is higher.

[0138] In some implementations, if the dimensional matching degree between at least one original description segment and the target attention dimension is greater than the matching threshold, it means that relevant content has already been described in the original resume document. In this case, the description of the existing content can be corrected. Therefore, all original description segments with dimensional matching degrees between the target attention dimension greater than the matching threshold can be imported into the target behavior conversion model to generate description segments that conform to the target attention dimension.

[0139] In S405, if the dimensional matching degree between all the original description segments and any target attention dimension is less than a preset matching threshold, the historical behavior record of the target user is obtained.

[0140] In this embodiment, if the dimensional matching degree of all original description segments in the original resume document with the target attention dimension is less than the matching threshold, it means that there is no description segment about the target attention dimension in the original resume document. At this time, it is impossible to generate the corresponding description segment based on the existing original resume document. In this case, the electronic device can refer to the historical behavior record of the target user from the user data to determine whether there is a behavior event related to the target attention dimension through the historical behavior record of the target user, and then generate a description segment associated with the target user based on the related behavior event.

[0141] In this embodiment, the above historical behavior records may include the target user's class records, relevant work records of the school student union, off-campus internship records, on-campus competition records, and work records, etc. The specific type and quantity of historical behavior records to be obtained can be selected according to actual conditions. For example, all historical behavior records of the target user within a certain period of time are selected, or all historical behavior records related to the school are obtained.

[0142] In S406, user events associated with any target attention dimension are extracted from all the historical behavior records.

[0143] In this embodiment, each target attention dimension may correspond to at least one behavior type. The electronic device may extract all historical behavior records of the behavior type corresponding to the target attention dimension according to the behavior type of each historical behavior record, and then determine the user event corresponding to the target attention dimension for the target object according to all historical behavior records.

[0144] For example, if the above historical behavior records include multiple class records of a certain course, the electronic device can determine the course progress and course evaluation of the target user based on all the class records of the course, thereby obtaining the corresponding user event for the course. Specifically, the electronic device can integrate the content of multiple historical behavior records belonging to the same event, thereby merging them into one user event.

[0145] In S407, the user event is imported into the target behavior conversion model corresponding to any target attention dimension to obtain the description segment corresponding to any target attention dimension.

[0146] In this embodiment, after obtaining the user event corresponding to the target attention dimension, the electronic device can import the user event described in text form into the above-mentioned target behavior conversion model, so as to obtain the description paragraph corresponding to the target attention dimension, so that when the relevant content is not mentioned in the original resume document, it can also be automatically supplemented, thereby improving the accuracy of the subsequent corrected resume document.

[0147] Figure 5 The specific implementation flow chart of a resume generation method applied to both school and enterprise ends before S203 provided in the fourth embodiment of the present application is shown. Figure 5 , relative to Figure 2 The embodiment provides a resume generation method for both school and enterprise, and before S203, further includes: S501 to S504, which are described in detail as follows:

[0148] Further, the first object is the enterprise cluster; the second object is the campus cluster;

[0149] The determining, according to the correspondence between the preset attention feature dimension and the cluster, a target attention dimension associated with the second object includes:

[0150] In S501, a professional cluster and at least one custom cluster of the target user in the campus cluster are determined.

[0151] In this embodiment, the second object is a campus cluster. In this case, the electronic device can query the user database for the professional cluster and at least one custom cluster corresponding to the target user corresponding to the user ID in the campus cluster according to the user ID of the target user. The professional cluster is used to determine the major of the target user, that is, to specifically determine the course content that the target user has studied; the custom cluster is used to determine the group in which the target user participates voluntarily, such as the student union, social work organization, internship group, etc. The professional cluster and custom cluster can be entered into the user information of the target user, and the electronic device can determine them from the corresponding user information according to the user ID of the target user.

[0152] In S502, a course list of the target user is determined according to the professional type corresponding to the professional cluster; the course list includes at least one professional course of the target user; and the professional course is used to determine a course feature dimension.

[0153] In this embodiment, after determining the professional cluster to which the target user belongs, the electronic device can determine at least one professional type corresponding to the professional cluster. For example, if a user's major is a single major, the number of corresponding professional types is one; conversely, if a user's major is a double major, that is, the professional cluster to which the user belongs has two professional degrees, the number of the above professional types can be two. The electronic device can determine the corresponding professional type according to the cluster identifier of the professional cluster.

[0154] In this embodiment, the electronic device can store the correspondence between the professional type and the professional courses. After determining the professional type of the professional cluster where the target user is located, the course list corresponding to the target user can be determined through the above correspondence. The course list records the professional courses that the target user has studied during school. The number of the above professional courses can be multiple.

[0155] In S503, based on the custom cluster, the extracurricular behavior range of the target user is obtained; the extracurricular behavior range is used to determine the extracurricular feature dimension.

[0156] In this embodiment, after the electronic device determines the custom cluster of the target user, it can determine the corresponding extracurricular behavior scope according to the cluster type of the custom cluster. For example, if the custom cluster is one of the organizations of the student union, then the scope of organizational activities of the organization can be determined; for example, if the custom cluster is an internship group in which the target user participates, then the internship scope corresponding to the internship group can be determined; for example, if the custom cluster is an interest class joined by the target user, then the learning scope corresponding to the interest class can be determined, etc.

[0157] In this embodiment, each extracurricular behavior range may correspond to at least one extracurricular feature dimension. The electronic device may determine at least one extracurricular behavior subject according to the extracurricular behavior range determined above, and obtain the extracurricular feature dimension based on the extracurricular behavior subject.

[0158] In S504, the target attention dimension is obtained according to the course characteristic dimension and the extracurricular characteristic dimension.

[0159] In this embodiment, the electronic device can determine the relevant characteristic dimensions of the target user on and off campus as the target attention dimensions of the target user, so as to judge the authenticity of the relevant content about the target user in the original resume document, thereby improving the accuracy of subsequent content correction.

[0160] Figure 6 The specific implementation flow chart of a resume generation method applied to both school and enterprise provided in the fifth embodiment of the present application in S204 and S205 is shown. Figure 6 , relative to Figure 5 In the embodiment, a resume generation method applied to both school and enterprise provided in this embodiment includes: S601 to S604 in S204, and the above S205 includes: S605 to S608, which are described in detail as follows:

[0161] Furthermore, the preset behavior description model is corrected based on all the target attention dimensions of the second object to obtain the target behavior conversion model of the second object, including:

[0162] In S601, a degree reference coefficient is determined according to the object level of the second object.

[0163] In this embodiment, the second object is a campus cluster, and the learning level of the students is related to the teaching level of the campus where they are located. Therefore, when the electronic device calibrates the original resume document of the target user, it can determine the corresponding degree reference coefficient according to the object level corresponding to the second object. For example, the higher the object level of the second object, the stronger the teaching ability corresponding to the second object, and the larger the value of the corresponding degree reference coefficient; conversely, the lower the object level of the second object, the weaker the teaching ability corresponding to the second object, and the smaller the value of the corresponding degree reference coefficient. Therefore, through the above-mentioned degree reference coefficient, the degree of the content related to the course in the original resume document can be calibrated to improve the accuracy of the subsequent content.

[0164] In S602, based on the course feature dimension, a course description segment associated with the course feature dimension is obtained.

[0165] In this embodiment, after determining the professional courses offered in the campus cluster where the target user is located, that is, the above-mentioned course feature dimensions, the electronic device can obtain the course description paragraphs associated with the above-mentioned course feature dimensions. The above-mentioned course description paragraphs can be extracted from the historical resume documents in the resume library.

[0166] In S603, the history extracurricular language segments containing the extracurricular feature dimension are extracted from the history resume database.

[0167] In this embodiment, after determining the extracurricular activities that the target user can carry out outside of school, that is, determining the extracurricular feature dimension, the electronic device can extract the historical extracurricular language segment corresponding to the above extracurricular feature dimension from the historical resume document in the resume library. For example, the above extracurricular feature dimension can correspond to at least one extracurricular activity keyword, and through the extracurricular activity keyword, it is determined whether the historical resume document contains a description paragraph of the above extracurricular activity keyword. If so, the above description paragraph is used as the above historical extracurricular language segment.

[0168] In S604, the behavior description model is calibrated according to the degree benchmark coefficient, the course description paragraph and the history extracurricular paragraph to obtain the target behavior conversion model.

[0169] In this embodiment, the electronic device can calibrate the behavior description model based on all history extracurricular language segments to obtain a target behavior conversion model calibrated for the extracurricular language segments; correspondingly, the electronic device can also calibrate the behavior description model based on all course description segments and corresponding benchmark calibration coefficients to obtain a target behavior conversion model related to the course.

[0170] Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including:

[0171] In S605, a first description paragraph corresponding to the course feature dimension and a second description paragraph corresponding to the extracurricular feature dimension in the original resume document are obtained.

[0172] In this embodiment, the electronic device can classify the description paragraphs in the original resume document into a first description paragraph related to the course content and a second description paragraph related to the outside of the campus (i.e., extracurricular content). The number of the first description paragraphs and the second description paragraphs can be multiple. Of course, in some scenarios, some original resume documents may not contain the first description paragraph or the second description paragraph, which can be determined according to actual conditions.

[0173] In S606, the initial degree adverbs in the first description segment are calibrated based on the degree calibration coefficient to obtain calibrated degree adverbs.

[0174] In this embodiment, the electronic device can identify the initial degree adverb in the first description paragraph, wherein the initial degree adverb includes words describing the degree in the first description paragraph, such as "very", "skilled" and "slightly", etc., and can also be characters describing the degree through numbers, such as "100%", "90" in "assessment score is 90", etc. After determining the above initial degree adverb, the electronic device can perform weighted correction on the above initial degree adverb according to the calibration degree coefficient corresponding to the object level of the second object, so as to obtain the corresponding calibration degree adverb.

[0175] In S607, the first description paragraph and the calibration degree adverb are imported into the course conversion model of the course feature dimension, other characters in the first description paragraph whose correlation with the course description paragraph is less than a preset correlation threshold are removed, and the initial degree adverb is replaced by the calibration degree adverb to obtain a calibration course paragraph.

[0176] In this embodiment, after adjusting the degree adverbs in the description paragraph, the electronic device can import the first description paragraph and the adjusted calibrated degree adverbs into the course conversion model. The course conversion model can determine the content related to the course dimension based on the course description paragraph, that is, obtain the course keywords related to the course, so as to identify other characters in the first description paragraph that are not related to the course content, and remove other characters in the above-mentioned first description paragraph. Correspondingly, the initial degree adverbs in the first description paragraph are replaced with calibrated degree adverbs, and other characters irrelevant to the content and the first description paragraph after degree calibration are removed as the above-mentioned calibrated course paragraph, so that the content can be calibrated and irrelevant content can be removed to improve the reading speed of enterprise employees.

[0177] In S608, the second description paragraph is imported into the extracurricular conversion model of the extracurricular feature dimension, and special characters in the second description paragraph whose correlation with the extracurricular description paragraph is less than a preset correlation threshold are marked to obtain the extracurricular feature paragraph.

[0178] In this embodiment, unlike the relevant content on campus, the course content generally has a greater similarity, so the course description paragraph can be used to identify the content with a correlation less than the correlation threshold as invalid content; and extracurricular activities often have personalized characteristics, so the content with a lower correlation with the extracurricular description paragraph belongs to the personalized content of the target user, and this part of the content can be highlighted by marking. The above-mentioned marking includes highlighting, text enlargement, and text color adjustment. After identifying the special characters in the second description paragraph, the electronic device can mark the characteristic characters in a preset manner, and use the second description paragraph marked with the characteristic characters as the above-mentioned extracurricular characteristic paragraph, so that corporate employees can notice the personalized content of the target user in the original resume text.

[0179] In the embodiment of the present application, different methods are used to calibrate descriptions of on-campus activities and off-campus activities, thereby improving the accuracy of content calibration and also improving the speed at which subsequent corporate employees read resume texts.

[0180] Figure 7 The specific implementation flow chart of a resume generation method applied to both school and enterprise ends in S201 provided in the sixth embodiment of the present application is shown. Figure 7 , relative to Figure 2-6 In any of the embodiments, the resume generation method for both school and enterprise provided in this embodiment includes S2011 to S2015, which are described in detail as follows:

[0181] In S2011, based on the user ID of the target user, a search is performed to determine whether the user database contains an existing resume document corresponding to the user ID.

[0182] In this embodiment, the electronic device can store a resume library. After receiving a resume generation request, the user identifier of the target user can be extracted from the resume generation request, and it can be determined whether an existing resume document corresponding to the user identifier is already stored in the resume library; if the resume library contains an existing resume document corresponding to the user identifier, the existing resume document is used as the original resume document of the target user; conversely, if the resume library does not store an existing resume document corresponding to the user identifier, the corresponding original resume document can be automatically generated by the electronic device.

[0183] In S2012, if the user database does not contain an existing resume document of the user identifier, multiple historical behavior records associated with the user identifier are obtained.

[0184] In S2013, based on the description dimensions included in the preset resume template, associated behavior records corresponding to each description dimension are determined from multiple historical behavior records.

[0185] In S2014, initial description segments corresponding to the description dimensions are obtained according to the associated behavior records.

[0186] In this embodiment, when the electronic device generates the original resume document of the target user, it can extract multiple historical behavior records corresponding to the target user from the user database, so as to determine multiple user events that occurred to the user on and off campus, and then determine the associated behavior records corresponding to each description dimension according to the description dimensions contained in the resume template, so as to determine the corresponding user behavior event of the description dimension based on all the associated behavior records corresponding to the description dimension, and import the user behavior event corresponding to the description dimension into the description conversion model based on the user behavior event of the target user in the description dimension, so as to obtain the corresponding initial description segment.

[0187] In S2015, the original resume document is generated based on the initial description paragraphs of all description dimensions.

[0188] In this embodiment, after obtaining the initial description paragraphs corresponding to all description dimensions, they can be imported into the display area of ​​the corresponding description dimension in the resume template, thereby generating the corresponding original resume document, which can realize the automatic generation of the original resume document and improve the degree of automation of resume generation.

[0189] Figure 8 FIG. 2 shows a specific implementation flow chart of a resume generation method applied to both school and enterprise provided in the seventh embodiment of the present application after S206. Figure 8, relative to Figure 2-6 In any of the embodiments, the resume generation method for school-enterprise dual-end provided in this embodiment further includes, after S206: S801 to S803, which are described in detail as follows:

[0190] In S801, evaluation information about the calibration resume document sent by the second object is received.

[0191] In S802, semantic analysis is performed on the evaluation information to determine an evaluation level corresponding to the evaluation information and at least one target attention dimension described by the evaluation information;

[0192] In S803, if the evaluation level is lower than a preset evaluation threshold, model training is performed on the target behavior conversion model corresponding to the target attention dimension based on the evaluation information so that the description segment output by the target behavior model matches the evaluation information.

[0193] In this embodiment, after the electronic device generates a calibration resume document, the calibration resume document can be sent to the corresponding user terminal. For example, in scenario 1, it is sent to the relevant employees in the enterprise for viewing. In scenario 2, it can also be sent to the user who initiated the resume generation request. After the user terminal receives the calibration resume document, the content can be displayed in the preset interactive interface. After the relevant user has viewed the calibration resume document and interviewed the target user, the calibration resume document can be evaluated and fed back, that is, the above-mentioned evaluation information is obtained. The evaluation information may include content in multiple dimensions, namely, content accuracy, content completeness, and modification opinions. After receiving the above-mentioned calibration information, the electronic device can obtain the evaluation level for the calibration resume document and the evaluation information description for at least one target attention dimension. If the above-mentioned evaluation level is higher than the evaluation threshold, it means that the content of the calibration resume document is consistent with the actual situation, and the corresponding model does not need to be adjusted; on the contrary, if the evaluation level is lower than the preset rating threshold, the target behavior conversion model of a certain target attention dimension can be trained according to the evaluation information description in the evaluation information to improve the accuracy of subsequent output.

[0194] In this embodiment, Fig. 9 The structure diagram of a resume generation device applied to both school and enterprise provided by an embodiment of the present application is shown. The resume generation device applied to both school and enterprise includes various units for executing Figure 2 The steps implemented by the first device in the corresponding embodiment. For details, please refer to Figure 2 and Figure 2 For the convenience of explanation, only the parts related to this embodiment are shown.

[0195] See also Fig. 9 , a resume generation device used in both schools and enterprises, including:

[0196] A request response unit 91 is used to respond to a resume generation request of a first object and obtain an original resume document of a target user in the resume generation request; the first object is a campus cluster or an enterprise cluster;

[0197] An object determination unit 92 is used to determine at least one second object corresponding to the resume to be generated according to the resume generation request; the second object is another cluster of a different cluster type from that of the first object;

[0198] an attention dimension determination unit 93, configured to determine a target attention dimension associated with the second object according to a correspondence between a preset attention feature dimension and a cluster;

[0199] A model correction unit 94 is used to perform model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object;

[0200] A description paragraph correction unit 95 is used to import the original resume document into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension;

[0201] The resume generating unit 96 is used to generate a calibration resume document corresponding to the second object based on all the description segments.

[0202] It should be understood that Fig. 9 In the structural block diagram of the device shown, each module is used to execute Figures 2 to 8 The steps in the corresponding embodiments, and for Figures 2 to 8 Each step in the corresponding embodiment has been explained in detail in the above embodiment. Figures 2 to 8 as well as Figures 2 to 8 The relevant descriptions in the corresponding embodiments are not repeated here.

[0203] Fig.10 is a structural block diagram of an electronic device provided by another embodiment of the present application. Fig.10 The electronic device 1000 of this embodiment includes: a processor 1010, a memory 1020, and a computer program 1030 stored in the memory 1020 and executable by the processor 1010, such as a program for a resume generation method applied to both school and enterprise. When the processor 1010 executes the computer program 1030, the steps in each embodiment of the resume generation method applied to both school and enterprise are implemented, such as Figure 2 Alternatively, the processor 1010 implements the above when executing the computer program 1030 Fig.10 The functions of each module in the corresponding embodiment are, for example, Fig. 9 For details on the functions of the units 91 to 96, please refer to Fig.10 Related description in the corresponding embodiment.

[0204] Exemplarily, the computer program 1030 may be divided into one or more modules, one or more modules are stored in the memory 1020, and are executed by the processor 1010 to complete the present application. One or more modules may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program 1030 in the electronic device 1000. For example, the computer program 1030 may be divided into various unit modules, and the specific functions of each module are as described above.

[0205] The electronic device 1000 may include, but is not limited to, a processor 1010 and a memory 1020. Those skilled in the art will appreciate that Fig.10 It is only an example of the electronic device 1000 and does not constitute a limitation of the electronic device 1000. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0206] The processor 1010 may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0207] The memory 1020 may be an internal storage unit of the electronic device 1000, such as a hard disk or memory of the electronic device 1000. The memory 1020 may also be an external storage device of the electronic device 1000, such as a plug-in hard disk, a smart memory card, a flash memory card, etc. equipped on the electronic device 1000. Furthermore, the memory 1020 may also include both an internal storage unit of the electronic device 1000 and an external storage device.

[0208] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A resume generation method applied to both school and enterprise ends, characterized in that: include: In response to a resume generation request of the first object, obtaining an original resume document of a target user in the resume generation request; The first object is a campus cluster or an enterprise cluster; Determine, according to the resume generation request, at least one second object corresponding to the resume to be generated; the second object is another cluster of a different cluster type from that of the first object; Determining a target attention dimension associated with the second object according to a correspondence between a preset attention feature dimension and a cluster; Performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object; Importing the original resume document into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension; Based on all the description segments, generating a calibration resume document corresponding to the second object; The first object is the campus cluster; the second object is the enterprise cluster; Before determining the target attention dimension associated with the second object according to the correspondence between the preset attention feature dimension and the cluster, the method further includes: Determine a plurality of job keywords according to the job description information of the second object, and convert the job keywords into corresponding job word vectors; The vector distance between each position word vector is calculated respectively; the vector distance is: Among them, Dist (Work i ,Work j ) is the vector distance between the i-th job word vector and the j-th job word vector; Work i (n) is the value of the word vector of the i-th position in the n-th word dimension; j (n) is the value of the j-th position word vector in the n-th word dimension; N is the total number of word dimensions; Based on the vector distance, keyword merging is performed on the plurality of job keywords to obtain a plurality of characteristic keywords; Determine the attention feature dimension of the second object based on each of the feature keywords; or The first object is the enterprise cluster; the second object is the campus cluster; The determining, according to the correspondence between the preset attention feature dimension and the cluster, a target attention dimension associated with the second object includes: Determine a professional cluster and at least one custom cluster of the target user in the campus cluster; Determine a course list for the target user according to the professional type corresponding to the professional cluster; the course list includes at least one professional course for the target user; the professional course is used to determine a course feature dimension; Based on the custom cluster, the extracurricular behavior range of the target user is obtained; the extracurricular behavior range is used to determine the extracurricular feature dimension; The target attention dimension is obtained according to the course characteristic dimension and the extracurricular characteristic dimension.

2. The resume generation method according to claim 1, characterized in that: The method of performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object includes: According to the characteristic keywords, a plurality of historical description paragraphs corresponding to the characteristic keywords in the historical resume are obtained; Performing word frequency statistics on the plurality of historical description segments, and selecting, from the word groups in the plurality of historical description segments, a word group whose word frequency value is greater than a preset word frequency threshold as a high-frequency description word; Performing model correction on the behavior description model based on all the high-frequency description words to obtain the target behavior conversion model of the target attention dimension corresponding to the characteristic keyword; Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including: According to each original description paragraph in the original resume document, calculating the dimensional matching degree between each original description paragraph and each target attention dimension; If the dimensional matching degree between all the original description segments and any target attention dimension is less than a preset matching threshold, then obtaining the historical behavior record of the target user; Extracting user events associated with any target attention dimension from all the historical behavior records; The user event is imported into the target behavior conversion model corresponding to any target attention dimension to obtain the description segment corresponding to any target attention dimension.

3. The resume generation method according to claim 1, characterized in that: The method of performing model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object includes: determining a degree reference coefficient according to the object level of the second object; Based on the course feature dimension, obtaining a course description segment associated with the course feature dimension; Extracting history extracurricular language segments containing the extracurricular feature dimensions from the history resume database; Calibrate the behavior description model according to the degree benchmark coefficient, the course description paragraph and the history extracurricular paragraph to obtain the target behavior conversion model; Correspondingly, the original resume document is imported into the target behavior conversion model to obtain the description paragraphs corresponding to each target attention dimension, including: Obtaining a first description paragraph corresponding to the course feature dimension and a second description paragraph corresponding to the extracurricular feature dimension in the original resume document; Calibrate the initial degree adverbs in the first description paragraph based on the degree reference coefficient to obtain calibrated degree adverbs; Importing the first description paragraph and the calibration degree adverb into the course conversion model of the course feature dimension, removing other characters in the first description paragraph whose correlation with the course description paragraph is less than a preset correlation threshold, and replacing the initial degree adverb with the calibration degree adverb to obtain a calibration course paragraph; The second description paragraph is introduced into the extracurricular conversion model of the extracurricular feature dimension, and special characters in the second description paragraph whose correlation with the extracurricular description paragraph is less than a preset correlation threshold are marked to obtain the extracurricular feature paragraph.

4. The resume generation method according to any one of claims 1 to 3, characterized in that: The step of responding to the resume generation request of the first object and obtaining the original resume document of the target user in the resume generation request includes: According to the user ID of the target user, searching the user database to see whether there is an existing resume document corresponding to the user ID; If the user database does not contain an existing resume document of the user ID, obtaining multiple historical behavior records associated with the user ID; Based on the description dimensions included in the preset resume template, determine the associated behavior records corresponding to each description dimension from multiple historical behavior records; According to each of the associated behavior records, an initial description segment corresponding to each of the description dimensions is obtained; The original resume document is generated based on the initial description paragraphs of all description dimensions.

5. The resume generation method according to any one of claims 1 to 3, characterized in that: After generating a calibration resume document corresponding to the second object based on all the description segments, the method further includes: receiving evaluation information about the calibration resume document sent by the second object; Performing semantic analysis on the evaluation information to determine an evaluation level corresponding to the evaluation information and at least one target attention dimension described by the evaluation information; If the evaluation level is lower than a preset evaluation threshold, model training is performed on the target behavior conversion model corresponding to the target attention dimension based on the evaluation information so that the description segment output by the target behavior model matches the evaluation information.

6. A resume generation device applied to both school and enterprise, characterized in that: include: A request response unit, configured to respond to a resume generation request of a first object and obtain an original resume document of a target user in the resume generation request; The first object is a campus cluster or an enterprise cluster; An object determination unit, configured to determine, according to the resume generation request, at least one second object corresponding to the resume to be generated; the second object being another cluster of a different cluster type from that of the first object; an attention dimension determination unit, configured to determine a target attention dimension associated with the second object according to a correspondence between a preset attention feature dimension and a cluster; A model correction unit, configured to perform model correction on a preset behavior description model based on all the target attention dimensions of the second object to obtain a target behavior conversion model of the second object; A description paragraph correction unit, used for importing the original resume document into the target behavior conversion model to obtain description paragraphs corresponding to each target attention dimension; A resume generating unit, configured to generate a calibration resume document corresponding to the second object based on all the description segments; The first object is the campus cluster; the second object is the enterprise cluster; The attention dimension determination unit is also used for: Determine a plurality of job keywords according to the job description information of the second object, and convert the job keywords into corresponding job word vectors; The vector distance between each position word vector is calculated respectively; the vector distance is: Among them, Dist (Work i ,Work j ) is the vector distance between the i-th job word vector and the j-th job word vector; Work i (n) is the value of the word vector of the i-th position in the n-th word dimension; j (n) is the value of the j-th position word vector in the n-th word dimension; N is the total number of word dimensions; Based on the vector distance, keyword merging is performed on the plurality of job keywords to obtain a plurality of characteristic keywords; Determine the attention feature dimension of the second object based on each of the feature keywords; or The first object is the enterprise cluster; the second object is the campus cluster; The attention dimension determination unit is also used for: Determine a professional cluster and at least one custom cluster of the target user in the campus cluster; Determine a course list for the target user according to the professional type corresponding to the professional cluster; the course list includes at least one professional course for the target user; the professional course is used to determine a course feature dimension; Based on the custom cluster, the extracurricular behavior range of the target user is obtained; the extracurricular behavior range is used to determine the extracurricular feature dimension; The target attention dimension is obtained according to the course characteristic dimension and the extracurricular characteristic dimension.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images