Enterprise investigation data analysis method and system for application education

Abstract

The invention relates to the technical field of data processing, in particular to an application education enterprise research data analysis method and system. According to the method, firstly, employment post information investigation is performed on an enterprise, investigation data samples are obtained, then regions, occupations, posts, occupational roles, action fields and work situations in the investigation data samples are clustered, localization characteristics of enterprises among different regions are fully considered in the further deep analysis process of the work situations, and the work situations of the enterprises among different regions are analyzed. Through transverse and longitudinal comparison, differences of different posts and different occupations under different real work tasks and links are extracted, so that analysis of enterprise investigation data is more real, the accuracy and the structured degree of investigation content are enhanced, and a sufficient basis is further provided for talent cultivation targets and processes of application education.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for analyzing enterprise survey data of applied education. Background Art

[0002] Applied education is an education model oriented towards practical application, emphasizing the combination of theoretical knowledge and practical skills to cultivate students' ability to solve practical problems. Enterprises are always the important subjects of applied education. Scientific talent training plans and theoretical support are formulated according to the action fields, work situations, and work tasks corresponding to different enterprises. Among them, enterprise research is an important way for enterprises to participate in the construction of applied education majors. Through enterprise research, we can fully understand the actual needs of enterprises, so as to adjust and optimize the curriculum setting, ensure that the applied talents cultivated can meet the needs of the market, and at the same time improve the social adaptability of applied education.

[0003] In the process of analyzing enterprise survey data, existing methods usually directly use clustering algorithms to determine the core employment positions of the surveyed enterprises, clarify the main professional roles in each position, summarize the typical action fields and typical work processes of each main professional role, and finally sort out the professional standards of each typical work link to complete the analysis of the survey data. However, when a certain enterprise to be surveyed has businesses in different cities at the same time, this analysis method has certain limitations. The local characteristics of different cities will have a certain impact on the actual work process of the enterprise, which is mainly reflected in the changes in real work tasks and real work links, which will lead to errors in the cluster analysis results of the survey data (typical work tasks, typical work links, typical professional requirements, typical work standards), affecting the accuracy of the subsequent talent training plan and the degree of matching between talents and positions. Summary of the invention

[0004] In order to solve the technical problem that the local characteristics of different regions are not considered in the applied education research process, resulting in errors in the cluster analysis results of the research data, the purpose of the present invention is to provide an applied education enterprise research data analysis method and system, and the technical solution adopted is as follows: In a first aspect, a method for analyzing enterprise survey data of applied education is provided, the method comprising: Step S1: Conduct a job information survey on an enterprise to obtain survey data samples; Step S2: Classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples using the K-means clustering algorithm; Step S3: Based on different work situation types, the real work situation in the same area is obtained through a hierarchical clustering algorithm; Step S4: construct a bag-of-words model for each real work situation in the same region, and obtain a typical work situation in the region based on the similarities of keywords between different real work situations in the same region; Step S5: Compare the common words and unique words in the typical work situations of any two regions to obtain the typical work characteristics and localized characteristics of the occupational role.

[0005] Furthermore, the step S1 specifically includes: Conduct a survey on the region, occupation, position, professional role, action field, and work situation information of the same enterprise through questionnaires; Collect survey questionnaires, clean up missing and abnormal survey data, and standardize the survey data; Get a sample of survey data on your business.

[0006] Furthermore, the standardization of the survey data includes: replacing and unifying synonyms and antonyms in the survey data.

[0007] Further, in step S3 and step S4: The real work situation includes four dimensions: real work tasks, real work links, real professional requirements, and real work standards; The typical work situation includes typical work tasks, typical work links, typical occupational requirements, and typical work standards corresponding to actual work situations.

[0008] Furthermore, step S3 is specifically as follows: S31: construct a bag-of-words model for samples of different work situation types in a region, and obtain the similarity between the two samples in real work tasks according to the number of words in the real work tasks in any two samples and the number of common words in the two samples; S32: Repeat step S31 to obtain the similarity between all samples in real work tasks and construct a similarity matrix; S33: using a hierarchical clustering algorithm to sequentially classify the two samples with the highest similarity in the similarity matrix into a cluster, and obtaining an optimal clustering hierarchy, wherein each cluster in the optimal clustering hierarchy represents a real work task in the region; S34: Repeat steps S31 to S33, traverse other dimensions of the real work situation, and obtain each cluster in the best clustering level of the real work links, real professional requirements, and real work standards.

[0009] Furthermore, step S4 specifically includes: S41: Build a bag-of-words model for each real work task in the same region; S42: Obtain a key index of a word according to the average occurrence frequency of the word in each enterprise survey data sample, and when the key index of the word is greater than a preset first threshold, the word is considered to be a keyword; S43: comparing and extracting similarities between keywords of real work tasks in the same region, and obtaining typical work tasks corresponding to the real work tasks; S44: Repeat steps S41 to S43, traverse other dimensions of the actual work situation, and obtain typical work links, typical occupational requirements, and typical work standards in the region.

[0010] Furthermore, in step S5, the typical work characteristics include typical work tasks, typical work links, typical occupational requirements, and typical work standards common to the same work situations in all regions; the localized characteristics include typical work tasks, typical work links, typical occupational requirements, and typical work standards unique to the same work situation in each region.

[0011] Furthermore, the typical work characteristics of the occupational role obtained in step S5 specifically include: Step S501: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S502: Compare the data samples of the two regions to obtain common words; Step S503: Obtain the key index of the common words in the two regions according to the key index average of each common word in the two regions; Step S504: repeat steps S501-S503, traverse any two other regions, and obtain key indexes of common words in all regions; Step S505: presetting a second threshold value of the key index of the common words in all regions, and when the key index of the common words in all regions is greater than the preset second threshold value, it is considered to be a typical work pronoun; Step S506: Acquire typical work characteristics of the occupational role according to the semantics of typical work pronouns.

[0012] Furthermore, the step S5 of obtaining the localized features of the professional role specifically includes: Step S511: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S512: Compare the data sample differences between the two regions, obtain unique words, and traverse any other two regions to obtain unique words for each region; Step S513: presetting a third threshold of the key index of any region-specific vocabulary. When the key index of any region-specific vocabulary is greater than the preset third threshold, it is considered to be a typical work pronoun in the region. Step S514: According to the typical work pronoun semantics of the region, the localized characteristics of the occupational role in the region are obtained.

[0013] In another aspect, the present invention provides an enterprise survey data analysis system for applied education, the system comprising: The data acquisition module is used to obtain survey data samples based on the enterprise's employment information survey; The data preliminary classification module is used to classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples through the K-means clustering algorithm; The real work situation acquisition module is used to obtain the real work situation in the same area based on different work situation types through a hierarchical clustering algorithm; The typical work situation acquisition module is used to construct a bag-of-words model for each real work situation in the same region, and obtain the typical work situation in the region based on the similarities of keywords between different real work situations in the same region; A typical work feature acquisition module is used to acquire the typical work features of the occupational role based on the comparison of common words in typical work situations in any two regions; The localization feature acquisition module is used to obtain the localization features of the professional role based on the unique vocabulary of any region.

[0014] The present invention has the following beneficial effects: The present invention can obtain information such as the job group of a certain profession and the corresponding professional role by clustering the enterprise survey data, thus providing accurate guidance and support for the formulation of talent training plans. At the same time, in the process of cluster analysis, the localization characteristics of enterprises in different regions are fully considered, and the differences of different positions and professions under different real work tasks and links are extracted through horizontal and vertical comparisons, making the analysis of enterprise survey data more realistic, and enhancing the accuracy and structured degree of the survey content, further providing sufficient basis for the talent training goals and processes of applied education. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1A flow chart of a method for analyzing enterprise survey data for applied education provided by one embodiment of the present invention.

[0017] Figure 2 This is a corporate occupation structure tree after preliminary classification of the survey data samples provided by one embodiment of the present invention.

[0018] Figure 3 A schematic diagram of a clustering tree obtained by optimal hierarchical clustering of samples provided in one embodiment of the present invention.

[0019] Figure 4 A block diagram of an enterprise survey data analysis system for applied education provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the enterprise survey data analysis method and system for applied education proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. Unless otherwise defined, all technical and scientific terms used in this article have the same meaning as those commonly understood by technicians in the technical field of the present invention.

[0021] The scenario targeted by the present invention is: for the same enterprise, the same position in different cities may have some differences in the data such as workplace, work situation, and work tasks, that is, the localization differences caused by the market, culture, and policies of different cities. For example, the chef of a catering company, whose professional role is to cook, but for different dishes (work tasks), there may be certain changes in eating habits due to regional differences, so there are differences in the production of different dishes (real work links), for example, less pepper is put in areas where spicy food cannot be eaten, and pork is not used in some areas, that is, there are localization differences in different work tasks and work links in the same position. At this time, if the differences caused by localization of the same position are ignored when conducting data research and analysis on the enterprise and planning talent training, serious errors will occur during the enterprise research data analysis, which will lead to errors in the content of typical work tasks, typical work links, etc., and ultimately cause the talent training program of applied education to be mismatched with the actual needs of the enterprise. Therefore, by eliminating or specifically considering the localization characteristics of the same position in different regions, and then obtaining various typical characteristics of the current position, the accuracy of subsequent data analysis can be improved, and the quality of subsequent talent training can be improved.

[0022] The specific scheme of the enterprise survey data analysis method and system for applied education provided by the present invention is described in detail below with reference to the accompanying drawings.

[0023] First, see Figure 1 , which shows a flow chart of a method for analyzing enterprise survey data of applied education provided by an embodiment of the present invention, the method comprising the following steps: Step S1: Conduct a job information survey on an enterprise and obtain survey data samples.

[0024] Among them, step S1 specifically includes: conducting a survey on the region, occupation, position, professional role, action field, and work situation information of the same enterprise in the form of a survey questionnaire; collecting the survey questionnaires, cleaning up missing and abnormal survey data, and standardizing the survey data; obtaining survey data samples for the enterprise.

[0025] More specifically, we first select the enterprise samples for investigation and formulate the survey questionnaire. In the process of enterprise investigation, we mainly formulate the survey questionnaire according to the survey logic of "enterprise-occupation-position-occupational role-action field-work situation", that is, first determine the distribution area of ​​the investigated enterprise, then clarify the occupations in each area of ​​the enterprise, further clarify the specific positions of these occupations, and then clarify the main occupational roles in each position, and then summarize the typical action field and specific work situation of each main occupational role; then, summarize the collected questionnaires, and perform data cleaning to remove missing and abnormal data samples; further, standardize the data of different dimensions, such as replacing and unifying the synonyms and synonyms in the survey data, so as to facilitate the subsequent cluster analysis; finally, store the data obtained from the enterprise investigation in the designated location to complete the acquisition of the survey data sample. It should be noted here that the action field refers to the physical and psychological fields required for each occupational role to complete the work tasks.

[0026] It should be noted that before conducting enterprise research, it is usually necessary to demonstrate the necessity of a certain major, that is, to analyze and research the current and future talent needs of the corresponding professional positions in the enterprise and the society's recognition of it. Generally, the greater the demand and the higher the recognition, the more sufficient the necessity demonstration of the major and the higher the necessity of enterprise research.

[0027] Step S2: Classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples using the K-means clustering algorithm.

[0028] Taking the occupations in the survey data sample as an example, the K value of K-means clustering is taken as the maximum number of occupational categories in the current enterprise. When the occupations in the questionnaire are the same, they are considered to belong to the same cluster.

[0029] It should be further explained that the occupations, positions, professional roles, action fields, and work situations in the survey data samples are in a progressive form. For example, for the action field, the preceding data (occupation, position, professional role) must remain consistent to belong to the same cluster. In some companies, an employee may have multiple occupations. Therefore, after completing a clustering, the survey data samples of the same cluster are marked and put back into the database for clustering again. That is, a survey data sample can participate in multiple clustering processes at the same level.

[0030] According to the above clustering results, the structure tree of the current enterprise is constructed as shown in the attached figure. Figure 2 It should be noted that the attached Figure 2 The dashed line indicates that there is no correlation, the solid line indicates that there is a clear correlation, and the solid line plus the dashed line indicates that there is a certain correlation.

[0031] Since the localization characteristics of the survey data samples will affect the actual working conditions of the current enterprises in the local area (real work tasks, real work links, etc.), this impact has both the same parts and changes according to different tasks. For example, catering enterprises in some areas do not have pork options when preparing dishes, side dishes, and cooking (common impact), but the impact of the lack of pork in different real work tasks and real work links is also different (individual impact), for example, the staff who cut vegetables no longer need to cut pork, the staff who prepare side dishes need to replace pork with other meats, and the staff who cook dishes need to pay attention to how to cook the replaced meat, etc. Therefore, this embodiment further sets the following steps.

[0032] Step S3: Based on different work situation types, the real work situation in the same area is obtained through a hierarchical clustering algorithm.

[0033] Among them, the real working situation includes four dimensions: real work tasks, real work links, real professional requirements, and real work standards.

[0034] Step S3 specifically includes: S31: construct a bag-of-words model for samples of different work situation types in a region, and obtain the similarity between the two samples in real work tasks according to the number of words in the real work tasks in any two samples and the number of common words in the two samples; S32: Repeat step S31 to obtain the similarity between all samples in real work tasks and construct a similarity matrix; S33: using a hierarchical clustering algorithm to sequentially classify the two samples with the highest similarity in the similarity matrix into a cluster, and obtaining an optimal clustering hierarchy, wherein each cluster in the optimal clustering hierarchy represents a real work task in the region; S34: Repeat steps S31 to S33, traverse other dimensions of the real work situation, and obtain each cluster in the best clustering level of the real work links, real professional requirements, and real work standards.

[0035] More specifically, the regions to which the survey data samples belong are first classified, and the survey data samples belonging to the same region are divided into one category. In step S31, a similarity calculation formula between two samples in a certain dimension of the real working situation is constructed as follows: In the formula, and Respectively represent the number of words in the real work task of the i-th sample and the j-th sample. Obviously, and All are greater than 0; represents the number of identical words between the i-th sample and the j-th sample, It represents the similarity between the i-th sample and the j-th sample in the real working situation in dimension z.

[0036] In the above-mentioned similarity calculation formula between the two samples in a certain dimension of the real working situation, the ratio of the number of the same words between the i-th sample and the j-th sample to the total number of words in the real working tasks of the i-th sample and the j-th sample represents the similarity between the two samples in a certain dimension of the real working situation.

[0037] Then, repeat the above calculation process, traverse all samples, obtain the similarity between all samples in the zth dimension, and construct the similarity matrix as follows: In the formula, It indicates the similarity between sample 1 and sample 2 in the z dimension of the real working situation; Indicates the similarity between sample 2 and sample 1 in the real working situation z dimension; Indicates the similarity between the i-th sample and sample 1 in the real working situation z dimension; It indicates the similarity between sample 1 and sample 2 in the real working situation z dimension; It indicates the similarity between sample 2 and sample 2 in the z dimension of the real working situation; Indicates the similarity between the i-th sample and sample 2 in the real working situation z dimension; Indicates the similarity between sample 1 and the jth sample in the real working situation z dimension; Indicates the similarity between sample 2 and the jth sample in the real working situation z dimension; Indicates the similarity between the i-th sample and the j-th sample in the real working situation z dimension; Represents the similarity matrix of the real working situation in the z dimension.

[0038] The similarity matrix is ​​processed using a hierarchical clustering algorithm, that is, the two samples with the highest similarity are classified into a cluster in turn, and a hierarchical clustering tree is constructed, as shown in the attached figure. Figure 3 As shown in the figure, a character represents a node, that is, a sample of different work situation types in a region in a corporate survey data. In this way, the best clustering level is obtained, and each cluster in the best clustering level represents each real work situation. Taking the real work task dimension of the real work situation as an example, each cluster in the best clustering level represents each real work task. By traversing other dimensions of the real work situation, each cluster in the best clustering level of the real work link, real professional requirements, and real work standards can be obtained.

[0039] After obtaining each real working situation in the survey data sample, it is necessary to summarize and extract each real working situation to further obtain the typical working situation in the survey data sample. Therefore, this embodiment further sets the following steps.

[0040] Step S4: construct a bag-of-words model for each real working situation in the same region, and obtain the typical working situation in the region based on the similarities of keywords between different real working situations in the same region.

[0041] The typical working situation includes typical working tasks, typical working links, typical professional requirements, and typical working standards corresponding to the real working situation. That is, relative to the real working tasks, real working links, real professional requirements, and real working standards in the real working situation, the typical working situation corresponds to typical working tasks, typical working links, typical professional requirements, and typical working standards.

[0042] Step S4 specifically includes: S41: Build a bag-of-words model for each real work task in the same region; S42: Obtain a key index of a word according to the average occurrence frequency of the word in each enterprise survey data sample, and when the key index of the word is greater than a preset first threshold, the word is considered to be a keyword; S43: comparing and extracting similarities between keywords of real work tasks in the same region, and obtaining typical work tasks corresponding to the real work tasks; S44: Repeat steps S41 to S43, traverse other dimensions of the actual work situation, and obtain typical work links, typical occupational requirements, and typical work standards in the region.

[0043] More specifically, since keywords can be used to represent typical information of the current real work task, we first construct a bag-of-words model for each real work task in the same region, extract keywords for each real work task based on the bag-of-words model, and construct the key index calculation formula for each word as follows: In the formula, Indicates the number of enterprise survey data samples with the u-th word; Indicates the total number of enterprise survey data samples; represents the average frequency of occurrence of the u-th word in each enterprise survey data sample; Represents the key index of the u-th word.

[0044] In the above formula for calculating the key index of each word, there is a ratio of the number of enterprise survey data samples for the uth word to the total number of enterprise survey data samples. , which is positively correlated with the key index of the u-th word; the average frequency of the u-th word in each enterprise survey data sample , and its key index of the u-th word is positively correlated.

[0045] The first threshold of the key index is preset to 0.6. When the key index of the uth word , it is considered a keyword, otherwise, it is a non-keyword. Then, the similarities of keywords between different real work tasks are compared, and the similarities are extracted, that is, the typical work tasks in the area. For example, a real work task is "later, stir-fry a slightly spicy mapo tofu, without chopped green onions". If "stir-fry" and "mapo tofu" are also used as keywords in other work tasks, then "stir-fry mapo tofu" is a typical work task.

[0046] Finally, repeat the above steps to traverse other dimensions of the real work situation and obtain the typical work links, typical occupational requirements, and typical work standards in the area.

[0047] After obtaining the typical working conditions of each region, it is necessary to extract local features of the typical working conditions of each region, and further obtain typical working features and local features corresponding to the typical working conditions of all regions in the survey data sample. Therefore, this embodiment further sets the following steps.

[0048] Step S5: Compare the common words and unique words in the typical work situations of any two regions to obtain the typical work characteristics and localized characteristics of the occupational role.

[0049] Among them, typical work characteristics include the typical work tasks, typical work links, typical occupational requirements, and typical work standards common to the same work situations in all regions; localized characteristics include the typical work tasks, typical work links, typical occupational requirements, and typical work standards unique to the same work situations in each region.

[0050] The typical work characteristics of the occupational role obtained in step S5 specifically include: Step S501: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S502: Compare the data samples of the two regions to obtain common words; Step S503: Obtain the key index of the common words in the two regions according to the key index average of each common word in the two regions; Step S504: repeat steps S501-S503, traverse any two other regions, and obtain key indexes of common words in all regions; Step S505: presetting a second threshold value of the key index of the common words in all regions, and when the key index of the common words in all regions is greater than the preset second threshold value, it is considered to be a typical work pronoun; Step S506: Acquire typical work characteristics of the occupational role according to the semantics of typical work pronouns.

[0051] More specifically, firstly, a data sample of typical working conditions in the same working situation in any two regions is constructed, where the typical working conditions include typical working tasks, typical working links, typical professional requirements, and typical working standards. Then, the data sample differences between the two regions are compared to obtain the common words. Furthermore, the key index mean calculation formula for each common word in the two regions is constructed as follows: In the formula, represents the key index of vocabulary u in region 1, The key index of vocabulary u in region 2; Represents the average key index of the common vocabulary u in regions 1 and 2.

[0052] The average key index of the common vocabulary u in regions 1 and 2 mentioned above The key index of the common words in the two regions is represented. Repeat the above steps, traverse any two other regions, and obtain the key index of each common word in all regions. The larger the key index of a common word in all regions, the more typical the working situation represented by the word. The second threshold of the key index of the common words in all regions is preset to 0.6. Finally, based on the semantics of typical work pronouns, we obtain the typical work characteristics of the occupational role, and count and analyze all typical work pronouns, that is, based on the known semantics of typical work pronouns, we can deduce the typical work tasks, typical work links, typical occupational requirements, and typical work standards in the same work situations in all regions.

[0053] On the other hand, the localization characteristics of the occupational role obtained in step S5 specifically include: Step S511: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S512: Compare the data sample differences between the two regions, obtain unique words, and traverse any other two regions to obtain unique words for each region; Step S513: presetting a third threshold of the key index of any region-specific vocabulary. When the key index of any region-specific vocabulary is greater than the preset third threshold, it is considered to be a typical work pronoun in the region. Step S514: According to the typical work pronoun semantics of the region, the localized characteristics of the occupational role in the region are obtained.

[0054] More specifically, first, we construct data samples of typical working conditions in the same working situation in any two regions, compare the differences in the data samples of the two regions, obtain unique words, traverse any two other regions, obtain the unique words of any region, and when the unique words are in the key index of the region where they are located The larger the value, the more localized the task or link represented by the word. Therefore, the third threshold of the key index of any region-specific word is preset to 0.8. When the key index of any region-specific word is greater than 0.8, it is considered to be a typical work pronoun in the region. Finally, based on the semantics of the typical work pronouns in the region, the typical work tasks, typical work links, typical professional requirements, and typical work standards in the region under the same work situation of the occupational role are obtained.

[0055] In this way, we obtained the typical work characteristics such as typical work tasks, typical work links, typical occupational requirements, and typical work standards that are common in the same work situations in all regions; at the same time, we obtained the localized characteristics such as typical work tasks, typical work links, typical occupational requirements, and typical work standards that are unique to the same work situation in each region. So far, we have completed the analysis of the survey data sample of the enterprise, and obtained the regional distribution of the enterprise, the occupational composition of each region, the job groups of each occupation, the occupational roles under specific positions, the action fields under each occupational role, the specific work situations of each action field, and the typical work tasks, typical work links, typical occupational requirements, and typical work standards under each work situation.

[0056] It should be further explained that this embodiment can be expanded to obtain industry groups corresponding to enterprises and industrial groups corresponding to various industries for analysis. The specific survey data analysis method is similar and will not be elaborated in detail in this embodiment.

[0057] Furthermore, through the above steps, we can complete the classification of survey data of the same enterprise, the analysis of work characteristics of all regions under the same work situation of the same professional role, and the analysis of localized characteristics of different regions under the same work situation of the same professional role. Based on this, we can develop corresponding talent training programs in the field of applied education. The specific implementation steps are as follows: First, according to the results of the division of job groups, determine the grouping logic of professional groups, and then divide professional groups and professional directions; then, according to the work content and nature of the positions, divide the talent training levels; further, according to the number of graduates, retirees, unemployed people, population growth, population mobility and other data, predict the future supply and demand of labor, so as to determine the scale of talent training; finally, according to job standards, work needs and regional characteristics, determine talent training goals and methods.

[0058] In the second aspect, this embodiment provides an enterprise survey data analysis system for applied education, see Figure 4 , which shows a block diagram of an enterprise survey data analysis system for applied education provided by an embodiment of the present invention, the system comprising: The data acquisition module 101 is used to obtain survey data samples according to the enterprise's job information survey; The data preliminary classification module 102 is used to classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples by using a K-means clustering algorithm; The real work situation acquisition module 103 is used to acquire the real work situation in the same area through a hierarchical clustering algorithm based on different work situation types; The typical working situation acquisition module 104 is used to construct a bag-of-words model for each real working situation in the same region, and acquire the typical working situation in the region based on the similarities of keywords between different real working situations in the same region; A typical work feature acquisition module 105 is used to acquire the typical work features of the occupational role based on the comparison of common words in typical work situations of any two regions; The localization feature acquisition module 106 is used to acquire the localization features of the professional role according to the specific vocabulary of any region.

[0059] The present embodiment provides a method and system for analyzing enterprise survey data of applied education. Through a clustering algorithm, the survey data of the same enterprise is classified into regions, action fields, work situations, etc., and the actual working conditions under different work situations are further analyzed, so as to obtain the localized characteristics in the enterprise survey data, and sort out the typical work characteristics such as typical work tasks, typical work links, typical occupational requirements, and typical work standards shared by different regions. The analysis results of these enterprise survey data improve the accuracy of the clustering analysis results of the survey data, and can be used as a reference for the subsequent formulation of talent training plans, thereby improving the matching degree between talent training and enterprise positions in applied education.

[0060] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The embodiments described above 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, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the 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 method for analyzing enterprise survey data for applied education, characterized in that: The method comprises: Step S1: Conduct a job information survey on an enterprise to obtain survey data samples; Step S2: Classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples using the K-means clustering algorithm; Step S3: Based on different work situation types, the real work situation in the same area is obtained through a hierarchical clustering algorithm; Step S4: construct a bag-of-words model for each real work situation in the same region, and obtain a typical work situation in the region based on the similarities of keywords between different real work situations in the same region; Step S5: Compare the common words and unique words in the typical work situations of any two regions to obtain the typical work characteristics and localized characteristics of the occupational role.

2. The enterprise survey data analysis method for applied education according to claim 1 is characterized in that: The step S1 specifically includes: Conduct a survey on the region, occupation, position, professional role, action field, and work situation information of the same enterprise through questionnaires; Collect survey questionnaires, clean up missing and abnormal survey data, and standardize the survey data; Get a sample of survey data on your business.

3. The enterprise survey data analysis method for applied education according to claim 2 is characterized in that: The standardization of the survey data includes: replacing and unifying synonyms and antonyms in the survey data.

4. The enterprise survey data analysis method for applied education according to claim 1 is characterized in that: In step S3 and step S4: The real work situation includes four dimensions: real work tasks, real work links, real professional requirements, and real work standards; The typical work situation includes typical work tasks, typical work links, typical occupational requirements, and typical work standards corresponding to actual work situations.

5. The enterprise survey data analysis method for applied education according to claim 4 is characterized in that: Step S3 is specifically as follows: S31: construct a bag-of-words model for samples of different work situation types in a region, and obtain the similarity between the two samples in real work tasks according to the number of words in the real work tasks in any two samples and the number of common words in the two samples; S32: Repeat step S31 to obtain the similarity between all samples in real work tasks and construct a similarity matrix; S33: using a hierarchical clustering algorithm to sequentially classify the two samples with the highest similarity in the similarity matrix into a cluster, and obtaining an optimal clustering hierarchy, wherein each cluster in the optimal clustering hierarchy represents a real work task in the region; S34: Repeat steps S31 to S33, traverse other dimensions of the real work situation, and obtain each cluster in the best clustering level of the real work links, real professional requirements, and real work standards.

6. The enterprise survey data analysis method for applied education according to claim 4 is characterized in that: Step S4 specifically includes: S41: Build a bag-of-words model for each real work task in the same region; S42: Obtain a key index of a word according to the average occurrence frequency of the word in each enterprise survey data sample, and when the key index of the word is greater than a preset first threshold, the word is considered to be a keyword; S43: comparing and extracting similarities between keywords of real work tasks in the same region, and obtaining typical work tasks corresponding to the real work tasks; S44: Repeat steps S41 to S43, traverse other dimensions of the actual work situation, and obtain typical work links, typical occupational requirements, and typical work standards in the region.

7. The enterprise survey data analysis method for applied education according to claim 1 is characterized in that: In step S5, the typical working characteristics include: Typical work tasks, typical work links, typical professional requirements, and typical work standards common to the same work situations in all regions; The localized characteristics include typical work tasks, typical work links, typical professional requirements, and typical work standards that are unique to the same work situation in each region.

8. The enterprise survey data analysis method for applied education according to claim 7 is characterized in that: The typical work characteristics of the occupational role obtained in step S5 specifically include: Step S501: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S502: Compare the data samples of the two regions to obtain common words; Step S503: Obtain the key index of the common words in the two regions according to the key index average of each common word in the two regions; Step S504: repeat steps S501-S503, traverse any two other regions, and obtain key indexes of common words in all regions; Step S505: presetting a second threshold value of the key index of the common words in all regions, and when the key index of the common words in all regions is greater than the preset second threshold value, it is considered to be a typical work pronoun; Step S506: Acquire typical work characteristics of the occupational role according to the semantics of typical work pronouns.

9. The enterprise survey data analysis method for applied education according to claim 7 is characterized in that: The step S5 of obtaining the localized features of the professional role specifically includes: Step S511: constructing data samples of typical working conditions in any two regions under the same working situation of a professional role; Step S512: Compare the data sample differences between the two regions, obtain unique words, and traverse any other two regions to obtain unique words for each region; Step S513: presetting a third threshold of the key index of any region-specific vocabulary. When the key index of any region-specific vocabulary is greater than the preset third threshold, it is considered to be a typical work pronoun in the region. Step S514: According to the typical work pronoun semantics of the region, the localized characteristics of the occupational role in the region are obtained.

10. An enterprise survey data analysis system for applied education, characterized in that: The system comprises: The data acquisition module is used to obtain survey data samples based on the enterprise's employment information survey; The data preliminary classification module is used to classify the regions, occupations, positions, professional roles, action fields, and work situations in the survey data samples through the K-means clustering algorithm; The real work situation acquisition module is used to obtain the real work situation in the same area based on different work situation types through a hierarchical clustering algorithm; The typical work situation acquisition module is used to construct a bag-of-words model for each real work situation in the same region, and obtain the typical work situation in the region based on the similarities of keywords between different real work situations in the same region; A typical work feature acquisition module is used to acquire the typical work features of the occupational role based on the comparison of common words in typical work situations in any two regions; The localization feature acquisition module is used to obtain the localization features of the professional role based on the unique vocabulary of any region.

Images