Personalized nurse training management method based on data mining

Abstract

The present invention discloses a personalized nurse training management method based on data mining, and specifically relates to the technical field of training management, including constructing a knowledge graph of nurse training content, analyzing user behavior information according to user query history data, analyzing current demand information for different types of nurse talents, expressing the importance of users to different nurse training contents through content evaluation coefficients, determining user preferences for different nurse training contents in different monitoring periods, performing cluster analysis on content evaluation coefficients in different monitoring periods, generating content recommendation scores for different nurse training contents, determining nurse training contents recommended to users according to the content recommendation scores for different nurse training contents, monitoring recommended contents through a logistic regression model, and judging the accuracy of recommended contents. The present invention helps to optimize the recommendation and management of nurse training contents, and improve user satisfaction and training effects.

Description

Technical Field

[0001] The present invention relates to the technical field of training management, and more specifically, to a personalized nurse training management method based on data mining. Background Art

[0002] In the modern medical environment, the training and development of nurses is crucial because their professional knowledge and skills directly affect the quality of patient care and the efficiency of medical services. However, effectively managing and optimizing nurse training content and its recommendations is a complex issue that requires a combination of user behavior analysis, nurse type demand analysis, and advanced data analysis technology to solve;

[0003] Traditional training content and nurse skills information may be scattered across different systems and documents, lacking relevance. In addition, traditional nurse training usually adopts a unified training plan, ignoring individual differences and users' specific needs as well as dynamic changes over time, resulting in the nurse training content recommended to different users being inaccurate and incomplete.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a personalized nurse training management method based on data mining to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The personalized nurse training management method based on data mining specifically includes the following steps:

[0008] S1: Build a knowledge graph of nurse training content, analyze user behavior information based on user query history data, and analyze the current demand information for different types of nurses based on the types of nurses that can be trained by the nurse training content and the demand for nurses of different types;

[0009] S2: Quantify the importance users place on different nurse training contents, express the importance of users on different nurse training contents through content evaluation coefficients, and determine users’ preferences for different nurse training contents in different monitoring periods;

[0010] S3: cluster analysis is performed on the content evaluation coefficients in different monitoring cycles to obtain the content evaluation coefficients at the cluster center with the minimum time difference between the current monitoring cycle and the cluster center and the minimum time difference, and generate content recommendation scores for different nurse training contents;

[0011] S4: Determine the nurse training content recommended to the user based on the content recommendation scores of different nurse training contents, and monitor the recommended content through a logistic regression model to determine the accuracy of the recommended content;

[0012] Among them, the knowledge graph of nurse training content is constructed, and the user's behavior information is analyzed, including:

[0013] The user's behavior information in the application or website is analyzed to obtain the content preference weight coefficient and the content deviation coefficient, and the content preference weight coefficient and the content deviation coefficient are marked as and , where n = 1, 2, 3, …, N, N is a positive integer, and n represents different types of nurse training content in the application or website;

[0014] Construct a knowledge graph of nurse training content, which helps to systematically organize and manage training content by subdividing various types of training content and defining their relationships. In the knowledge graph, nodes represent entities including nursing skills and knowledge points, and edges represent the relationship between entities. Each edge has a weight, which represents the distance or correlation between two nodes.

[0015] Through the knowledge graph, the user's query content is mapped to the corresponding training content category, and the path query is used to determine the distance between the user's query content and different nurse training content, and the monitoring cycle is set;

[0016] Through the path query algorithm, the shortest path between the user query content and the nurse training content within the monitoring period is obtained, and the shortest path between the user query content and the nurse training content within the monitoring period is marked as: , where i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the user's query content within the monitoring period;

[0017] Calculate the content preference weight coefficient using the following formula: ;in, is the content preference weight coefficient of the nth type of nurse training content;

[0018] Set the shortest path threshold, compare the shortest path between the user query content and the nurse training content within the monitoring period with the shortest path threshold, retain the user query content that is less than the shortest path threshold, and re-mark the shortest path of the user query content that is less than the shortest path threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the user query content that is less than the shortest path threshold;

[0019] Calculate the content deviation coefficient using the following formula: ;in, is the content deviation coefficient of the nth type of nurse training content, is the shortest path threshold.

[0020] In a preferred embodiment, the current demand information for different types of nurses is analyzed, including:

[0021] According to the current demand information of different types of nurse talents, the content importance coefficient is obtained and marked as ;

[0022] The logic for obtaining the content importance coefficient is: obtaining the types of nurses trained by different nurse training contents during the monitoring period, and obtaining the demand for different types of nurses during the monitoring period, and marking the demand for different types of nurses during the monitoring period as: , where p = 1, 2, 3, ..., P, P is a positive integer, and p is the number of different nurse types;

[0023] Obtain the proportion of different nurse training content requirements and mark them as follows: ,in, , where The demand for training different types of nurses with different nurse training contents, m=1, 2, 3, ..., M, M is a positive integer, and m is the number of different types of nurses trained with different nurse training contents;

[0024] Get the rarity of different nurse training contents and mark the rarity of different nurse training contents as: ,in, , where is an indicator function. When the type of nurses trained by a certain nurse training content appears in other nurse training contents, =1, when the type of nurses trained in a certain nurse training content does not appear in other nurse training contents, then is 0;

[0025] Calculate the importance of different nurse types trained by different nurse training contents, and label the importance of different nurse types trained by different nurse training contents as: ,in, ;

[0026] Calculate the content importance coefficient, the calculation formula is: ,in, is the content importance coefficient of the nth type of nurse training content.

[0027] In a preferred embodiment, quantifying the importance of users on different nurse training contents includes:

[0028] The content preference weight coefficient, content deviation coefficient and content importance coefficient are weighted and calculated to generate the content evaluation coefficient. The calculation formula of the content evaluation coefficient is: ;in, is the content evaluation coefficient of the nth type of nurse training content during the monitoring period, , , are the proportional coefficients of content preference weight coefficient, content deviation coefficient, and content importance coefficient, respectively. , , Greater than 0.

[0029] In a preferred embodiment, cluster analysis is performed on content evaluation coefficients in different monitoring periods, including:

[0030] Set the monitoring interval, which consists of multiple monitoring cycles. The monitoring interval is marked as: T, ,in, represents the rth monitoring period of the nth nurse training content, r=1, 2, 3, ... R, R is a positive integer, and r is the number of the monitoring period included in the monitoring interval;

[0031] According to the content of the user in the monitoring interval, the content evaluation coefficients in different monitoring periods in the monitoring interval are collected, and the content evaluation coefficients in different monitoring periods and different monitoring periods are used as inputs of the clustering model;

[0032] Use K-means clustering to initialize K initial cluster centers, assign each sample to the nearest cluster center to form K clusters, recalculate the center point of each cluster as the new cluster center, and repeat the steps until the cluster center no longer changes or changes very little;

[0033] The obtained cluster centers are marked as: ,in, represents the monitoring period of the kth cluster center of the nth nurse training content, The content evaluation coefficient of the kth cluster center representing the nth nurse training content is ;

[0034] Get the current monitoring period, subtract the current monitoring period from the monitoring period of the cluster center, get the time difference, and mark the time difference as: ,in, , represents the time difference of the kth cluster center of the nth nurse training content;

[0035] Get the minimum time difference between the current monitoring period and the cluster center, and mark the minimum time difference between the current monitoring period and the cluster center as: , and mark the cluster center with the smallest time difference as: ,in, is the content evaluation coefficient at the cluster center with the smallest time difference;

[0036] The content recommendation score is generated by weighting the content evaluation coefficient at the cluster center with the minimum time difference between the current monitoring period and the cluster center and the minimum time difference.

[0037] In a preferred embodiment, determining the nurse training content recommended to the user includes:

[0038] A content recommendation score threshold is set, and the content recommendation score is compared with the content recommendation score threshold. If the content recommendation score is greater than the content recommendation score threshold, the nurse training content is recorded as recommended content. If the content recommendation score is less than the content recommendation score threshold, the nurse training content is not marked.

[0039] In a preferred embodiment, determining the accuracy of the recommended content includes:

[0040] The content evaluation coefficients of different monitoring periods within the monitoring interval are summed to obtain the total content evaluation coefficients of different nurse training contents within the monitoring interval. The total content evaluation coefficients of different nurse training contents within the monitoring interval are normalized by Min-Max, and the normalized total content evaluation coefficients of different nurse training contents are re-labeled as: ;

[0041] Obtain the proportion of different nurse training contents in the monitoring period within the monitoring interval, and mark the proportion of different nurse training contents in the monitoring period within the monitoring interval as: ;

[0042] The normalized total content evaluation coefficient of different nurse training contents and the proportion of different nurse training contents in the monitoring period are used as the input features of the logistic regression model, and the certainty coefficient is used as the output of the logistic regression model. The logistic regression formula of the certainty coefficient is: ;in, is the certainty coefficient about the user, , , ,……, Indicates the types of training content for different nurses;

[0043] Set a certainty coefficient threshold. If the certainty coefficient is less than the certainty coefficient threshold, the user will be marked as a suspicious user and the monitoring period will be extended. If the certainty coefficient is greater than the certainty coefficient threshold, the recommended content will be used as the user's training target.

[0044] Technical effects and advantages of the present invention:

[0045] The present invention analyzes the user's behavioral information in the application or website and the current demand information for different types of nurse talents, determines the nurse training content that is relatively important to the trained user in the user's query history, quantifies the nurse training content recommended to the user through a clustering model, and monitors through a logistic regression model to determine the accuracy of the recommended content. The present invention helps to optimize the recommendation and management of nurse training content and improve user satisfaction and training effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0047] Figure 1 The figure is a flow chart of the personalized nurse training management method based on data mining of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Example 1

[0050] The present invention provides Figure 1 The flowchart of the personalized nurse training management method based on data mining shown in the figure specifically includes the following steps:

[0051] S1: Build a knowledge graph of nurse training content, analyze user behavior information based on user query history data, and analyze the current demand information for different types of nurses based on the types of nurses that can be trained by the nurse training content and the demand for nurses of different types;

[0052] S2: Quantify the importance users place on different nurse training contents, express the importance of users on different nurse training contents through content evaluation coefficients, and determine users’ preferences for different nurse training contents in different monitoring periods;

[0053] S3: cluster analysis is performed on the content evaluation coefficients in different monitoring cycles to obtain the content evaluation coefficients at the cluster center with the minimum time difference between the current monitoring cycle and the cluster center and the minimum time difference, and generate content recommendation scores for different nurse training contents;

[0054] S4: Determine the nurse training content recommended to users based on the content recommendation scores of different nurse training contents, and monitor the recommended content through a logistic regression model to determine the accuracy of the recommended content.

[0055] In a personalized nurse training application or website, a dedicated ID is assigned to each user being trained to record each user's learning history, areas of interest, and progress, thereby providing a more personalized training program and determining the user's preference for training content by collecting user behavior information in the application or website.

[0056] The user's behavior information in the application or website is analyzed to obtain the content preference weight coefficient and the content deviation coefficient, and the content preference weight coefficient and the content deviation coefficient are marked as and , where n = 1, 2, 3, …, N, N is a positive integer, and n represents different types of nurse training content in the application or website.

[0057] It should be noted that the content of nurse training includes basic nursing skills, clinical specialized skills, advanced nursing skills, communication and collaboration, etc., which are further divided into multiple subcategories under different nurse training contents, and each subcategory contains specific knowledge points and skills.

[0058] Construct a knowledge graph of nurse training content, which helps to systematically organize and manage training content by subdividing various types of training content and defining their relationships. In the knowledge graph, nodes represent entities including nursing skills and knowledge points, and edges represent the relationship between entities. Each edge has a weight, which represents the distance or correlation between two nodes.

[0059] Collect the user's query history data, which should include user ID, query content, query time, etc., and pre-process the query content, including word segmentation, removal of stop words, etc.;

[0060] Through the knowledge graph, the user's query content is mapped to the corresponding training content category, including using natural language processing (NLP) technology to match the user's query content with the nodes in the knowledge graph;

[0061] Use path query to determine the distance between the user query content and different nurse training contents, and evaluate the relevance of the user query content and nurse training content, including using path query algorithms (such as Dijkstra algorithm) to determine the shortest path from the user query content node to different nurse training content nodes, reflecting the distance between the user query content and each nurse training content.

[0062] Set a monitoring period. The monitoring period is a short, specific period of time, usually one day. The monitoring period is used to capture the user's daily activities and interest changes by collecting the nurse training content that the user queries in the application or website during the monitoring period.

[0063] Through the path query algorithm, the shortest path between the user query content and the nurse training content within the monitoring period is obtained, and the shortest path between the user query content and the nurse training content within the monitoring period is marked as: , where i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the user's query content within the monitoring period;

[0064] It should be noted that the shorter the shortest path between the user query content and the nurse training content, the more relevant the user query content is to the nurse training content, and the more times the user queries the content, the greater the user's interest in the nurse training content.

[0065] Calculate the content preference weight coefficient using the following formula: ;in, is the content preference weight coefficient of the nth type of nurse training content.

[0066] It can be seen from the formula that the larger the content preference weight coefficient is, the higher the user's preference for a certain type of nurse training content is, which means that the user's nurse training management should allocate more resources to this type and increase the depth and breadth of the content, including providing more advanced courses and professional knowledge.

[0067] Set the shortest path threshold, compare the shortest path between the user query content and the nurse training content within the monitoring period with the shortest path threshold, retain the user query content that is less than the shortest path threshold, and re-mark the shortest path of the user query content that is less than the shortest path threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the user query content that is less than the shortest path threshold;

[0068] It should be noted that the shortest path threshold is set by professional staff. The shortest path lengths from different user query content to different nurse training content types are usually different. If the shortest path from the user query content to the nurse training content type is less than the shortest path threshold, it means that the user query content and the nurse training content type are strongly correlated.

[0069] Calculate the content deviation coefficient using the following formula: ;in, is the content deviation coefficient of the nth type of nurse training content, is the shortest path threshold.

[0070] From the formula, we can see that the larger the content deviation coefficient is, the higher the frequency of users' searches for a certain type of nurse training content is, and the higher the relevance of each search is, indicating that users have a strong interest in and demand for this type of nurse training content. The number of courses for this type of content can be increased, and more in-depth and advanced training content can be provided to meet the learning needs of users at different levels.

[0071] According to the current demand information of different types of nurse talents, the content importance coefficient is obtained and marked as .

[0072] The logic for obtaining the content importance coefficient is: obtaining the types of nurses trained by different nurse training contents during the monitoring period, and obtaining the demand for different types of nurses during the monitoring period, and marking the demand for different types of nurses during the monitoring period as: , where p = 1, 2, 3, ..., P, P is a positive integer, and p is the number of different nurse types;

[0073] Obtain the proportion of different nurse training content requirements and mark them as follows: ,in, , where The demand for training different types of nurses with different nurse training contents, m=1, 2, 3, ..., M, M is a positive integer, and m is the number of different types of nurses trained with different nurse training contents;

[0074] It should be noted that different nurse training contents may train different types of nurses. For example, clinical nurses and nursing nurses may be trained in the nurse training content of clinical specialized skills, and nursing nurses may be trained in basic nursing skills, but clinical nurses will not be trained. Therefore, the more types of nurses cultivated by the nurse training content, the greater the demand for that type of nurse, which means that the nurse training content is more important.

[0075] Get the rarity of different nurse training contents and mark the rarity of different nurse training contents as: ,in, , where is an indicator function. When the type of nurses trained by a certain nurse training content appears in other nurse training contents, =1, when the type of nurses trained in a certain nurse training content does not appear in other nurse training contents, then is 0;

[0076] It should be noted that the higher the rarity of the type of nurses to be trained in a certain nurse training content, the higher the importance of the nurse training content, because if the same type of nurses can be trained in other types of nurse training content, it means that the nurse training content is substitutable.

[0077] Calculate the importance of different nurse types trained by different nurse training contents, and label the importance of different nurse types trained by different nurse training contents as: ,in, ;

[0078] Calculate the content importance coefficient, the calculation formula is: ,in, is the content importance coefficient of the nth type of nurse training content.

[0079] It can be seen from the formula that the larger the content importance coefficient is, the more important the nurse training content is, and the more priority these training contents should be recommended to users.

[0080] The content preference weight coefficient, content deviation coefficient and content importance coefficient are weighted and calculated to generate the content evaluation coefficient. The calculation formula of the content evaluation coefficient is: ;in, is the content evaluation coefficient of the nth type of nurse training content during the monitoring period, , , are the proportional coefficients of content preference weight coefficient, content deviation coefficient, and content importance coefficient, respectively. , , Greater than 0.

[0081] It can be seen from the formula that the larger the content preference weight coefficient, content deviation coefficient and content importance coefficient, the larger the content evaluation coefficient, which means that the importance of the nth nurse training content to the user during the monitoring period is higher; conversely, the smaller the content preference weight coefficient, content deviation coefficient and content importance coefficient, the smaller the content evaluation coefficient, which means that the importance of the nth nurse training content to the user during the monitoring period is lower.

[0082] Set the monitoring interval, which consists of multiple monitoring cycles. The monitoring interval is marked as: T, ,in, Represents the rth monitoring period of the nth nurse training content, r=1, 2, 3, ...R, R is a positive integer, r is the number of the monitoring period included in the monitoring interval, and the monitoring interval is used to observe the content evaluation coefficients of different monitoring periods in the monitoring interval, which helps to recommend appropriate nurse training content to users. The monitoring interval is set by professional staff.

[0083] According to the content of the user in the monitoring interval, the content evaluation coefficients in different monitoring periods in the monitoring interval are collected, and the content evaluation coefficients in different monitoring periods and different monitoring periods are used as inputs of the clustering model;

[0084] Use K-means clustering to initialize K initial cluster centers, assign each sample to the nearest cluster center to form K clusters, recalculate the center point of each cluster as the new cluster center, and repeat the steps until the cluster center no longer changes or changes very little;

[0085] The obtained cluster centers are marked as: ,in, represents the monitoring period of the kth cluster center of the nth nurse training content, The content evaluation coefficient of the kth cluster center representing the nth nurse training content is ;

[0086] Get the current monitoring period, subtract the current monitoring period from the monitoring period of the cluster center, get the time difference, and mark the time difference as: ,in, , represents the time difference of the kth cluster center of the nth nurse training content;

[0087] Get the minimum time difference between the current monitoring period and the cluster center, that is, The minimum value of , marks the minimum time difference between the current monitoring period and the cluster center as: , and mark the cluster center with the smallest time difference as: ,in, is the content evaluation coefficient at the cluster center with the smallest time difference;

[0088] The content evaluation coefficient at the cluster center with the minimum time difference between the current monitoring period and the cluster center and the minimum time difference is weighted to generate a content recommendation score. The calculation formula for the content recommendation score is: ;in, Score the content recommendation for the nth nurse training content. , is the ratio coefficient of the minimum time difference between the current monitoring period and the cluster center and the content evaluation coefficient at the cluster center with the minimum time difference, , Greater than 0.

[0089] It should be noted that the smaller the time difference between the current monitoring period and the minimum time of the cluster center, the closer the time point when the user is interested in a certain nurse training content is to the current time point, indicating that the cluster center can better reflect the user's true preference for nurse training content. The larger the time difference between the current monitoring period and the minimum time of the cluster center, the farther the time point when the user is interested in a certain nurse training content is from the current time point. Because the time difference is long, the user is likely to lose interest in a certain nurse training content, so the content recommendation score of a certain nurse training content is low;

[0090] The larger the content evaluation coefficient at the cluster center with the smallest time difference, the more important the user is to a certain nurse training content, which means that the cluster center can better reflect the importance of the user to a certain nurse training content. Conversely, it means that the user is less important to a certain nurse training content.

[0091] A content recommendation score threshold is set, and the content recommendation score is compared with the content recommendation score threshold. If the content recommendation score is greater than the content recommendation score threshold, the nurse training content is recorded as recommended content, and a large amount of relevant content is pushed in the personalized nurse training application or website. If the content recommendation score is less than the content recommendation score threshold, the nurse training content is not marked.

[0092] The content evaluation coefficients of different monitoring periods within the monitoring interval are summed to obtain the total content evaluation coefficients of different nurse training contents within the monitoring interval, and the total content evaluation coefficients of different nurse training contents within the monitoring interval are marked as: , the total content evaluation coefficients of different nurse training contents within the monitoring interval are processed by Min-Max normalization, and the normalized total content evaluation coefficients of different nurse training contents are re-labeled as: ;

[0093] Obtain the proportion of different nurse training contents in the monitoring period within the monitoring interval, and mark the proportion of different nurse training contents in the monitoring period within the monitoring interval as: For example, if the content evaluation coefficient of a nurse training content in each monitoring cycle within the monitoring interval is not 0, then the proportion of the nurse training content in the monitoring cycle within the monitoring interval is 1; if the content evaluation coefficient of a nurse training content in half of the monitoring cycles within the monitoring interval is not 0, then the proportion of the nurse training content in the monitoring cycle within the monitoring interval is 0.5;

[0094] The normalized total content evaluation coefficient of different nurse training contents and the proportion of different nurse training contents in the monitoring period are used as the input features of the logistic regression model, and the certainty coefficient is used as the output of the logistic regression model. The logistic regression formula of the certainty coefficient is: ;in, is the certainty coefficient about the user, , , ,……, Indicates the types of different nurse training contents.

[0095] A certainty coefficient threshold is set. If the certainty coefficient is less than the certainty coefficient threshold, the user will be marked as a questionable user, indicating that the user is interested in multiple nurse training contents, and it is not easy to determine which specific nurse training content the user is most interested in. The monitoring period is extended to ensure accuracy. If the certainty coefficient is greater than the certainty coefficient threshold, the recommended content will be used as the user's training goal, and a large number of relevant results of the recommended content will be pushed to the user.

[0096] The present invention analyzes the user's behavioral information in the application or website and the current demand information for different types of nurse talents, determines the nurse training content that is relatively important to the trained user in the user's query history, quantifies the nurse training content recommended to the user through a clustering model, and monitors through a logistic regression model to determine the accuracy of the recommended content. The present invention helps to optimize the recommendation and management of nurse training content and improve user satisfaction and training effects.

[0097] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0098] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0099] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0100] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0101] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0102] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0103] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0104] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A personalized nurse training management method based on data mining, characterized in that: The specific steps include: S1: Build a knowledge graph of nurse training content, analyze user behavior information based on user query history data, and analyze the current demand information for different types of nurses based on the types of nurses that can be trained by the nurse training content and the demand for nurses of different types; S2: Quantify the importance users place on different nurse training contents, express the importance of users on different nurse training contents through content evaluation coefficients, and determine users’ preferences for different nurse training contents in different monitoring periods; S3: cluster analysis is performed on the content evaluation coefficients in different monitoring cycles to obtain the content evaluation coefficients at the cluster center with the minimum time difference between the current monitoring cycle and the cluster center and the minimum time difference, and generate content recommendation scores for different nurse training contents; S4: Determine the nurse training content recommended to the user based on the content recommendation scores of different nurse training contents, and monitor the recommended content through a logistic regression model to determine the accuracy of the recommended content; Among them, the knowledge graph of nurse training content is constructed, and the user's behavior information is analyzed, including: The user's behavior information in the application or website is analyzed to obtain the content preference weight coefficient and the content deviation coefficient, and the content preference weight coefficient and the content deviation coefficient are marked as and , where n = 1, 2, 3, …, N, N is a positive integer, and n represents different types of nurse training content in the application or website; Construct a knowledge graph of nurse training content, which helps to systematically organize and manage training content by subdividing various types of training content and defining their relationships. In the knowledge graph, nodes represent entities including nursing skills and knowledge points, and edges represent the relationship between entities. Each edge has a weight, which represents the distance or correlation between two nodes. Through the knowledge graph, the user's query content is mapped to the corresponding training content category, and the path query is used to determine the distance between the user's query content and different nurse training content, and the monitoring cycle is set; Through the path query algorithm, the shortest path between the user query content and the nurse training content within the monitoring period is obtained, and the shortest path between the user query content and the nurse training content within the monitoring period is marked as: , where i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the user's query content within the monitoring period; Calculate the content preference weight coefficient using the following formula: ;in, is the content preference weight coefficient of the nth type of nurse training content; Set the shortest path threshold, compare the shortest path between the user query content and the nurse training content within the monitoring period with the shortest path threshold, retain the user query content that is less than the shortest path threshold, and re-mark the shortest path of the user query content that is less than the shortest path threshold as: , where j=1, 2, 3, ..., J, J is a positive integer, and j is the number of the user query content that is less than the shortest path threshold; Calculate the content deviation coefficient using the following formula: ;in, is the content deviation coefficient of the nth type of nurse training content, is the shortest path threshold.

2. The personalized nurse training management method based on data mining according to claim 1 is characterized in that: Analyze the current demand information for different types of nurses, including: According to the current demand information of different types of nurse talents, the content importance coefficient is obtained and marked as ; The logic for obtaining the content importance coefficient is: obtaining the types of nurses trained by different nurse training contents during the monitoring period, and obtaining the demand for different types of nurses during the monitoring period, and marking the demand for different types of nurses during the monitoring period as: , where p = 1, 2, 3, ..., P, P is a positive integer, and p is the number of different nurse types; Obtain the proportion of different nurse training content requirements and mark them as follows: ,in, , where The demand for training different types of nurses with different nurse training contents, m=1, 2, 3, ..., M, M is a positive integer, and m is the number of different types of nurses trained with different nurse training contents; Get the rarity of different nurse training contents and mark the rarity of different nurse training contents as: ,in, , where is an indicator function. When the type of nurses trained by a certain nurse training content appears in other nurse training contents, =1, when the type of nurses trained in a certain nurse training content does not appear in other nurse training contents, then is 0; Calculate the importance of different nurse types trained by different nurse training contents, and label the importance of different nurse types trained by different nurse training contents as: ,in, ; Calculate the content importance coefficient, the calculation formula is: ,in, is the content importance coefficient of the nth type of nurse training content.

3. The personalized nurse training management method based on data mining according to claim 2 is characterized in that: Quantify the importance users place on different nurse training components, including: The content preference weight coefficient, content deviation coefficient and content importance coefficient are weighted and calculated to generate the content evaluation coefficient. The calculation formula of the content evaluation coefficient is: ;in, is the content evaluation coefficient of the nth type of nurse training content during the monitoring period, , , are the proportional coefficients of content preference weight coefficient, content deviation coefficient, and content importance coefficient, respectively. , , Greater than 0.

4. The personalized nurse training management method based on data mining according to claim 3 is characterized in that: Cluster analysis of content evaluation coefficients in different monitoring cycles, including: Set the monitoring interval, which consists of multiple monitoring cycles. The monitoring interval is marked as: T, ,in, represents the rth monitoring period of the nth nurse training content, r=1, 2, 3, ... R, R is a positive integer, and r is the number of the monitoring period included in the monitoring interval; According to the content of the user in the monitoring interval, the content evaluation coefficients in different monitoring periods in the monitoring interval are collected, and the content evaluation coefficients in different monitoring periods and different monitoring periods are used as inputs of the clustering model; Use K-means clustering to initialize K initial cluster centers, assign each sample to the nearest cluster center to form K clusters, recalculate the center point of each cluster as the new cluster center, and repeat the steps until the cluster center no longer changes or changes very little; The obtained cluster centers are marked as: ,in, represents the monitoring period of the kth cluster center of the nth nurse training content, The content evaluation coefficient of the kth cluster center representing the nth nurse training content is ; Get the current monitoring period, subtract the current monitoring period from the monitoring period of the cluster center, get the time difference, and mark the time difference as: ,in, , represents the time difference of the kth cluster center of the nth nurse training content; Get the minimum time difference between the current monitoring period and the cluster center, and mark the minimum time difference between the current monitoring period and the cluster center as: , and mark the cluster center with the smallest time difference as: ,in, is the content evaluation coefficient at the cluster center with the smallest time difference; The content recommendation score is generated by weighting the content evaluation coefficient at the cluster center with the minimum time difference between the current monitoring period and the cluster center and the minimum time difference.

5. The personalized nurse training management method based on data mining according to claim 4 is characterized in that: Determine the nurse training content recommended for users, including: A content recommendation score threshold is set, and the content recommendation score is compared with the content recommendation score threshold. If the content recommendation score is greater than the content recommendation score threshold, the nurse training content is recorded as recommended content. If the content recommendation score is less than the content recommendation score threshold, the nurse training content is not marked.

Images