Nursing education resource intelligent recommendation method

Abstract

The invention provides an intelligent recommendation method for nursing education resources, and the method comprises the steps: obtaining the skill level of a worker according to the skill information inputted by the worker who nurses an old person in a nursing home, and determining a target screening condition; searching in an educational resource library according to the disease information of the target old person input by a worker to obtain a plurality of initial educational resources; dividing all the old people into a plurality of clusters according to the relation topology network corresponding to all the old people in the nursing home; according to the personal attribute information of the target old person and a support vector machine, matching with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target old person belongs; and according to the feedback information, optimizing a target screening condition, and according to the optimized target screening condition and each initial educational resource, determining an optimal educational resource and recommending the optimal educational resource to the worker. According to the embodiment of the invention, the accuracy of recommending the educational resources to the target user is improved, and personalized recommendation to the user is realized.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular, to an intelligent recommendation method for nursing education resources. Background Art

[0002] In current various high-end nursing homes, especially in the late-stage care of various diseases of the elderly by caregivers, health education, as an important means to promote rehabilitation and prevent recurrence of diseases, its effectiveness and pertinence are particularly important. However, with the continuous progress of medical technology and the increasing diversification of the needs of the elderly in nursing homes, the traditional health education model for disease care has exposed many limitations.

[0003] Currently, affected by the environment of a large number of elderly people and a shortage of caregivers in the nursing home, many caregivers often only perform routine procedures in the care of the elderly's diseases, lacking in-depth understanding of the diseases suffered by each elderly person and implementing targeted care mechanisms, reducing the experience of the elderly in the nursing home and the satisfaction of customers with the nursing home; moreover, the turnover rate of caregivers in nursing homes is very high, and some newly recruited caregivers lack industry experience and are difficult to get started quickly, affecting the daily work of the nursing home. Summary of the Invention

[0004] The present invention provides an intelligent recommendation method for nursing education resources. In view of the shortage of caregivers in nursing homes, through this intelligent recommendation method for nursing education resources, the most suitable education resources are provided for each elderly person. By referring to these education resources, the nursing efficiency of the staff for the elderly can be improved, and one staff member can care for more elderly people, thereby reducing the pressure of staff shortage in the nursing home and reducing the labor cost of the nursing home; even if there is a shortage of caregivers in the nursing home, inexperienced staff can quickly master nursing skills according to these education resources, solving the employment pressure of the nursing home.

[0005] In a first aspect, an embodiment of the present invention provides an intelligent recommendation method for nursing education resources, which is used in a nursing home, and the method includes:

[0006] According to the skill information input by the staff who take care of the elderly in the nursing home, obtain the skill level of the staff, and according to the skill level, determine the target screening conditions, which are weighted and determined by language difficulty, content depth, and presentation form, and the skill information includes the education background, professional experience, and working years of the staff;

[0007] According to the disease information of the target elderly person input by the staff, search in the education resource library to obtain a plurality of initial education resources;

[0008] According to the relationship topology network corresponding to all the elderly in the nursing home, all the elderly are divided into multiple clusters. The relationship topology network includes all the elderly in the nursing home, the disease similarity relationship and preference similarity relationship between any two elderly in the nursing home, and the personal attribute information of the elderly. The personal attribute information includes the age, occupation, and hobbies of the elderly.

[0009] Match the personal attribute information of the target elderly with the attribute feature matrix corresponding to each cluster according to the support vector machine to obtain the cluster to which the target elderly belongs.

[0010] Regularly receive the feedback information corresponding to the cluster, optimize the target screening conditions according to the feedback information, and determine the best educational resources based on the optimized target screening conditions and each initial educational resource, and recommend them to the staff. The feedback information includes the problems reflected by the elderly and / or their relatives in the cluster.

[0011] In a second aspect, an embodiment of the present invention provides an intelligent recommendation system for nursing education resources. The system is used in a nursing home and includes:

[0012] A level module for obtaining the skill level of the staff according to the skill information input by the staff who take care of the elderly in the nursing home, and determining the target screening conditions according to the skill level. The target screening conditions are determined by weighting the language difficulty, content depth, and presentation form. The skill information includes the education background, professional experience, and working years of the staff.

[0013] A search module for searching in the educational resource library according to the disease information of the target elderly input by the staff to obtain multiple initial educational resources.

[0014] A clustering module for dividing all the elderly into multiple clusters according to the relationship topology network corresponding to all the elderly in the nursing home. The relationship topology network includes all the elderly in the nursing home, the disease similarity relationship and preference similarity relationship between any two elderly in the nursing home, and the personal attribute information of the elderly. The personal attribute information includes age, occupation, and hobbies.

[0015] A matching module for matching the personal attribute information of the target elderly with the attribute feature matrix corresponding to each cluster according to the support vector machine to obtain the cluster to which the target elderly belongs.

[0016] An optimization module, which is used to regularly receive the feedback information corresponding to the affiliated cluster, optimize the target screening conditions according to the feedback information, and determine the best educational resources based on the optimized target screening conditions and each initial educational resource, and recommend them to the staff. The feedback information includes the problems reflected by the elderly and / or their relatives in the affiliated cluster.

[0017] In a third aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned intelligent recommendation method for nursing education resources are implemented.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned intelligent recommendation method for nursing education resources are implemented.

[0019] An intelligent recommendation method for nursing education resources proposed by the present invention is applicable to nursing homes, especially in an environment where there are many disabled elderly people and a shortage of nursing workers in nursing homes. First, according to the skill information of the staff in the nursing home, the skill level of the staff is determined, and the target screening conditions are determined accordingly, so as to recommend different educational resources to staff with different educational backgrounds and occupations, and increase the matching degree between the staff and the educational resources; then, according to the disease information of the target elderly person input by the staff, a search is performed in the educational resource library to obtain multiple initial educational resources, so as to narrow the search scope of subsequent educational resources.

[0020] Then, according to the relationship topology network corresponding to all the elderly people in the nursing home, all the elderly people are divided into multiple clusters, the affiliated cluster with the highest similarity to the target elderly person is determined, and the target screening conditions are further optimized according to the feedback information on the previously recommended educational resources in these affiliated clusters. After obtaining the best educational resources, they are recommended to the staff to facilitate the staff to care for the elderly according to the best educational resources.

[0021] An embodiment of the present invention provides an intelligent recommendation method for nursing education resources, which has the following advantages:

[0022] (1) According to the skill information of the staff in the nursing home, the skill level of the staff is determined, and the target screening conditions are determined accordingly, so as to recommend different educational resources to staff with different educational backgrounds and occupations, and increase the matching degree between the staff and the educational resources.

[0023] (2) According to the disease information of the target elderly person, initial educational resources are screened out from the educational resource library, shortening the search time for subsequent best educational resources and improving the recommendation efficiency of nursing education resources.

[0024] (3) Through the relationship topology network, the elderly with similarities are divided into the same cluster. Through the support vector machine, the cluster with the highest similarity to the disease information of the target elderly is matched. Combining the feedback information of the affiliated cluster, the target screening conditions are optimized, so that the scores of each initial educational resource are more accurate in the end, improving the accuracy of the recommended best educational resources. Through artificial intelligence technology, the best match among the staff, educational resources, and the elderly can be achieved, further enhancing the satisfaction of the target users.

[0025] (4) Through this nursing education resource recommendation method, even if there is a shortage of nursing staff in the nursing home, inexperienced staff can quickly master nursing skills based on this educational resource, solving the employment pressure in the nursing home. Moreover, by referring to this educational resource, the nursing efficiency of the staff towards the elderly can be improved, and a single staff member can care for a larger number of elderly people, thus further reducing the pressure of personnel shortage in the nursing home and reducing the labor cost in the nursing home. Description of the Drawings

[0026] Figure 1 Schematic diagram of the application scenario of an intelligent nursing education resource recommendation method provided by an embodiment of the present invention;

[0027] Figure 2 Flowchart of an intelligent nursing education resource recommendation method provided by an embodiment of the present invention;

[0028] Figure 3 Schematic diagram of the structure of an intelligent nursing education resource recommendation system provided by an embodiment of the present invention;

[0029] Figure 4 Schematic diagram of the structure of a computer device provided by an embodiment of the present invention.

[0030] The realization, functional characteristics, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0031] The following details the embodiments of the present application. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the drawings are exemplary only for explaining the present application and should not be construed as a limitation of the present application.

[0032] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope protected by this application.

[0033] In the embodiments of this application, "at least one" means one or more; "a plurality" means two or more. In the description of this application, terms such as "first", "second", and "third" are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying an order. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of this application, "a plurality" means two or more, unless otherwise specifically defined.

[0034] The reference to "an embodiment" or "some embodiments" etc. described in this specification means that a specific feature, structure, or characteristic described in conjunction with the embodiment is included in one or more embodiments of this application. Thus, the terms "including", "comprising", "having" and their variants in this specification all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0035] Figure 1 The following is a schematic diagram of the application scenario of an intelligent recommendation method for nursing education resources provided for the embodiments of the present invention. As Figure 1 shown, in the case of a shortage of nursing staff in a nursing home, newly recruited and less skilled staff can input skill information and the disease information of the target elderly person who needs nursing on the page provided by the client. There are pre-set data points on this page. When the user operates on the page, the corresponding data points will be triggered. According to the data point code, the corresponding process node number is found. The client collects all the process node numbers corresponding to the operations and sends all the process node numbers to the server. After receiving the process node number, the server executes this intelligent recommendation method for nursing education resources to push the best educational resources to this staff, helping this staff quickly master the nursing work of the target elderly person, reducing the adaptation time of newly recruited staff, and being able to provide a relatively satisfactory and personalized service that meets the needs of the target elderly person.

[0036] It should be noted that the server can be implemented by an independent server or a server cluster composed of multiple servers. The client can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The client and the server can be connected through Bluetooth, USB (Universal Serial Bus), or other communication connection methods, and the embodiments of the present invention do not limit this here.

[0037] The embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (abbreviated as AI) is a theory, method, technology, and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0038] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics technology, biometric technology, speech processing technology, natural language processing technology, and machine learning, deep learning.

[0039] Figure 2 It is a flowchart of an intelligent recommendation method for nursing education resources provided by an embodiment of the present invention. As Figure 2 shown, the method includes:

[0040] S110, according to the skill information input by the staff caring for the elderly in the nursing home, obtain the skill level of the staff, and according to the skill level, determine the target screening conditions, where the target screening conditions are determined by weighting the language difficulty, content depth, and presentation form, and the skill information includes the education background, work experience, and working years of the staff;

[0041] This method mainly targets some high-end nursing homes. Generally, such high-end nursing homes charge higher fees. However, in today's environment with greater pension pressure, the shortage of nursing workers has also become a headache problem for nursing homes; moreover, the mobility of nursing workers is also very large, and generally their education level is not high. Therefore, how to enable nursing workers to quickly master nursing skills is very important.

[0042] First, obtain the skill information input by the staff who take care of the elderly in the nursing home. The staff are internal staff in the nursing home who need to learn educational resources, such as newly recruited nursing workers or those without industry experience who need to start from scratch. Specifically, it can be determined according to the actual situation, and the embodiments of the present invention do not make specific limitations on this. The skill information includes the staff's education level, occupation, working years, etc.; the education level can include three categories, such as low education level, middle school education level, and high education level; the occupation can be divided into two categories, occupations related to medical care and occupations not related to medical care, and the target user can select according to their actual situation; the working experience of the staff can be seen from the working years.

[0043] As an implementation manner, the steps of obtaining the skill level of the staff according to the skill information input by the staff who take care of the elderly in the nursing home include:

[0044] According to the skill information of the staff, obtain the education level grade, occupation grade, and experience grade. Among them, the education level grade includes low education level, middle education level, and high education level, the occupation grade includes occupations related to medicine and occupations not related to medicine, and the experience grade includes proficient, general, and unfamiliar;

[0045] Determine the skill level of the target user according to the education level grade, the occupation grade, and the experience grade.

[0046] In the embodiments of the present invention, according to the education level in the education and occupation information, if the education level is primary school or below, the education level grade is set as low education level; if it is middle school or secondary vocational school, the education level grade is set as middle education level; if it is university or above, the education level grade is set as high education level. Occupations are divided into occupations related to medicine and occupations not related to medicine. For different combinations of education and occupation, corresponding screening conditions are set. For example, for patients with low education level and occupations not related to medicine, the screening conditions tend to be videos with simple and easy-to-understand, life-like language expressions, and multiple case displays; for patients with high education level and occupations related to medicine, the screening conditions pay more attention to the professionalism and academic nature of the video content and the introduction of cutting-edge research results.

[0047] After determining the skill level of the target user, then determine the target screening conditions according to this skill level. In the embodiments of the present invention, the target screening conditions are determined by weighted combination of language difficulty, content depth, and presentation form. The weights corresponding to different skill levels are different. For example, the target screening conditions = the first weight * language difficulty + the second weight * content depth + the third weight * presentation form. The first weight, the second weight, and the third weight corresponding to different skill levels are different, and can be specifically determined according to the actual situation.

[0048] As an example, the steps of determining the target screening conditions according to the skill level include:

[0049] If the educational level in the skill level is a low educational level, the occupational level is a non-medical-related occupation, and the experience level is inexperienced, then set the language difficulty weight in the target screening conditions to 0.4, the content depth weight to 0.3, and the presentation form weight to 0.3;

[0050] If the educational level in the skill level is a high educational level, the occupational level is a medical-related occupation, and the experience level is proficient, then set the language difficulty weight in the target screening conditions to 0.2, the content depth weight to 0.5, and the presentation form weight to 0.3.

[0051] In the embodiments of the present invention, if the educational level is a low educational level and the occupational level is a non-medical-related occupation, then the first weight in the target screening conditions is 0.4, the second weight is 0.3, and the third weight is 0.3; if the educational level is a high educational level and the occupational level is a medical-related occupation, then the first weight in the target screening conditions is 0.2, the second weight is 0.5, and the third weight is 0.3.

[0052] If the educational level is a medium educational level and the occupational level is a non-medical-related occupation, then the first weight in the target screening conditions is 0.35, the second weight is 0.3, and the third weight is 0.35; if the educational level is a medium educational level and the occupational level is a medical-related occupation, then the first weight in the target screening conditions is 0.3, the second weight is 0.35, and the third weight is 0.35.

[0053] If the educational level is a high educational level and the occupational level is a non-medical-related occupation, then the first weight in the target screening conditions is 0.25, the second weight is 0.5, and the third weight is 0.35.

[0054] S120. Search in the educational resource library according to the disease information of the target elderly person input by the staff to obtain a plurality of initial educational resources;

[0055] Then, search in the educational resource library according to the disease information of the target elderly person input by the staff. For example, first extract the disease name, conduct a keyword search in the huge educational resource library, and screen out all educational resources containing the disease name or related professional terms to form a preliminary video set.

[0056] As an implementation manner, the step of searching in the educational resource library according to the disease information of the target elderly person input by the staff to obtain a plurality of initial educational resources includes:

[0057] Extract the disease keywords of the target elderly person according to the disease information of the target elderly person;

[0058] According to the disease keywords, screen in the educational resource library to screen out multiple initial educational resources that exactly match the disease keywords.

[0059] For example, if the target user inputs "diabetes", all educations related to diabetes care, treatment, diet, etc. will be included in this preliminary set, and the educational resources contained in this set are the initial educational resources.

[0060] In addition, it should be noted that the educational resources in this educational resource library can be videos, documents, comics, etc., which can be determined according to the actual situation specifically, and the initial educational resources can also be videos, documents, comics, etc. User preference information refers to the types of educational resources that users like. Some users like videos, and some users like documents. According to this user preference information, for example, if the user likes videos, then the educational resources of the video category in the initial set are retained, and the educational resources of other categories are deleted, so as to obtain each initial educational resource.

[0061] S130, according to the relationship topology network corresponding to all the elderly in the nursing home, divide all the elderly into multiple clusters. The relationship topology network includes all the elderly in the nursing home, the disease similarity relationship and preference similarity relationship between any two elderly in the nursing home, and the personal attribute information of the elderly. The personal attribute information includes age, occupation, and hobby.

[0062] In the embodiment of the present invention, after screening out the initial educational resources according to the disease information of the target elderly input by the staff, the feedback information of various types of clustered elderly on the recommended best videos is further integrated to further optimize the matching function of the educational resources.

[0063] In the embodiment of the present invention, a corresponding relationship topology network is constructed. The relationship topology network in the embodiment of the present invention is a network constructed according to the disease similarity relationship between any two elderly in the nursing home. In this relationship topology network, each elderly corresponds to a node in this relationship topology network separately; the relationship topology network also includes the similarity relationship between any two elderly. If the disease similarity between any two historical patients is greater than the first threshold or the hobby similarity is higher than the second threshold, then there is an edge between the two nodes corresponding to the two elderly in this relationship topology network. For example, the first threshold is 50%, and the second threshold is 70%.

[0064] The relationship topology network in the embodiment of the present invention also includes the personal attribute information of each elderly. The personal attribute information can include gender, age, occupation, education level, etc. The specific information included can be determined according to the actual situation, and the embodiments of the present application do not make specific limitations on this.

[0065] As an implementation manner, based on the relationship topology network corresponding to all the elderly in the nursing home, all the elderly are divided into multiple clusters, and the steps include:

[0066] Based on the disease information and preference information of all the elderly in the nursing home, establish a relationship topology network, and obtain an adjacency matrix and an attribute feature matrix;

[0067] Input the adjacency matrix corresponding to each elderly person and the attribute feature matrix corresponding to each elderly person into the trained support vector machine to obtain the optimal low-dimensional representation matrix corresponding to each elderly person;

[0068] Obtain each cluster according to the optimal low-dimensional representation matrix corresponding to each elderly person and the clustering algorithm.

[0069] According to the relationship topology network, an adjacency matrix and an attribute feature matrix are obtained. In the embodiments of the present invention, through the similarity relationship between users in the relationship topology network, the adjacency matrix is converted. In the embodiments of the present invention, the similarity relationship between users is quantified through the adjacency matrix. Specifically, if there is an edge between two nodes in the relationship topology network, the element value corresponding to these two nodes in the adjacency matrix is set to 1, and if there is no edge, the element value corresponding to these two nodes is set to 0. The attribute feature matrix is used to quantify the personal attribute information of users. The attribute feature matrix is obtained by quantifying all the personal attribute information of users in the database. In this attribute feature matrix, different rows in the matrix represent different users, and different columns represent different personal attribute information of the elderly.

[0070] In the specific implementation process, the relationship topology network is represented by G=(V, E, X), where V={v 1 , v 2 , …, v n} represents the node set of the relationship topology network G, v 1 represents the first node, and this first node is the node corresponding to the first user in the nursing home. v 2 represents the second node, and this second node is the node corresponding to the second user in the nursing home. v n represents the nth node, that is, the node corresponding to the nth elderly person in the nursing home, and n represents the number of all the elderly in the nursing home; in the embodiments of the present invention, each node represents an elderly person; in the embodiments of the present invention, the similarity relationship between the elderly is represented by E={e i,j}, where e i,j represents the edge between node v i and node v j ; in the embodiments of the present invention, X represents the attribute feature matrix, and X∈R n×f, each row vector in the attribute feature matrix represents the personal attribute information of different elderly people, and f represents the attribute dimension of the target user. For example, if the personal attribute information includes gender, age, occupation, and disease information, then the value of f is 4.

[0071] As an implementation, the step of establishing a relationship topology network according to the disease information and preference information of all the elderly in the nursing home includes:

[0072] Assign a network node to each elderly person;

[0073] Calculate the disease similarity and preference similarity between every two elderly people;

[0074] If the disease similarity is greater than the first preset threshold or the preference similarity is greater than the second preset threshold, then set a connection between the network nodes corresponding to every two elderly people. Otherwise, do not set a connection between the network nodes corresponding to every two elderly people, and obtain the relationship topology network.

[0075] Specifically, in the embodiment of the present invention, the method for obtaining the adjacency matrix according to the relationship topology network is as follows: Assume A ij is the element corresponding to the i-th row and the j-th column of the adjacency matrix A. If there is an edge between node v i and node v j , then A ij = 1, which means that the disease similarity between the two elderly people represented by v i and v j is greater than 50% or the hobby similarity between the two elderly people is greater than 70%; if there is no edge between node v i and node v j , then A ij = 0, which means that the similarity between the two historical patients represented by v i and v j is less than 50% and the hobby similarity between the two elderly people is less than 70%.

[0076] In the embodiment of the present invention, input the adjacency matrix corresponding to each elderly person and the attribute feature matrix corresponding to each elderly person into the trained vector representation learning model, and the best low-dimensional representation matrix can be obtained.

[0077] Then, according to the optimal low-dimensional representation matrix corresponding to each elderly person, input the optimal low-dimensional representation matrix corresponding to each elderly person into the target clustering algorithm to obtain the target category corresponding to each elderly person, and then each cluster can be obtained. Clustering is to divide a data set into different classes or clusters according to a specific criterion (such as distance), so that the similarity of data objects within the same cluster is as large as possible, and at the same time, the difference of data objects not in the same cluster is also as large as possible. That is, data of the same class are gathered together as much as possible, and data of different classes are separated as much as possible. The algorithms of clustering analysis can be divided into partitioning methods, hierarchical methods, density-based methods, grid-based methods, model-based methods, etc. The clustering analysis method in the embodiments of the present invention can be specifically determined according to the actual situation.

[0078] S140. Match the personal attribute information of the target elderly person and the support vector machine with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target elderly person belongs.

[0079] As an implementation manner, the step of matching the personal attribute information of the target elderly person and the support vector machine with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target elderly person belongs includes:

[0080] First, generate simulated cluster data. Each simulated cluster has a corresponding attribute feature matrix, and assign a corresponding label to each simulated cluster data.

[0081] Use the support vector machine to train the simulated cluster data with labels.

[0082] Input the personal attribute information of the target elderly person into the trained support vector machine to predict the label of the target elderly person, so as to determine the cluster to which the target elderly person belongs.

[0083] In an actual application scenario, the attribute information data of the elderly in different clusters may have been collected. To simulate this situation, we use the numpy library to generate simulated cluster data. Here, it is assumed that there are multiple clusters, and the data of each cluster has certain distribution characteristics. For the convenience of subsequent model training, we merge the data of all clusters into a large feature matrix.

[0084] To enable the support vector machine model to learn the differences between different clusters, we need to assign corresponding labels to the data of each cluster. The support vector machine (SVM) is used to train the merged feature matrix and the corresponding labels.

[0085] In practical applications, the attribute information of the target elderly is known. To simulate this situation, we randomly generate the attribute information of the target elderly and use the trained support vector machine model to predict the attribute information of the target elderly to determine the cluster to which they belong.

[0086] S150, regularly receive the feedback information corresponding to the cluster to which it belongs, and according to the feedback information, optimize the target screening conditions, and determine the best educational resources based on the optimized target screening conditions and each initial educational resource, and recommend them to the staff. The feedback information includes the problems reflected by the elderly and / or their relatives in the cluster to which they belong.

[0087] In this embodiment, the feedback information corresponding to the cluster to which the elderly belong is regularly received. This feedback information can be provided by the elderly's family members or the elderly themselves. The feedback information can include evaluations of various aspects of the professional services and basic services of the caregiver. Professional services include nursing comfort, satisfaction with nursing procedures, and post-operative evaluations of nursing, etc. Basic services include whether the elderly's needs are promptly reported, and the satisfaction of the elderly with food, clothing, housing, etc. According to this feedback information, the target screening conditions are optimized, and two parameters, namely professional services and basic services, are added to the target screening conditions. Thus, each initial educational resource is scored according to the optimized target screening conditions, and the best educational resource is selected from each initial educational resource according to the score results.

[0088] As an implementation method, the step of optimizing the target screening conditions according to the feedback information and determining the best educational resource based on the optimized target screening conditions and each initial educational resource includes:

[0089] If the problem contains key indicators related to basic services, then add the basic services as a weight parameter item to the target screening conditions, and adjust the weights of the basic services, the language difficulty, the content depth, and the presentation form;

[0090] If the problem contains key indicators related to professional services, then use the professional services as the proportionality coefficient of the target screening conditions to obtain the optimized target screening conditions;

[0091] Determine the best educational resource according to the optimized target screening conditions and each initial educational resource.

[0092] After determining the cluster to which the target elderly person belongs, the elderly people in this cluster are the ones who are most similar to the target elderly person in all aspects. Therefore, in order to further improve the accuracy of educational resource recommendation, the feedback information of the relatives of the elderly people in the cluster is used. According to whether the feedback information contains professional service opinions and basic service opinions, the target screening conditions are optimized, and the best educational resources are selected using the optimized target screening conditions and recommended to the target user.

[0093] Among them, if the feedback information contains key indicators related to basic services, such as food, clothing, housing, and transportation, etc., then this basic service is added as a weight parameter to the target screening conditions. In this case, the optimized target screening conditions == the first weight * language difficulty + the second weight * content depth + the third weight * presentation form + the fourth weight * basic service; if the feedback information contains key indicators related to professional services, such as nursing techniques, etc., then this basic service is added as a weight parameter to the target screening conditions. In this case, the optimized target screening conditions == the proportionality coefficient corresponding to professional services * (the first weight * language difficulty + the second weight * content depth + the third weight * presentation form).

[0094] If the feedback information contains both professional services and basic services, then the optimized target screening conditions == the proportionality coefficient corresponding to professional services * (the first weight * language difficulty + the second weight * content depth + the third weight * presentation form + the fourth weight * basic service score).

[0095] According to the optimized target screening conditions, calculate the scores of each initial educational resource to obtain the best educational resources.

[0096] As an implementation method, the step of obtaining the initial score corresponding to each initial educational resource according to each initial educational resource and the target screening conditions includes:

[0097] With the help of natural language processing tools, evaluate the proportion of professional vocabulary, average sentence length, and complexity of grammatical structure of each initial educational resource in the educational resource library to obtain the language difficulty score corresponding to each initial educational resource;

[0098] Analyze the depth of medical knowledge involved in each initial educational resource, and judge whether each initial educational resource contains basic nursing knowledge, advanced treatment principles, and cutting-edge research results. According to the judgment results, obtain the content depth score corresponding to each initial educational resource;

[0099] Use image recognition and video analysis technologies to evaluate the presentation method of each initial educational resource to obtain the presentation form score corresponding to each initial educational resource;

[0100] Using the trained neural network model, the professional score and basic score of each initial educational resource are extracted;

[0101] According to the language difficulty score, content depth score, expression form score, the professional score and the basic score corresponding to each initial educational resource, calculation is performed according to the target screening conditions to obtain the score corresponding to each initial educational resource, and the optimal educational resource is determined based on the score.

[0102] After obtaining the initial educational resources, it is necessary to analyze the language difficulty, content depth, presentation form, basic services and professional services of each initial educational resource, as follows:

[0103] Language difficulty analysis: The embodiment of the present invention uses natural language processing tools to perform vocabulary and sentence complexity analysis on the text after the video explanation audio is converted into text. Statistics include indicators such as the proportion of professional vocabulary, average sentence length, and complexity of grammatical structure. For example, in a video text explaining cardiovascular disease care, if professional vocabulary such as "coronary atherosclerosis" and "arrhythmia" appear frequently, and the sentences are mostly complex medical terms, the language difficulty is greater for patients with low education. In this way, the language difficulty score corresponding to each initial educational resource is determined.

[0104] Content depth analysis: An embodiment of the present invention analyzes the depth of medical knowledge involved in the video content. Determine whether it contains different levels of content such as basic nursing knowledge, advanced treatment principles, and cutting-edge research results. For example, for cancer care videos, the basic content may be daily dietary precautions, the advanced content is the principles of chemotherapy, and the cutting-edge results are introductions to new immunotherapy methods. For patients in medical-related occupations, videos containing the latter two in-depth contents may be more suitable. In this way, the content depth score corresponding to each initial educational resource is determined.

[0105] Expression evaluation: The embodiment of the present invention uses image recognition and video analysis technology to evaluate the presentation of video images. Analyze the proportion of elements such as charts, animations, and real-life demonstrations in the screen. If the video uses a large number of professional medical charts to show the principles of the disease, it may be difficult for non-medical related patients to understand; while vivid animation demonstrations or real-life nursing operation demonstrations are more suitable for various types of patients. In this way, the expression score corresponding to each initial educational resource is determined.

[0106] When calculating the professional score and the basic score, each initial educational resource is input into the trained neural network model, and the professional score and the basic score of each initial educational resource can be obtained. The neural network model is a machine learning network model, specifically a multi-layer perceptron. A multi-layer perceptron is a most basic feedforward neural network, composed of an input layer, multiple hidden layers, and an output layer. Each layer consists of multiple neurons, and the neurons transmit information through weighted connections. Nonlinear factors are introduced through activation functions, enabling the model to learn complex nonlinear relationships.

[0107] Finally, according to the first weight, second weight, third weight, and fourth weight corresponding to the language difficulty, content depth, presentation form, and basic score in the target screening conditions, combined with the language difficulty score, content depth score, presentation form score, basic score, and professional score corresponding to each initial educational resource, weighted calculation is performed to obtain the score corresponding to each initial educational resource.

[0108] An intelligent recommendation method for nursing educational resources proposed by the present invention is applicable to nursing homes, especially in an environment where there are many disabled elderly people and a shortage of nursing workers in nursing homes. First, according to the skill information of the staff in the nursing home, the skill level of the staff is determined, and based on this, the target screening conditions are determined, so as to recommend different educational resources to staff with different educational backgrounds and occupations, increasing the matching degree between the staff and the educational resources; then, according to the disease information of the target elderly person input by the staff, a search is performed in the educational resource library to obtain multiple initial educational resources, so as to narrow the search scope of subsequent educational resources.

[0109] Then, according to the relationship topology network corresponding to all the elderly people in the nursing home, all the elderly people are divided into multiple clusters, the cluster with the highest similarity to the target elderly person is determined, and based on the feedback information on the previously recommended educational resources in these clusters, the target screening conditions are further optimized. After obtaining the best educational resources, they are recommended to the staff to facilitate the staff to care for the elderly according to the best educational resources.

[0110] The embodiment of the present invention provides an intelligent recommendation method for nursing educational resources, which has the following advantages:

[0111] (1) According to the skill information of the staff in the nursing home, the skill level of the staff is determined, and based on this, the target screening conditions are determined, so as to recommend different educational resources to staff with different educational backgrounds and occupations, increasing the matching degree between the staff and the educational resources.

[0112] (2) According to the disease information of the target elderly person, initial educational resources are screened out from the educational resource library, shortening the search time for subsequent best educational resources and improving the recommendation efficiency of nursing educational resources.

[0113] (3) Through the relationship topology network, the elderly with similarities are divided into the same cluster. By matching the cluster with the highest similarity of disease information to the target elderly and combining the feedback information of the cluster to which it belongs, the target screening conditions are optimized, so that the scores of each initial educational resource are more accurate in the end, improving the accuracy of the recommended best educational resources, enabling the best match among the staff, educational resources, and the elderly, and further enhancing the satisfaction of the target users.

[0114] (4) Through this nursing education resource recommendation method, even if there is a shortage of nursing staff in the nursing home, inexperienced staff can quickly master nursing skills based on this educational resource, solving the employment pressure in the nursing home. Moreover, by referring to this educational resource, the nursing efficiency of the staff towards the elderly can be improved, and a single staff member can care for a relatively large number of elderly people, thereby further reducing the pressure of personnel shortage in the nursing home and reducing the labor cost in the nursing home.

[0115] Figure 3 The following is a schematic structural diagram of a nursing education resource intelligent recommendation system provided by an embodiment of the present invention, as Figure 3 shown. The system includes:

[0116] A level module 310, configured to obtain the skill level of the staff according to the skill information input by the staff who take care of the elderly in the nursing home, and determine target screening conditions according to the skill level. The target screening conditions are determined by weighting the language difficulty, content depth, and presentation form. The skill information includes the staff's education background, work experience, and working years;

[0117] A search module 320, configured to search in the educational resource library according to the disease information of the target elderly input by the staff to obtain a plurality of initial educational resources;

[0118] A clustering module 330, configured to divide all the elderly into multiple clusters according to the relationship topology network corresponding to all the elderly in the nursing home. The relationship topology network includes all the elderly in the nursing home, the disease similarity relationship and preference similarity relationship between any two elderly in the nursing home, and the personal attribute information of the elderly. The personal attribute information includes age, occupation, and hobby;

[0119] A matching module 340, configured to match the personal attribute information of the target elderly and a support vector machine with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target elderly belongs;

[0120] An optimization module 350 is configured to periodically receive feedback information corresponding to the affiliated cluster, optimize the target screening criteria according to the feedback information, and determine the best educational resources based on the optimized target screening criteria and each initial educational resource, and recommend them to the staff. The feedback information includes problems reflected by the elderly and / or relatives in the affiliated cluster.

[0121] This embodiment is a system embodiment corresponding to the above method embodiment. The specific implementation process is the same as that of the above method embodiment. For details, please refer to the above method embodiment. This system embodiment will not be elaborated here.

[0122] Each module in the above intelligent nursing education resource recommendation system can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in the form of hardware or independent of it, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0123] Figure 4 FIG. is a schematic structural diagram of a computer device provided by an embodiment of the present invention. The computer device can be a server, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a computer storage medium and an internal memory. The computer storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the computer storage medium. The database of the computer device is used to store data generated or obtained during the execution of an intelligent nursing education resource recommendation method, such as educational background and occupation information, disease information, and user preference information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes an intelligent nursing education resource recommendation method.

[0124] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps of an intelligent nursing education resource recommendation method in the above embodiment. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in an embodiment of an intelligent nursing education resource recommendation system.

[0125] In one embodiment, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of an intelligent recommendation method for nursing education resources in the above embodiment are implemented. Alternatively, when the computer program is executed by a processor, the functions of the various modules / units in the above embodiment of an intelligent recommendation system for nursing education resources are implemented.

[0126] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0128] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for intelligently recommending nursing education resources, characterized in that: The method is used in a nursing home, and comprises: According to the skill information input by the staff who take care of the elderly in the nursing home, the skill level of the staff is obtained, and according to the skill level, the target screening condition is determined, and the target screening condition is determined by weighting the language difficulty, content depth and expression form. The skill information includes the education level, professional experience and years of work of the staff; According to the disease information of the target elderly person input by the staff, searching in the education resource library to obtain a plurality of initial education resources; Divide all the elderly people in the nursing home into multiple clusters according to the relationship topology network corresponding to all the elderly people in the nursing home, the relationship topology network includes all the elderly people in the nursing home, the disease similarity relationship and preference similarity relationship between any two elderly people in the nursing home, and the personal attribute information of the elderly people, and the personal attribute information includes age, occupation, and hobbies; According to the personal attribute information of the target elderly and the support vector machine, matching is performed with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target elderly belongs; Regularly receive feedback information corresponding to the cluster to which the employee belongs, and optimize the target screening conditions based on the feedback information, and determine the best educational resources based on the optimized target screening conditions and each initial educational resource and recommend them to the staff, wherein the feedback information includes problems reflected by the elderly and / or relatives in the cluster to which the employee belongs.

2. The intelligent recommendation method for nursing education resources according to claim 1, characterized in that: The steps of matching the personal attribute information of the target elderly person and the support vector machine with the attribute feature matrix corresponding to each cluster to obtain the cluster to which the target elderly person belongs include: First, simulated cluster data is generated. Each simulated cluster has a corresponding attribute feature matrix, and each simulated cluster data is assigned a corresponding label. Use support vector machines to train on simulated clustering data with labels; The personal attribute information of the target elderly is input into a trained support vector machine, and the label of the target elderly is predicted, so as to determine the cluster to which the target elderly belongs.

3. The intelligent recommendation method for nursing education resources according to claim 1, characterized in that: The step of optimizing the target screening condition according to the feedback information and determining the best educational resource according to the optimized target screening condition and each initial educational resource comprises: If the question contains key indicators related to basic services, the basic services are added to the target screening conditions as weight parameters, and the weights of the basic services, the language difficulty, the content depth and the presentation form are adjusted; If the problem contains key indicators related to professional services, the professional services are used as the proportional coefficient of the target screening condition to obtain the optimized target screening condition; The optimal educational resource is determined based on the optimized target screening conditions and each initial educational resource.

4. The method for intelligent recommendation of nursing education resources according to claim 3, characterized in that: The step of determining the best educational resource according to the optimized target screening condition and each initial educational resource comprises: By means of natural language processing tools, the professional vocabulary proportion, average sentence length, and grammatical structure complexity of each initial educational resource in the educational resource library are evaluated to obtain a language difficulty score corresponding to each initial educational resource; Analyze the depth of medical knowledge involved in each initial education resource, determine whether each initial education resource contains basic nursing knowledge, advanced treatment principles, and cutting-edge research results, and obtain the content depth score corresponding to each initial education resource based on the judgment results; Using image recognition and video analysis technology, evaluate the presentation of each initial educational resource and obtain the presentation score corresponding to each initial educational resource; Using the trained neural network model, the professional score and basic score of each initial educational resource are extracted; According to the language difficulty score, content depth score, expression form score, the professional score and the basic score corresponding to each initial educational resource, calculation is performed according to the target screening conditions to obtain the score corresponding to each initial educational resource, and the optimal educational resource is determined based on the score.

5. The method for intelligent recommendation of nursing education resources according to claim 1, characterized in that: The step of dividing all the elderly people in the nursing home into multiple clusters according to the relationship topology network corresponding to all the elderly people in the nursing home includes: According to the disease information and preference information of all the elderly people in the nursing home, a relationship topology network is established, and an adjacency matrix and an attribute feature matrix are obtained; Input the adjacency matrix and attribute feature matrix corresponding to each elderly person into the trained support vector machine to obtain the optimal low-dimensional representation matrix corresponding to each elderly person; Each cluster is obtained according to the best low-dimensional representation matrix and clustering algorithm corresponding to each elderly person.

6. The method for intelligent recommendation of nursing education resources according to claim 5, characterized in that: The step of establishing a relationship topology network based on the disease information and preference information of all the elderly people in the nursing home includes: Assign a network node to each elderly person; Calculate the disease similarity and preference similarity between every two elderly people; If the disease similarity is greater than a first preset threshold or the preference similarity is greater than a second preset threshold, a connection is set between the network nodes corresponding to every two elderly people. Otherwise, no connection is set between the network nodes corresponding to every two elderly people to obtain the relationship topology network.

7. The method for intelligently recommending nursing education resources according to claim 1, characterized in that: The step of searching in the education resource library according to the disease information of the target elderly person input by the staff to obtain a plurality of initial education resources comprises: Extracting disease keywords of the target elderly person according to the disease information of the target elderly person; According to the disease keywords, screening is performed in the education resource library to screen out a plurality of initial education resources that completely match the disease keywords.

8. The method for intelligent recommendation of nursing education resources according to claim 1, characterized in that: The step of obtaining the skill level of the staff member who takes care of the elderly in the nursing home according to the skill information input by the staff member includes: According to the skill information of the staff member, the educational level, occupational level and experience level are obtained, wherein the educational level includes low education, medium education and high education, the occupational level includes medical-related occupations and non-medical-related occupations, and the experience level includes skilled, general and unskilled; The skill level of the target user is determined according to the educational level, the occupation level and the experience level.

9. The method for intelligently recommending nursing education resources according to claim 8, characterized in that: The step of determining the skill level of the target user according to the educational level, the occupation level and the experience level comprises: If the educational level is low educational level, the occupation level is non-medical related occupation and the experience level is unfamiliar, the language difficulty weight in the target screening condition is set to 0.4, the content depth weight is set to 0.3, and the expression form weight is set to 0.3; If the educational level in the skill level is high education, the occupation level is medical-related occupation and the experience level is proficient, the language difficulty weight in the target screening condition is set to 0.2, the content depth weight is set to 0.5, and the expression form weight is set to 0.

3.

10. The method for intelligently recommending nursing education resources according to claim 1, characterized in that: The educational resource library is obtained in the following way: Establish a whitelist of resource sources, which includes information released by authoritative medical institutions, professional medical education platforms, and personal accounts of qualified medical experts; Obtain various educational resources from the whitelist, and use natural language processing technology and image recognition technology to conduct a preliminary review of the various educational resources to obtain the educational resources after preliminary screening; Invite professional medical staff to form an evaluation team to manually review the educational resources after the initial screening, and add the educational resources that have passed the manual review to the educational resource library.

Images