Medical education guidance method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of medical education, and discloses a medical education guidance method and system based on artificial intelligence. The method comprises the following steps: setting a medical education resource library in a cloud data center, and constructing an artificial intelligence model; performing feature extraction and classification on a plurality of medical education resource data in the medical education resource library; generating a user portrait according to the real-time basic information and the real-time behavior data of the user; performing matching in the plurality of real-time resource classification results according to the real-time user portrait; performing medical education guidance according to the real-time user portrait and the plurality of matched real-time resource data features; sending the real-time medical education strategy and the plurality of target medical education resource data to a user terminal; and performing parameter optimization on the artificial intelligence model according to real-time feedback data sent by the user terminal. According to the invention, the problems of high cost investment, low efficiency, poor effect, lack of customization and low intelligent degree in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the field of medical education technology, and specifically relates to a medical education guidance method and system based on artificial intelligence. Background Art

[0002] Medical education refers to educational activities dedicated to the training of medical professionals, covering all aspects from basic medical knowledge to clinical skills. Medical education guidance technology refers to providing teachers with guidance and suggestions on teaching methods, teaching resources, and course arrangements during the teaching process of medical education courses, so as to improve the efficiency and experience of medical education.

[0003] The existing medical education guidance has the following defects: 1) Existing medical education guidance often relies on the experience of medical education experts. When providing medical education guidance, it is costly, inefficient, and ineffective, and cannot be promoted; 2) Existing medical education guidance lacks customization and cannot deeply explore the personalized needs of users, resulting in inaccurate recommendations of educational resources and failure to meet the specific needs of different users; 3) The existing medical education guidance has a low level of intelligence, making it difficult to achieve intelligent integration and efficient use of educational resources. In addition, there is a problem of insufficient data processing capabilities, resulting in low resource utilization. Summary of the invention

[0004] In order to solve the problems of high cost investment, low efficiency, poor effect, lack of customization and low intelligence in the prior art, the present invention aims to provide a medical education guidance method and system based on artificial intelligence.

[0005] The technical solution adopted by the present invention is: A medical education guidance method based on artificial intelligence comprises the following steps: Set up a medical education resource library in the cloud data center and use artificial intelligence algorithms to build a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model; Using the medical education resource classification model, feature extraction and classification are performed on a number of medical education resource data in the medical education resource library to obtain a number of real-time resource data features and a number of real-time resource classification results; Based on the real-time basic information and real-time behavior data of the user sent by the user terminal, a user portrait generation model is used to generate a user portrait to obtain a real-time user portrait of the user; According to the real-time user portrait, matching is performed among several real-time resource classification results, and several matching real-time resource data features corresponding to the successfully matched matching real-time resource classification results are obtained; According to the real-time user portrait and several matching real-time resource data features, the medical education guidance model is used to provide medical education guidance, and the real-time medical education strategy and several target real-time resource data features are obtained; According to the characteristics of several target real-time resource data, searching is performed in the medical education resource library, and the real-time medical education strategy and the obtained several target medical education resource data are sent to the user terminal; According to the real-time feedback data sent by the user terminal, the parameter optimization model is used to optimize the parameters of the medical education resource classification model, the user portrait generation model and the medical education guidance model.

[0006] Furthermore, a medical education resource library is set up in the cloud data center, and an artificial intelligence algorithm is used to construct a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model, including the following steps: Collect a number of heterogeneous medical education resource data in different medical fields, and transform the data structure according to the preset data structure of the cloud data center to obtain a number of homogeneous medical education resource data; Preprocess a number of isomorphic medical education resource data to obtain a number of final medical education resource data, summarize and organize them, and build a medical education resource library; Based on some final medical education resource data, a content feature extraction and classification algorithm based on deep learning is used to build a medical education resource classification model and generate some historical resource data features; Based on the historical basic information and historical behavior data of several users, a user portrait generation model is constructed using a key feature screening and portrait generation algorithm based on deep learning, and several historical user portraits are generated; Based on several historical user portraits and several historical resource data features, a medical education guidance model is constructed using an enhanced reinforcement learning algorithm; According to several historical model parameters of the medical education resource classification model, user portrait generation model and medical education guidance model in the construction process, a parameter optimization model was constructed using a swarm intelligence optimization algorithm.

[0007] Furthermore, the classification model of medical education resources is constructed based on the CNN-LSTM-MLP algorithm; The user portrait generation model is built based on the LSTM-RF- Attention-MLP algorithm; The medical education guidance model is built based on the MPO-DQN algorithm; The parameter optimization model is built based on the ICPO algorithm.

[0008] Furthermore, based on several historical user portraits and several historical resource data features, an enhanced reinforcement learning algorithm is used to construct a medical education guidance model, including the following steps: The problem of generating medical education strategies is used as a simulation environment for the DQN algorithm, and the network parameter generation problem of the DQN algorithm is divided into several subtasks of the MPO algorithm according to the classification results of different preset resources. Use the MPO algorithm to build the initial meta-strategy optimization module, input several historical user portraits, train under several subtasks, obtain the final meta-strategy optimization module, and generate several historical meta-strategy optimization experiences; Use the final meta-strategy optimization module to output the historical initial network parameters of the DQN algorithm, and use the DQN algorithm to build the initial reinforcement learning module based on the historical initial network parameters; Define the state space based on historical user portraits and historical resource data characteristics, define the action space based on preset medical education actions, and set the reward function; Based on the state space, action space and reward function, several historical user portraits and several historical resource data features are input to train the agent and deep Q network of the initial reinforcement learning module to obtain the final reinforcement learning module and generate several historical reinforcement learning experiences; Several historical meta-strategy optimization experiences and several historical reinforcement learning experiences are stored in the experience replay pool of the final reinforcement learning module, and the final meta-strategy optimization module and the final reinforcement learning module are integrated to obtain a medical education guidance model.

[0009] Further, using the medical education resource classification model, feature extraction and classification are performed on a number of medical education resource data in the medical education resource library to obtain a number of real-time resource data features and a number of real-time resource classification results, including the following steps: Continuously collect a number of medical education resource data from the Internet, and store the collected medical education resource data in a medical education resource library; After the resource classification cycle is reached, the medical education resource classification model is used to extract the real-time unstructured content features and real-time structured content features of the medical education resource data; Perform feature fusion on real-time unstructured content features and real-time structured content features to obtain corresponding real-time resource data features; According to the characteristics of real-time resource data, the corresponding medical education resource data is classified to obtain the real-time resource classification results; All medical education resource data in the medical education resource library are traversed to obtain several real-time resource data features and several real-time resource classification results.

[0010] Further, based on the real-time basic information and real-time behavior data of the user sent by the user terminal, a user portrait generation model is used to generate a user portrait to obtain a real-time user portrait of the user, including the following steps: User terminals collect real-time basic information and real-time behavior data of users, and encrypt and upload the real-time basic information and real-time behavior data to the cloud data center; The cloud data center associates the user's real-time basic information with the real-time behavior data to obtain real-time user data and inputs it into the user portrait generation model; According to the key feature importance scores generated during the construction process, the attention weight values ​​of the user portrait generation model are set, and the user portrait generation model is used to extract the real-time user data features of the real-time user data; According to the attention weight value, the real-time user data features are weighted, and based on the obtained real-time weighted features, a user portrait is generated to obtain a real-time user portrait of the user.

[0011] Further, according to the real-time user portrait, matching is performed among several real-time resource classification results, and several matching real-time resource data features corresponding to the successfully matched matching real-time resource classification results are obtained, including the following steps: Obtain the real-time matching degree between each real-time resource classification result and the real-time user portrait, and take the real-time resource classification result with the highest real-time matching degree as the successfully matched real-time resource classification result; The real-time resource data features corresponding to the matching real-time resource classification results are used as matching real-time resource data features to obtain a plurality of matching real-time resource data features.

[0012] Further, according to the real-time user portrait and several matching real-time resource data features, the medical education guidance model is used to perform medical education guidance, and the real-time medical education strategy and several target real-time resource data features are obtained, including the following steps: According to the real-time user portrait, a number of historical meta-strategy optimization experiences stored in the experience playback pool of the medical education guidance model are matched to obtain the real-time meta-strategy optimization experience with successful matching; According to the real-time meta-strategy optimization experience, the meta-strategy optimization module is used to generate real-time initialization network parameters, and the reinforcement learning module is adjusted according to the initialization network parameters to obtain the adjusted reinforcement learning module; According to the real-time user portrait, a number of historical reinforcement learning experiences stored in the experience replay pool of the medical education guidance model are matched to obtain the real-time reinforcement learning experience with successful matching; According to the real-time reinforcement learning experience, the action space of the reinforcement learning module is updated and adjusted to obtain an updated action space; According to the real-time user portrait and several matching real-time resource data features, the state space of the reinforcement learning module is updated and adjusted to obtain an updated state space; According to the updated state space and the updated action space, the adjusted reinforcement learning module is used to provide medical education guidance and obtain a real-time medical education strategy; According to the real-time medical education strategy, a search is performed among several matching real-time resource data features to obtain several target real-time resource data features.

[0013] Furthermore, according to the real-time feedback data sent by the user terminal, the parameter optimization model is used to optimize the parameters of the medical education resource classification model, the user portrait generation model and the medical education guidance model, including the following steps: According to the real-time feedback data sent by the user terminal, the optimization target of the parameter optimization model is set, and according to the optimization target, the fitness function of the parameter optimization model is set; Encoding the comprehensive model parameter schemes of the medical education resource classification model, the user profile generation model, and the medical education guidance model into the individual vector format of the parameter optimization model; According to the individual vector format and fitness function, a parameter optimization model is used to perform iterative optimization, and the individual vector of the optimal individual is decoded to generate the optimal comprehensive real-time model parameter solution; According to the optimal comprehensive real-time model parameter scheme, the parameters of the medical education resource classification model, user portrait generation model and medical education guidance model are optimized.

[0014] A medical education guidance system based on artificial intelligence is used to implement a medical education guidance method, the system includes a cloud data center and a plurality of user terminals, each of which is in communication connection with the cloud data center; The cloud data center includes a model building unit, a medical education resource classification unit, a user portrait generation unit, a resource classification matching unit, a medical education guidance unit, a medical education resource sending unit and a parameter optimization unit which are connected in sequence.

[0015] The beneficial effects of the present invention are: The present invention discloses a medical education guidance method and system based on artificial intelligence, which provides a unified medical education guidance program based on a cloud data center, promotes the popularization and balance of medical education resources, avoids dependence on medical education experts, reduces the cost investment of medical education guidance, improves efficiency, and can be promoted in different regions and different medical education fields; the medical education resource library realizes the collection and aggregation of medical education resources, improves the richness of medical education resources, the medical education resource classification model provides powerful data processing capabilities, performs automatic resource classification, and improves the accuracy of resource retrieval and recommendation; the user portrait generation model can more accurately capture the characteristics of users and provide refined data support for subsequent education resource recommendations; the medical education guidance model can adapt to the user's behavioral characteristics, teaching progress, resource difficulty, etc., and provide accurate customized medical education strategies; the parameter optimization model realizes the continuous evolution of the artificial intelligence model according to user feedback; by constructing an automated medical education guidance system, the intelligence level of medical education guidance is enhanced and the resource utilization rate is improved.

[0016] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the medical education guidance method based on artificial intelligence in the present invention.

[0018] Figure 2 It is a structural block diagram of the medical education guidance system based on artificial intelligence in the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0020] Embodiment 1: like Figure 1 As shown, this embodiment provides a medical education guidance method based on artificial intelligence, comprising the following steps: S1: Set up a medical education resource library in the cloud data center and use artificial intelligence algorithms to build a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model, including the following steps: S1-1: Collect a number of heterogeneous medical education resource data in different medical fields, and transform the data structure according to the preset data structure of the cloud data center to obtain a number of homogeneous medical education resource data; Medical education resource data comes from different data sources, and the data structures are quite different. Data structure transformation and unification are required to provide support for subsequent model classification, resource matching, and resource data transmission. S1-2: Preprocess a number of isomorphic medical education resource data to obtain a number of final medical education resource data, summarize and organize them, and build a medical education resource library; Preprocessing includes data cleaning, format conversion, and normalization. While eliminating duplicate and erroneous data, it also performs standardized conversion, eliminates magnitude differences and noise, and converts the data into a format that can be recognized by subsequent models. S1-3: Based on some final medical education resource data, use the deep learning-based content feature extraction and classification algorithm to build a medical education resource classification model and generate some historical resource data features; The medical education resource classification model is constructed based on the Convolutional Neural Network (CNN)-Long Short-Term Memory Network (LSTM)-Multilayer Perceptron (MLP) algorithm, and the medical education resource classification model includes an unstructured content feature extraction module constructed based on the CNN algorithm, a structured content feature extraction module constructed based on the LSTM, and a medical education resource classification module constructed based on the MLP algorithm. Both the unstructured content feature extraction module and the structured content feature extraction module are connected to the medical education resource classification module; The unstructured content feature extraction module is used to extract the content features of unstructured data in medical education resource data, including pictures, videos, texts, etc. The structured content feature extraction module is used to extract the content features of structured data in medical education resource data, including tables, database data, etc. The medical education resource classification module is a general feedforward neural network, which is used to fuse the unstructured content features and structured content features, and classify and predict according to the fused features, thereby improving the accuracy of medical education resource classification; S1-4: Based on the historical basic information and historical behavior data of several users, a key feature screening and portrait generation algorithm based on deep learning is used to build a user portrait generation model and generate several historical user portraits; The user portrait generation model is built based on the LSTM-Random Forest (RF)-Attention-MLP algorithm, and includes a user feature extraction module built based on the LSTM algorithm, a key feature screening module built based on the RF algorithm, an attention weighting module built based on the Attention mechanism, and a user portrait generation module built based on the MLP algorithm. The user feature extraction module is used to extract the data features of user data, including the user's basic information features and usage behavior features. The key feature screening module screens the feature components of the input user data through the internal Classification And Regression Tree (CART). It can process a large number of feature components and select the most stable and discriminative feature components, and generate key feature importance scores for each feature component to set the attention weight value of the attention weighting module to enhance the characterization ability of key features. The key feature screening module only participates in training and key feature screening during the model construction process. After training, each time a new feature is input, it can be weighted according to the attention weight value of the attention weighting module, which enhances the model's attention to key features and improves the accuracy and efficiency of model prediction. The user portrait generation module is used to generate user portraits based on weighted features and output predicted user portrait labels. The working method of the key feature screening module includes the following steps: A-1: Use the CART tree of the RF algorithm to extract the feature contribution of several feature components in the historical user data; the historical user data includes the associated historical basic information and historical behavior data; The formula is:

[0021] In the formula, For the Feature contribution of feature components; For the Feature components in random forest The feature contribution of each tree; is the CART tree indicator; is the characteristic component indicator; is the total number of CARTs;

[0022] In the formula, CART tree node for random forest m ,node and nodes The Gini index of CART tree node m Medium Category The proportion of is the total number of categories; m , , is the node indicator; is the category indicator; A-2: normalizing the feature contributions of several feature components to obtain corresponding normalized feature contributions; The formula is:

[0023] In the formula, It is the contribution of the feature after normalization; J is the total number of characteristic components; A-3: Generate feature selection standard values ​​for several feature components based on the normalized feature contribution; The formula is:

[0024] In the formula, For the Feature selection criterion values ​​for feature components; For the The normalized feature contribution of the feature components; is the characteristic component indicator; A-4: The feature selection standard value of the feature component is used as the key feature importance score, and the first M feature components are selected as the key feature components according to the key feature importance score; A-5: According to the key feature importance scores of the M key feature components, the corresponding attention weight values ​​are set in the corresponding feature component channels in the attention weighting module, and the attention weight values ​​of the feature component channels of the remaining non-key feature components are set to 0; S1-5: Based on several historical user portraits and several historical resource data features, a medical education guidance model is constructed using an enhanced reinforcement learning algorithm; The medical education guidance model is constructed based on the Meta-Policy Optimization (MPO)-Deep Q Network (DQN) algorithm, and the medical education guidance model includes a meta-policy optimization module constructed based on the MPO algorithm and a reinforcement learning module constructed based on the DQN algorithm. The reinforcement learning module is provided with an intelligent agent, a deep Q network, and an experience replay pool. The intelligent agent is connected to the deep Q network, the experience replay pool, and the meta-policy optimization module, and the meta-policy optimization module is connected to the experience replay pool. The meta-strategy optimization module is used to optimize the deep Q network parameters in the reinforcement learning module so that these parameters can quickly adapt to new and unseen tasks, improve the generalization ability of the model, and update the deep Q network based on previous learning experience even under unseen data; through meta-strategy optimization, the DQN algorithm can converge faster, reduce the amount of data and time required for training, and significantly improve learning efficiency; the reinforcement learning module outputs personalized medical education strategies based on the characteristics of user portraits, and provides the most appropriate medical education resources, education paths and education methods to improve the user's teaching effect; Based on several historical user portraits and several historical resource data features, an enhanced reinforcement learning algorithm is used to build a medical education guidance model, including the following steps: S1-5-1: The medical education strategy generation problem is used as the simulation environment of the DQN algorithm, and the network parameter generation problem of the DQN algorithm is divided into several subtasks of the MPO algorithm according to the classification results of different preset resources; S1-5-2: Use the MPO algorithm to build an initial meta-strategy optimization module, input several historical user portraits, train under several subtasks, obtain the final meta-strategy optimization module, and generate several historical meta-strategy optimization experiences; S1-5-3: Use the final meta-strategy optimization module to output the historical initial network parameters of the DQN algorithm, and use the DQN algorithm to build the initial reinforcement learning module based on the historical initial network parameters; S1-5-4: Define the state space based on historical user portraits and historical resource data characteristics, define the action space based on preset medical education actions, and set the reward function; The state space includes user portrait features (such as medical education field, knowledge level, teaching style, teaching audience, teaching progress, etc.), content features of medical education resources (such as resource classification results, difficulty, type, discipline, etc.); the action space includes the content of medical education resources recommended to users (such as courses, textbooks, videos, exercises, etc.) and the teaching plan of medical education resources (such as teaching methods, teaching skills, course arrangements, etc.); the reward function is used to evaluate the impact of recommended medical education resources on the user's medical teaching effect, such as the user's knowledge mastery, learning satisfaction, etc.; S1-5-5: Based on the state space, action space and reward function, several historical user portraits and several historical resource data features are input to train the agent and deep Q network of the initial reinforcement learning module to obtain the final reinforcement learning module and generate several historical reinforcement learning experiences; S1-5-6: storing several historical meta-strategy optimization experiences and several historical reinforcement learning experiences in the experience replay pool of the final reinforcement learning module, and integrating the final meta-strategy optimization module and the final reinforcement learning module to obtain a medical education guidance model; S1-6: Based on several historical model parameters of the medical education resource classification model, user portrait generation model, and medical education guidance model during the construction process, a parameter optimization model is constructed using a swarm intelligence optimization algorithm; The parameter optimization model is constructed based on the Improved Crested Porcupine Optimizer (ICPO) algorithm, and the parameter optimization model includes a vector encoding module, an iterative optimization module, a vector decoding module, and a parameter optimization module connected in sequence; S2: Using the medical education resource classification model, feature extraction and classification of a number of medical education resource data in the medical education resource library are performed to obtain a number of real-time resource data features and a number of real-time resource classification results, including the following steps: S2-1: Continuously collect a number of medical education resource data from the Internet, and store the collected medical education resource data in the medical education resource library; S2-2: After the resource classification cycle is reached, the unstructured content feature extraction module of the medical education resource classification model is used to extract the real-time unstructured content features of the unstructured data in the medical education resource data, and the structured content feature extraction module is used to extract the real-time structured content features of the structured data in the medical education resource data; S2-3: Use the medical education resource classification module to fuse the real-time unstructured content features and the real-time structured content features to obtain the corresponding real-time resource data features; S2-4: using the medical education resource classification module, classifying the corresponding medical education resource data according to the real-time resource data characteristics, and obtaining the real-time resource classification result; S2-5: traverse all medical education resource data in the medical education resource library to obtain several real-time resource data features and several real-time resource classification results; S3: Generate a user portrait using a user portrait generation model based on the user's real-time basic information and real-time behavior data sent by the user terminal to obtain a real-time user portrait of the user, including the following steps: S3-1: User terminal, collects the user's real-time basic information and real-time behavior data, and encrypts and uploads the real-time basic information and real-time behavior data to the cloud data center; S3-2: Cloud data center, which associates the user's real-time basic information and real-time behavior data to obtain real-time user data and inputs it into the user portrait generation model; S3-3: according to the key feature importance score generated by the key feature screening module in the construction process, the attention weight value of the attention weighting module of the user portrait generation model is set, and the user feature extraction module of the user portrait generation model is used to extract the real-time user data features of the real-time user data; S3-4: weighting the real-time user data features according to the attention weight value of the attention weighting module, and generating a user portrait using the user portrait generation module according to the obtained real-time weighted features to obtain a real-time user portrait of the user; S4: matching is performed among several real-time resource classification results according to the real-time user portrait, and several matching real-time resource data features corresponding to the successfully matched matching real-time resource classification results are obtained, including the following steps: S4-1: Obtain the real-time matching degree between each real-time resource classification result and the real-time user portrait, and take the real-time resource classification result with the highest real-time matching degree as the successfully matched real-time resource classification result; S4-2: using the real-time resource data features corresponding to the matching real-time resource classification results as matching real-time resource data features, and obtaining a plurality of matching real-time resource data features; S5: Based on the real-time user portrait and several matching real-time resource data features, the medical education guidance model is used to provide medical education guidance, and a real-time medical education strategy and several target real-time resource data features are obtained, including the following steps: S5-1: According to the real-time user portrait, a number of historical meta-strategy optimization experiences stored in the experience playback pool of the medical education guidance model are matched to obtain the real-time meta-strategy optimization experience with successful matching; S5-2: Based on the real-time meta-strategy optimization experience, the meta-strategy optimization module is used to generate real-time initialization network parameters, and the reinforcement learning module is adjusted according to the initialization network parameters to obtain the adjusted reinforcement learning module; S5-3: According to the real-time user portrait, a number of historical reinforcement learning experiences stored in the experience replay pool of the medical education guidance model are matched to obtain the real-time reinforcement learning experience with successful matching; S5-4: Update the action space of the reinforcement learning module according to the real-time reinforcement learning experience to obtain an updated action space ,in, For the updated action, is the action indicator, is the total number of action space dimensions; S5-5: Update the state space of the reinforcement learning module according to the real-time user portrait and several matching real-time resource data features to obtain an updated state space ,in, For the updated state, is the state indicator, is the total number of state space dimensions; S5-6: According to the updated state space and the updated action space, using the adjusted reinforcement learning module, medical education guidance is performed to obtain a real-time medical education strategy, including the following steps: S5-6-1: Input the updated state space into the adjusted reinforcement learning module, and use the adjusted deep Q network to generate the Q value of each possible action in the updated action space based on the updated state space; The formula is:

[0025] In the formula, For updated status and update actions The corresponding updated Q value; Status and actions The corresponding predicted Q value; is the learning rate; is the highest predicted Q value; It is a comprehensive indicator; To update the parameters; Action-based The status Transition to the updated state The reward value of S5-6-2: Repeat the above Q value update steps until the number of iterations reaches the iteration number threshold, use the greedy strategy to select the possible action with the highest updated Q value as the execution action, and obtain the real-time medical education strategy based on the execution action; S5-7: searching among a number of matching real-time resource data features according to the real-time medical education strategy to obtain a number of target real-time resource data features; S6: searching in the medical education resource library according to the characteristics of the target real-time resource data, and sending the real-time medical education strategy and the obtained target medical education resource data to the user terminal; S7: According to the real-time feedback data sent by the user terminal, the parameter optimization model is used to optimize the parameters of the medical education resource classification model, the user portrait generation model, and the medical education guidance model, including the following steps: S7-1: according to the real-time feedback data sent by the user terminal, set the optimization target of the parameter optimization model, and set the fitness function of the parameter optimization model according to the optimization target; Real-time feedback data includes the accuracy score of the medical education resource classification model, the response efficiency score of the user portrait generation model, and the matching score of the medical education guidance model. If the score is lower than the threshold, the corresponding model needs to be optimized. The formula is:

[0026] In the formula, To optimize the objective function; The accuracy of the model for classifying medical education resources; Response efficiency time for user portrait generation model; The fit of the model for medical education guidance; is the weight factor adjusted according to the feedback data; Refers to the parameters for the solution; Refers to parameters for feedback data; is the weight value;

[0027] In the formula, is the fitness function; S7-2: Encode the comprehensive model parameter scheme of the medical education resource classification model, the user profile generation model, and the medical education guidance model into the individual vector format of the parameter optimization model; S7-3: According to the individual vector format and fitness function, the parameter optimization model is used to perform iterative optimization, and the individual vector of the optimal individual is decoded to generate the optimal comprehensive real-time model parameter solution, including the following steps: S7-3-1: According to the optimization objective function, set the fitness function of the parameter optimization model, and set the ICPO population parameters and maximum number of iterations of the parameter optimization model; S7-3-2: According to the individual vector format and ICPO population parameters, the Circle chaotic mapping sequence is used to initialize the ICPO population to obtain an initial ICPO population containing several initial ICPO individuals; The formula is:

[0028] In the formula, is the initial ICPO individual of the Circle chaos map, corresponding to an initial solution; is the randomly generated initial ICPO individual; It is the ICPO individual indicator; S7-3-3: Introduce a cyclic population reduction mechanism to limit the number of individuals in the ICPO population parameters and obtain updated ICPO population parameters for the next iteration; The formula is:

[0029] In the formula, For the The number of individuals in the ICPO population parameter of the iteration; For the The number of individuals in the ICPO population parameter of the iteration; is the minimum number of individuals in the ICPO population parameter; Evaluate arguments for functions; Evaluate loop parameters for a function; Evaluate loop parameters for the maximum function; t is the indicator of the number of iterations; S7-3-4: Calculate the initial fitness value of the initial ICPO individuals in the initial ICPO population according to the fitness function; S7-3-5: According to the initial fitness value and the updated ICPO population parameters, the first defense strategy, the second defense strategy, the third defense strategy and the fourth defense strategy are used to update the initial ICPO population to obtain an updated ICPO population; The formula for the first defense strategy is:

[0030] In the formula, For the updated ICPO individuals within the first defense range; The initial ICPO individual within the first defense range; is a random number based on normal distribution; is a random value in the interval [0,1]; It is the optimal solution within the first defense range; is the vector generated between the true optimal solution within the first defense range and the optimal solution randomly selected from the ICPO population; It is the ICPO individual indicator; is the iteration indicator; The formula for the second defense strategy is:

[0031] In the formula, For the updated ICPO individuals within the second defense range; The initial ICPO individual within the second defense range; is the search upper limit vector of the second defense range; is a random value in the interval [0,1]; Respectively Initial ICPO individuals; All are [1, ]; is the vector generated between the true optimal solution within the second defense range and the optimal solution randomly selected from the ICPO population; The formula for the third defense strategy is:

[0032] In the formula, For the updated ICPO individuals within the third defense range; The initial ICPO individual within the third defense range; is the search upper limit vector of the third defense range; Respectively Initial ICPO individuals; is [1, ]; The odor diffusion factor defined for the fitness function; It is a defense factor; Control parameters for search direction; The formula for the fourth defense strategy is:

[0033] In the formula, For the updated ICPO individuals within the fourth defense range; The initial ICPO individual within the fourth defense range; It is the optimal solution within the fourth defense range; All are random values ​​in the interval [0,1]; It is a defense factor; Control parameters for search direction; is the average force affecting the search direction; is the convergence speed factor; S7-3-6: Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the updated ICPO population to generate a dynamic reverse ICPO population; The formula is:

[0034] In the formula, It is a dynamically reversed ICPO individual; is the decreasing inertia coefficient; are the maximum and minimum values ​​of the vector space respectively; For the updated ICPO individuals; S7-3-7: According to the fitness function, calculate the fitness values ​​of all ICPO individuals in the updated ICPO population and the dynamically reversed ICPO population, take the ICPO individual with the minimum fitness value as the optimal individual, and retain the optimal individual; S7-3-8: If the number of iterations of iterative optimization reaches the maximum number of iterations or the fitness value of the optimal individual meets the requirements, the optimal solution corresponding to the optimal individual is output; S7-3-9: Decode the individual vectors of the optimal solution to generate the optimal comprehensive real-time model parameter solution; The comprehensive real-time model parameter scheme includes first real-time model parameters of a medical education resource classification model, second real-time model parameters of a user portrait generation model, and third real-time model parameters of a medical education guidance model; S7-4: Based on the optimal comprehensive real-time model parameter scheme, optimize the parameters of the medical education resource classification model, user portrait generation model, and medical education guidance model.

[0035] Embodiment 2: like Figure 2 As shown, this embodiment provides an artificial intelligence-based medical education guidance system for implementing a medical education guidance method. The system includes a cloud data center and a plurality of user terminals, each of which is in communication connection with the cloud data center; The user terminal is used to collect the user's real-time basic information, real-time behavior data, and real-time feedback data, and encrypt and upload them to the cloud data center; receive the real-time medical education strategy and several target medical education resource data sent by the cloud data center; The cloud data center includes a model building unit, a medical education resource classification unit, a user portrait generation unit, a resource classification matching unit, a medical education guidance unit, a medical education resource sending unit, and a parameter optimization unit which are connected in sequence; A model building unit is used to set up a medical education resource library and use artificial intelligence algorithms to build a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model; A medical education resource classification unit is used to extract and classify features of a number of medical education resource data in a medical education resource library using a medical education resource classification model, and obtain a number of real-time resource data features and a number of real-time resource classification results; A user portrait generation unit is used to generate a user portrait using a user portrait generation model according to the real-time basic information and real-time behavior data of the user sent by the user terminal, so as to obtain a real-time user portrait of the user; A resource classification matching unit is used to match a number of real-time resource classification results according to the real-time user portrait, and obtain a number of matching real-time resource data features corresponding to the successfully matched real-time resource classification results; A medical education guidance unit is used to provide medical education guidance using a medical education guidance model based on real-time user portraits and a number of matching real-time resource data features, and obtain a real-time medical education strategy and a number of target real-time resource data features; A medical education resource sending unit, used to search in a medical education resource library according to characteristics of a number of target real-time resource data, and send the real-time medical education strategy and the obtained number of target medical education resource data to a user terminal; The parameter optimization unit is used to optimize the parameters of the medical education resource classification model, the user portrait generation model and the medical education guidance model using the parameter optimization model according to the real-time feedback data sent by the user terminal.

[0036] The present invention discloses a medical education guidance method and system based on artificial intelligence, which provides a unified medical education guidance program based on a cloud data center, promotes the popularization and balance of medical education resources, avoids dependence on medical education experts, reduces the cost investment of medical education guidance, improves efficiency, and can be promoted in different regions and different medical education fields; the medical education resource library realizes the collection and aggregation of medical education resources, improves the richness of medical education resources, the medical education resource classification model provides powerful data processing capabilities, performs automatic resource classification, and improves the accuracy of resource retrieval and recommendation; the user portrait generation model can more accurately capture the characteristics of users and provide refined data support for subsequent education resource recommendations; the medical education guidance model can adapt to the user's behavioral characteristics, teaching progress, resource difficulty, etc., and provide accurate customized medical education strategies; the parameter optimization model realizes the continuous evolution of the artificial intelligence model according to user feedback; by constructing an automated medical education guidance system, the intelligence level of medical education guidance is enhanced and the resource utilization rate is improved.

[0037] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A medical education guidance method based on artificial intelligence, characterized by: The steps include: Set up a medical education resource library in the cloud data center and use artificial intelligence algorithms to build a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model; Using the medical education resource classification model, feature extraction and classification are performed on a number of medical education resource data in the medical education resource library to obtain a number of real-time resource data features and a number of real-time resource classification results; Based on the real-time basic information and real-time behavior data of the user sent by the user terminal, a user portrait generation model is used to generate a user portrait to obtain a real-time user portrait of the user; According to the real-time user portrait, matching is performed among several real-time resource classification results, and several matching real-time resource data features corresponding to the successfully matched matching real-time resource classification results are obtained; According to the real-time user portrait and several matching real-time resource data features, the medical education guidance model is used to provide medical education guidance, and the real-time medical education strategy and several target real-time resource data features are obtained; According to the characteristics of several target real-time resource data, searching is performed in the medical education resource library, and the real-time medical education strategy and the obtained several target medical education resource data are sent to the user terminal; According to the real-time feedback data sent by the user terminal, the parameter optimization model is used to optimize the parameters of the medical education resource classification model, the user portrait generation model and the medical education guidance model.

2. The medical education guidance method based on artificial intelligence according to claim 1, characterized in that: A medical education resource library is set up in the cloud data center, and artificial intelligence algorithms are used to build a medical education resource classification model, a user portrait generation model, a medical education guidance model, and a parameter optimization model, including the following steps: Collect a number of heterogeneous medical education resource data in different medical fields, and transform the data structure according to the preset data structure of the cloud data center to obtain a number of homogeneous medical education resource data; Preprocess a number of isomorphic medical education resource data to obtain a number of final medical education resource data, summarize and organize them, and build a medical education resource library; Based on some final medical education resource data, a content feature extraction and classification algorithm based on deep learning is used to build a medical education resource classification model and generate some historical resource data features; Based on the historical basic information and historical behavior data of several users, a user portrait generation model is constructed using a key feature screening and portrait generation algorithm based on deep learning, and several historical user portraits are generated; Based on several historical user portraits and several historical resource data features, a medical education guidance model is constructed using an enhanced reinforcement learning algorithm; According to several historical model parameters of the medical education resource classification model, user portrait generation model and medical education guidance model in the construction process, a parameter optimization model was constructed using a swarm intelligence optimization algorithm.

3. The medical education guidance method based on artificial intelligence according to claim 2, characterized in that: The medical education resource classification model is constructed based on the CNN-LSTM-MLP algorithm; The user portrait generation model is constructed based on the LSTM-RF-Attention-MLP algorithm; The medical education guidance model is constructed based on the MPO-DQN algorithm; The parameter optimization model is constructed based on the ICPO algorithm.

4. The medical education guidance method based on artificial intelligence according to claim 3, characterized in that: Based on several historical user portraits and several historical resource data features, an enhanced reinforcement learning algorithm is used to build a medical education guidance model, including the following steps: The problem of generating medical education strategies is used as a simulation environment for the DQN algorithm, and the network parameter generation problem of the DQN algorithm is divided into several subtasks of the MPO algorithm according to the classification results of different preset resources. Use the MPO algorithm to build the initial meta-strategy optimization module, input several historical user portraits, train under several subtasks, obtain the final meta-strategy optimization module, and generate several historical meta-strategy optimization experiences; Use the final meta-strategy optimization module to output the historical initial network parameters of the DQN algorithm, and use the DQN algorithm to build the initial reinforcement learning module based on the historical initial network parameters; Define the state space based on historical user portraits and historical resource data characteristics, define the action space based on preset medical education actions, and set the reward function; Based on the state space, action space and reward function, several historical user portraits and several historical resource data features are input to train the agent and deep Q network of the initial reinforcement learning module to obtain the final reinforcement learning module and generate several historical reinforcement learning experiences; Several historical meta-strategy optimization experiences and several historical reinforcement learning experiences are stored in the experience replay pool of the final reinforcement learning module, and the final meta-strategy optimization module and the final reinforcement learning module are integrated to obtain a medical education guidance model.

5. The medical education guidance method based on artificial intelligence according to claim 3, characterized in that: Using the medical education resource classification model, feature extraction and classification are performed on a number of medical education resource data in the medical education resource library to obtain a number of real-time resource data features and a number of real-time resource classification results, including the following steps: Continuously collect a number of medical education resource data from the Internet, and store the collected medical education resource data in a medical education resource library; After the resource classification cycle is reached, the medical education resource classification model is used to extract the real-time unstructured content features and real-time structured content features of the medical education resource data; Perform feature fusion on real-time unstructured content features and real-time structured content features to obtain corresponding real-time resource data features; According to the characteristics of real-time resource data, the corresponding medical education resource data is classified to obtain the real-time resource classification results; All medical education resource data in the medical education resource library are traversed to obtain several real-time resource data features and several real-time resource classification results.

6. The medical education guidance method based on artificial intelligence according to claim 3, characterized in that: According to the real-time basic information and real-time behavior data of the user sent by the user terminal, the user portrait generation model is used to generate the user portrait to obtain the real-time user portrait of the user, including the following steps: User terminals collect real-time basic information and real-time behavior data of users, and encrypt and upload the real-time basic information and real-time behavior data to the cloud data center; The cloud data center associates the user's real-time basic information with the real-time behavior data to obtain real-time user data and inputs it into the user portrait generation model; According to the key feature importance scores generated during the construction process, the attention weight values ​​of the user portrait generation model are set, and the user portrait generation model is used to extract the real-time user data features of the real-time user data; According to the attention weight value, the real-time user data features are weighted, and based on the obtained real-time weighted features, a user portrait is generated to obtain a real-time user portrait of the user.

7. The medical education guidance method based on artificial intelligence according to claim 5, characterized in that: According to the real-time user portrait, matching is performed among several real-time resource classification results, and several matching real-time resource data features corresponding to the successfully matched matching real-time resource classification results are obtained, including the following steps: Obtain the real-time matching degree between each real-time resource classification result and the real-time user portrait, and take the real-time resource classification result with the highest real-time matching degree as the successfully matched real-time resource classification result; The real-time resource data features corresponding to the matching real-time resource classification results are used as matching real-time resource data features to obtain a plurality of matching real-time resource data features.

8. The medical education guidance method based on artificial intelligence according to claim 4, characterized in that: According to the real-time user portrait and several matching real-time resource data features, the medical education guidance model is used to provide medical education guidance, and the real-time medical education strategy and several target real-time resource data features are obtained, including the following steps: According to the real-time user portrait, a number of historical meta-strategy optimization experiences stored in the experience playback pool of the medical education guidance model are matched to obtain the real-time meta-strategy optimization experience with successful matching; According to the real-time meta-strategy optimization experience, the meta-strategy optimization module is used to generate real-time initialization network parameters, and the reinforcement learning module is adjusted according to the initialization network parameters to obtain the adjusted reinforcement learning module; According to the real-time user portrait, a number of historical reinforcement learning experiences stored in the experience replay pool of the medical education guidance model are matched to obtain the real-time reinforcement learning experience with successful matching; According to the real-time reinforcement learning experience, the action space of the reinforcement learning module is updated and adjusted to obtain an updated action space; According to the real-time user portrait and several matching real-time resource data features, the state space of the reinforcement learning module is updated and adjusted to obtain an updated state space; According to the updated state space and the updated action space, the adjusted reinforcement learning module is used to provide medical education guidance and obtain a real-time medical education strategy; According to the real-time medical education strategy, a search is performed among several matching real-time resource data features to obtain several target real-time resource data features.

9. The medical education guidance method based on artificial intelligence according to claim 3, characterized in that: According to the real-time feedback data sent by the user terminal, the parameter optimization model is used to optimize the parameters of the medical education resource classification model, the user portrait generation model and the medical education guidance model, including the following steps: According to the real-time feedback data sent by the user terminal, the optimization target of the parameter optimization model is set, and according to the optimization target, the fitness function of the parameter optimization model is set; Encoding the comprehensive model parameter schemes of the medical education resource classification model, the user profile generation model, and the medical education guidance model into the individual vector format of the parameter optimization model; According to the individual vector format and fitness function, a parameter optimization model is used to perform iterative optimization, and the individual vector of the optimal individual is decoded to generate the optimal comprehensive real-time model parameter solution; According to the optimal comprehensive real-time model parameter scheme, the parameters of the medical education resource classification model, user portrait generation model and medical education guidance model are optimized.

10. A medical education guidance system based on artificial intelligence, used to implement the medical education guidance method according to any one of claims 1 to 9, characterized in that: The system includes a cloud data center and a plurality of user terminals, each of which is in communication connection with the cloud data center; The cloud data center includes a model building unit, a medical education resource classification unit, a user portrait generation unit, a resource classification matching unit, a medical education guidance unit, a medical education resource sending unit and a parameter optimization unit which are connected in sequence.

Images