Teaching case recommendation method and device, electronic terminal and storage medium

Abstract

The invention belongs to the technical field of digital medical treatment, and discloses a teaching case recommendation method and device, an electronic terminal and a storage medium, and the method comprises the steps: obtaining user data of a target user, and carrying out the preprocessing of the user data, and obtaining a user feature vector; inputting the user feature vector into a pre-trained case recommendation model, calculating the similarity between the target user and other users, searching for the nearest neighbor of the target user, evaluating the preference degree of the target user for the case difficulty level and the prediction score of the target user for the case, and recommending the case. And recommending the teaching case to the target user according to the prediction score. Teaching case recommendation is realized based on the user similarity, the basic information and learning process information of the user on the teaching platform are effectively utilized, the safety of case information is better protected, the recommendation result is more accurate, and the requirements of the target user can be better met.

Description

Technical Field

[0001] The present invention belongs to the field of digital medical technology, relates to medical education informatization technology, and particularly relates to a teaching case recommendation method, device, electronic terminal, and storage medium. Background Art

[0002] In order to share medical resources and narrow the gap in medical levels between regions, a digital standardized training teaching platform has emerged. Through this platform, the traditional standardized training learning process can be digitized, greatly improving the teaching efficiency of hospitals and the learning experience of users. As time goes by, a large amount of information will be generated when users learn on the platform, such as the basic information of users, case resources, and the learning history records of trainees. How to mine and analyze the massive data generated during the user's use process and then provide more personalized services for users has important research significance and application prospects.

[0003] If the platform does not have a learning case recommendation function, it will increase the mental burden of users when using it and cannot provide personalized services for users. Currently, similar case recommendation methods mainly use structured data such as the age and gender of patients in cases or unstructured data such as text data to construct case representation vectors, and then use deep learning or content-based collaborative filtering algorithms to find cases similar to the target case representation vector for recommendation. However, these recommendation methods are based on case information, but hospitals generally encrypt case information for information security considerations, resulting in a significant reduction in the effectiveness of these recommendation methods. Therefore, there is an urgent need for a teaching case recommendation method to solve the above technical problems. Summary of the Invention

[0004] Technical Objectives: Aiming at the above technical problems, the present invention proposes a teaching case recommendation method, device, electronic terminal, and storage medium.

[0005] Technical Solutions: To achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] A teaching case recommendation method, characterized by including the steps of:

[0007] Obtain the user data of the target user and perform preprocessing to obtain a user feature vector;

[0008] Input the user feature vector into a pre-trained case recommendation model to obtain the predicted score of the target user and the teaching cases recommended according to the predicted score;

[0009] The case recommendation model executes the following process:

[0010] Calculate the similarity between the target user and other users according to the user feature vector;

[0011] Find the nearest neighbors of the target user through the similarity;

[0012] Evaluate the preference degree of the target user for the difficulty level of cases;

[0013] Based on the user-case access frequency matrix in the case recommendation model, the nearest neighbors of the target user, and the preference degree of the target user for the difficulty level of cases, calculate the predicted score of the target user for cases, and recommend teaching cases to the target user according to the predicted score.

[0014] Preferably, calculate the similarity between the target user and other users by the following method:

[0015] Collect user data;

[0016] Preprocess the user data to obtain user feature vectors;

[0017] Cluster the user feature vectors to obtain user clustering results;

[0018] Construct an m×n order user-case access frequency matrix A(m,n) according to the user teaching case access data in the user data, where m rows represent m users and n columns represent n teaching cases;

[0019] Identify the identity information of the target user, and calculate the similarity between the target user and other users through the Pearson correlation coefficient based on the user-case access frequency matrix and the user clustering results.

[0020] Preferably, clustering the user feature vectors to obtain user clustering results includes the steps of:

[0021] Calculate the Euclidean distance between the user feature vector and the preset initial centroid vector;

[0022] Assign each user feature vector to the nearest clustering center, and iteratively optimize until the distance from the user feature vector to the clustering center converges to obtain user clustering results.

[0023] Preferably, calculate the similarity between the target user and other users through the Pearson correlation coefficient, and calculate according to formula (2):

[0024]

[0025] Among them, x represents the target user, y represents other users, sim(x,y) represents the similarity between the target user x and other users y, I xy represents the set of cases jointly accessed by the target user x and other users y in the user-case access frequency matrix, c represents a specific case in I xy inx,c and R y,c respectively represent the number of visits of the target user x and other user y to the case c, and respectively represent the average number of visits of the target user x and other user y to the case, G x,y represents the preset clustering coefficient between the target user x and other user y.

[0026] Preferably, the TF-IDF statistical method is used to evaluate the preference degree of the user for the difficulty level of the case, and the preference degree F of the user u for the difficulty level d is calculated according to Equations (3) to (5) u,d :

[0027] F u,d = 1 + TF u,d × IDF u,d (3)

[0028]

[0029] wherein, TF u,d and IDF u,d respectively represent the TF value and the IDF value of the difficulty level d in the access history of the user u; |{u.c k :c k .d’ = d}| represents the number of times the user u browses the cases with the difficulty level d, c k represents the kth case visited by the user, d’ represents the difficulty level corresponding to the case c k , |{v.c k :c k .d’ = d}| represents the number of times the user v browses the cases with the difficulty level d, |R u | and |R v | respectively represent the total number of case views of the users u and v, and U represents the set of all users.

[0030] Preferably, based on the user-case access frequency matrix in the case recommendation model and the nearest neighbors of the target user, the predicted score P of the target user u for the case is calculated according to Equation (6) u,i :

[0031]

[0032] wherein, u represents the target user, i represents the case for which the score is to be predicted, N u represents the set of the nearest neighbors of the target user u, n represents a certain user in N u , sim(u, n) represents the similarity between the target user u and the user n, R n,i represents the number of visits of the target user u to the case i, represents the average number of visits of the target user u to all cases, represents the average number of visits of user n to all cases, F u,i.d represents the preference degree of user u for the case difficulty level i.d.

[0033] A teaching case recommendation device, characterized by comprising:

[0034] A collection and preprocessing module, configured to obtain user data of a target user and perform preprocessing to obtain a user feature vector;

[0035] A medical record recommendation model, configured to input the user feature vector, output a predicted score, and recommend teaching cases according to the predicted score;

[0036] Wherein, the medical record recommendation model includes:

[0037] A user clustering module, configured to calculate the similarity between the target user and other users according to the user feature vector;

[0038] A searching module, configured to search for the nearest neighbor of the target user through the similarity;

[0039] An evaluation module, configured to evaluate the preference degree of the target user for the case difficulty level;

[0040] A recommendation module, configured to calculate a predicted score of the target user for the teaching cases to be recommended and rank the predicted scores based on the user-case access frequency matrix in the case recommendation model, the nearest neighbor of the target user, and the preference degree of the target user for the case difficulty level, and recommend teaching cases to the target user according to the predicted scores.

[0041] An electronic terminal, including a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method are executed.

[0042] A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method are implemented.

[0043] A computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method are implemented.

[0044] Beneficial effects: Due to the adoption of the above technical solutions, the present invention has the following beneficial effects:

[0045] The present invention realizes the recommendation of teaching cases based on user similarity, effectively utilizes the basic information and learning process information of users on the teaching platform, better protects the security of case information, and considers the user clustering results when calculating user similarity, so that the recommendation results are more accurate and better meet the needs of target users. Description of the Drawings

[0046] Figure 1 is a flowchart of a teaching case recommendation method provided in the first embodiment of the present invention;

[0047] Figure 2 is a complete execution flowchart of a teaching case recommendation method provided in the first embodiment of the present invention;

[0048] Figure 3 is an architecture diagram of a teaching case recommendation method provided in the first embodiment of the present invention. Detailed Embodiments

[0049] The technical solution of the present invention will be described in detail below through the drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present invention and the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. Without conflict, the technical features in the embodiments of the present invention and the embodiments can be combined with each other.

[0050] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0051] Embodiment 1:

[0052] Figure 1 is a flowchart of the teaching case recommendation method in the first embodiment of the present invention. This flowchart only shows the logical order of the method described in this embodiment. On the premise of no conflict, in other possible embodiments of the present invention, the steps shown or described can be completed in a different Figure 1 order as shown.

[0053] The teaching case recommendation method provided in this embodiment can be applied to a terminal and can be executed by a teaching case recommendation device. The device can be implemented in a software and / or hardware manner and can be integrated in the terminal. For example: any smart phone, tablet computer or computer device with communication functions. Refer to Figure 1 、 Figure 2 and Figure 3 As shown, the method of this embodiment specifically includes the following steps:

[0054] Step 1: Obtain the identity information of the target user;

[0055] Step 2: Input the identity information of the target user into a pre-trained case recommendation model to obtain the similarity between the target user and other users;

[0056] Step 3: Find the nearest neighbors of the target user through the similarity:

[0057] That is, for each target user x, find the user set C = {C1, C2, …, C k} in the overall user space, where k represents the number of users with the highest similarity to the target user x, and the similarity sim(x, C1) between C1 and x is the highest, the similarity sim(x, C2) between C2 and x is the second highest, and so on;

[0058] Step 4: Infer the difficulty level preference of the user based on the difficulty level of the cases accessed by the target user;

[0059] Step 5: Calculate the predicted score of the target user for the case based on the user-case access frequency matrix in the case recommendation model and the nearest neighbors of the target user:

[0060] The calculation formula for calculating the predicted score of the target user for the case based on the user-case access frequency matrix in the case recommendation model, the nearest neighbors of the target user, and the inferred difficulty level preference includes:

[0061]

[0062] where u represents the target user, i represents the case for which the score is to be predicted, N u represents the set of the nearest neighbors of the target user u, n represents a certain user in N u , sim(u, n) represents the similarity between the target user u and the user n, R n,i represents the number of times the target user u accesses the case i, represents the average number of times the target user u accesses all cases, represents the average number of times the user n accesses all cases, F u,i.d represents the preference degree of the user u for the difficulty level i.d of the case. If the target user has no access record, the difficulty level preference is set to the reference value 1; i represents the case i, and i.d represents the difficulty level corresponding to the case i. For a certain case, its difficulty level is a definite value.

[0063] Recommend the top N teaching cases with high predicted scores to the target user, where the specific value of N can be set according to actual needs.

[0064] Among them, the training process of the case recommendation model includes:

[0065] Collect user data, and obtain data such as the number of visits to teaching cases, gender, age, grade, major, professional and technical title, name of the teaching teacher, number of course previews, course attendance rate, and average score of course questions for each user from the teaching platform database;

[0066] Preprocess the user data:

[0067] Not all data in the user data contributes to the recommendation. For example, mobile phone numbers, employee numbers, and attendance machine numbers are filtered;

[0068] For text data such as gender, major, professional and technical title, and teaching teacher, convert it into numerical data. Specifically, map male and female genders to {1, 2}, map majors 1, 2,..., N to {1, 2,..., N}, map professional and technical titles 1, 2,..., N to {1, 2,..., N}, and map teaching teachers 1, 2,..., N to {1, 2,..., N};

[0069] The measurement units and ranges of the above user data are different, and the distribution of the data may deviate, which may affect the convergence speed and accuracy of the clustering algorithm model in the subsequent steps. To solve this problem, the data needs to be normalized. When normalizing the data, first replace the outliers with the mean value, and then use the Min-Max Normalization method, that is, linearly map the data to the [0, 1] interval. The formula is as follows:

[0070] X_normallized = (X - X_min) / (X_max - X_min),

[0071] Where X_min and X_max represent the minimum and maximum values of the data respectively, X represents the original data, and X_normallized represents the normalized data;

[0072] Convert the normalized user feature data into a two-dimensional user feature vector. One dimension A represents the basic information of the user, and the other dimension B represents the learning process information of the user on the medical teaching platform. The conversion method is: sum up the normalized attribute values representing the basic information of the user as vector dimension A, sum up the normalized attribute values representing the learning process information of the user as vector dimension B, and finally obtain the user feature vector;

[0073] Cluster the user feature vector to obtain the user clustering result;

[0074] Take all user feature vectors as a sample set, and randomly select k samples from the sample set as the initial centroid vectors.

[0075] Calculate the spatial distance between the user feature vector and the preset initial centroid vector through the Euclidean distance method:

[0076]

[0077] Among them, x represents the user feature vector, y represents the initial centroid vector, i and j respectively represent the user basic information dimension and the learning process information dimension in the user feature vector, x i and x j respectively represent the abscissa and ordinate of the user feature vector, y i and y j respectively represent the abscissa and ordinate of the initial centroid vector, d s represents the Euclidean distance between the user feature vector and the initial centroid vector;

[0078] Assign each of the user feature vectors to the clustering center closest to it. The clustering center and the sample vectors assigned to them represent a cluster. Iteratively optimize the above clustering until the distance from the user feature vector to the clustering center converges to obtain the user clustering result;

[0079] Construct a user-case access frequency matrix based on the user teaching case access data in the user data, that is, construct a user-case access frequency matrix based on the user teaching case access data in the user data. The user-case access frequency matrix can be represented by an m×n order matrix A(m,n). The m rows represent m users, and the n columns represent n teaching cases. The element V p,q in the p-th row and q-th column represents the number of times user p accesses teaching case q. This matrix is constructed only based on the number of times users access teaching cases, so it will not be affected by the problem of case information encryption.

[0080] Identify the identity information of the target user. Based on the user-case access frequency matrix and the user clustering result, calculate the similarity between the target user and other users through the Pearson correlation coefficient. The calculation formula is as follows:

[0081]

[0082] Among them, x represents the target user, y represents other users, sim(x,y) represents the similarity between the target user x and other users y, I xy represents the set of cases jointly accessed by the target user x and other users y in the user-case access frequency matrix, c represents a specific case in I xy , R x,c and Ry,c respectively represent the number of accesses of the target user x and other user y to the case c, and respectively represent the average number of accesses of the target user x and other user y to the case, G x,y represents the clustering coefficient between the target user x and other user y. If the target user x and other user y are not in the same cluster, then this coefficient is a reference value (such as 1 or 10). If they are in the same cluster, then this coefficient is a multiple of the reference value (should be greater than the reference value, such as a value between 1.2 and 5).

[0083] The case difficulty level does not belong to the case privacy information and usually includes simple, medium, difficult, rare, etc. Intuitively, if a user likes to access cases of a certain difficulty level, then the user will access more cases belonging to that difficulty level. In addition, if a user accesses cases belonging to a certain difficulty level that are rarely accessed by others, then the user may be more partial to that difficulty level. The TF-IDF statistical method can reflect the above-mentioned user preference psychology and is applicable to evaluating the user's preference degree for the difficulty level. Among them, the user's case access history record is regarded as a document, the difficulty level is regarded as a word in the document, and the preference degree F of user u for difficulty level d u,d is as follows:

[0084] F u,d = 1 + TF u,d × IDF u,d

[0085] TF in the equation u,d and IDF u,d respectively represent the TF value and IDF value of difficulty level d in the access history of user u, and are calculated by the following formula:

[0086]

[0087] where |{u.c k :c k .d’ = d}| represents the number of times user u browses cases with difficulty level d, c k represents a certain case accessed by the user, k represents the kth case among the cases accessed by the user, d’ represents the difficulty level corresponding to case c k , |{v.c k :c k .d’ = d}| represents the number of times user v browses cases with difficulty level d, |R u | and |R v | respectively represent the total number of case views of user u and v, and U represents the set of all users.

[0088] Since the case information is encrypted due to the information security requirements of the hospital, the recommendation effect of the method based on case similarity will be greatly reduced. The recommendation method based on user similarity in the present invention can effectively avoid this problem. The teaching case recommendation method described in the present invention effectively utilizes the basic information and learning process information of users on the teaching platform, fully considers multiple dimensions of patient information, and takes into account the user clustering results when calculating user similarity, so that the recommendation results are more in line with user interests.

[0089] Embodiment 2:

[0090] The second embodiment of the present invention provides a teaching case recommendation device, including:

[0091] An acquisition and preprocessing module, configured to obtain user data of a target user and perform preprocessing to obtain a user feature vector;

[0092] A medical record recommendation model, configured to input the user feature vector and output a predicted score and a teaching case recommended according to the predicted score;

[0093] Wherein, the medical record recommendation model includes:

[0094] A user clustering module, configured to calculate the similarity between the target user and other users according to the user feature vector;

[0095] A searching module, configured to search for the nearest neighbor of the target user through the similarity;

[0096] An evaluation module, configured to evaluate the preference degree of the target user for the difficulty level of the case;

[0097] A recommendation module, configured to calculate the predicted score of the target user for the case based on the user-case access frequency matrix in the case recommendation model, the nearest neighbor of the target user, and the preference degree of the target user for the difficulty level of the case, and recommend the teaching case to the target user according to the predicted score.

[0098] In the case recommendation model of the present invention, a network model in machine learning algorithms can be used to implement, and an average absolute error loss function is constructed to train the case recommendation model. However, it does not specifically refer to the network model in machine learning algorithms. This "model" is a general concept and is an abstraction of the recommendation process. The input data of the model includes the basic information of the user (gender, grade, major, etc.), the access times of each teaching case, and the learning process information (course preview times, course attendance rate, etc.), and the data format is a one-dimensional vector. User similarity, the nearest neighbor, and the preference degree of the target user for the difficulty level are process data calculated during the process of the input data. The data output by the model is the predicted score and the teaching cases to be recommended obtained after ranking the predicted scores.

[0099] The teaching case recommendation device provided in the second embodiment of the present invention can execute the teaching case recommendation method provided in the first embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0100] Embodiment Three:

[0101] The third embodiment of the present invention further provides an electronic terminal, including a processor and a memory connected to the processor. A computer program is stored in the memory, and the processor is used to operate according to the instructions to execute the steps of the method described in Embodiment One, and has the corresponding functional modules and beneficial effects for executing the method.

[0102] Embodiment Four:

[0103] The fourth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in Embodiment One are implemented, and it has the corresponding functional modules and beneficial effects for executing the method.

[0104] Embodiment Five:

[0105] The fifth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method are implemented, and it has the corresponding functional modules and beneficial effects for executing the method.

[0106] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0107] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (apparatuses), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one or more of these flows and / or multiple flows and / or blocks Figure 1 one or more of these blocks and / or multiple blocks.

[0108] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the function specified in one or more of the processes and / or blocks Figure 1 one or more of the processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the function specified in one or more of the processes and / or blocks Figure 1 one or more of the processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0110] The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A teaching case recommendation method, characterized in that: Includes steps: Obtain the user data of the target user and preprocess it to obtain the user feature vector; Inputting the user feature vector into a pre-trained case recommendation model to obtain a predicted score of the target user and a teaching case recommended according to the predicted score; The case recommendation model performs the following process: Calculating the similarity between the target user and other users based on the user feature vector; Find the nearest neighbor of the target user through the similarity; Assess target users' preferences for case difficulty levels; Based on the user-case access frequency matrix in the case recommendation model, the target user's nearest neighbors, and the target user's preference for the case difficulty level, the target user's predicted score for the case is calculated, and the teaching case is recommended to the target user based on the predicted score.

2. A teaching case recommendation method according to claim 1, characterized in that: The similarity between the target user and other users is calculated as follows: Clustering the user feature vectors to obtain user clustering results; Constructing an m×n-order user-case access frequency matrix A(m,n) according to the user teaching case access data in the user data, where m rows represent m users and n columns represent n teaching cases; The identity information of the target user is identified, and based on the user-case access frequency matrix and the user clustering result, the similarity between the target user and other users is calculated by the Pearson correlation coefficient.

3. A teaching case recommendation method according to claim 2, characterized in that: Clustering the user feature vectors to obtain user clustering results includes the following steps: Calculating the Euclidean distance between the user feature vector and a preset initial centroid vector; Each of the user feature vectors is assigned to the cluster center closest to it, and iterative optimization is performed until the distance from the user feature vector to the cluster center converges, thereby obtaining a user clustering result.

4. A teaching case recommendation method according to claim 2, characterized in that: The similarity between the target user and other users is calculated by the Pearson correlation coefficient, which is calculated according to formula (2): Where x represents the target user, y represents other users, sim(x,y) represents the similarity between the target user x and other users y, and I xy represents the set of cases that the target user x and other users y have visited together in the user-case access frequency matrix, and c represents I xy In a specific case, R x,c and R y,c Respectively represent the number of visits to case c by target user x and other users y, and represent the average number of visits to the case by the target user x and other users y, G x,y Represents the clustering coefficient between the preset target user x and other users y.

5. A teaching case recommendation method according to claim 1, characterized in that: The TF-IDF statistical method is used to evaluate the user's preference for the case difficulty level. According to formulas (3) to (5), the user u's preference for difficulty level d is calculated as F u,d : F u,d =1+TF u,d ×IDF u,d (3) Among them, TF u,d and IDF u,d Respectively represent the TF value and IDF value of the difficulty level d in the access history of user u; |{uc k :c k .d ’ =d}| represents the number of times user u browses cases with difficulty level d, c k represents the kth case visited by the user, d ’ Indicates case c k The corresponding difficulty level, |{vc k :c k .d ’ =d}| represents the number of times user v browses cases with difficulty level d, |R u | and |R v |represents the total number of case views of users u and v respectively, and U represents the set of all users.

6. A teaching case recommendation method according to claim 5, characterized in that: Based on the user-case access frequency matrix in the case recommendation model and the nearest neighbor of the target user, the predicted score P of the case by the target user u is calculated according to formula (6): u,i : Among them, u represents the target user, i represents the case to be predicted and scored, and N u represents the set of nearest neighbors of target user u, and n represents N u , sim(u,n) represents the similarity between target user u and user n, R n,i represents the number of visits of target user u to case i, represents the average number of visits to all cases by target user u, represents the average number of visits to all cases by user n, F u,i.d Represents the preference of user u for case difficulty level id.

7. A teaching case recommendation device, characterized in that: include: The acquisition and preprocessing module is used to acquire the user data of the target user and perform preprocessing to obtain the user feature vector; A medical record recommendation model, used for inputting the user feature vector, outputting a predicted score and a teaching case recommended according to the predicted score; Wherein, the medical record recommendation model includes: A user clustering module calculates the similarity between the target user and other users based on the user feature vector; A search module, used to find the nearest neighbor of the target user through the similarity; An evaluation module is used to evaluate the target users’ preferences for the difficulty level of the cases; The recommendation module is used to calculate the target user's predicted score for the recommended teaching case and rank the predicted score based on the user-case access frequency matrix in the case recommendation model, the target user's nearest neighbors, and the target user's preference for the case difficulty level, and recommend the teaching case to the target user according to the predicted score.

8. An electronic terminal, characterized in that: The method comprises a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are executed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, which implements the steps of the method according to any one of claims 1 to 6 when executed by a processor.

Images