An intelligent course selection method and system for research-based education

Abstract

The present invention relates to the technical field of course selection in an education system, and specifically to an intelligent course selection method and system for research-based education, which obtains the course duration of research-based courses obtained by users of a research-based education platform, obtains the course selection ratio of users of the research-based education platform, obtains a course selection recommendation value through a peak number of people and a course selection ratio, and matches research-based courses suitable for user time and courses according to the course selection recommendation value. For course projects that require research-based practice, the present invention can effectively improve the efficiency of school course selection, improve the efficiency of admission and management of research-based bases, improve the floor space efficiency of research-based bases, make full and reasonable use of the sites of research-based practice bases, and avoid the gathering of research-based practice activities.

Description

Technical Field

[0001] The present invention relates to the technical field of course selection in an education system, and in particular to an intelligent course selection method and system for research-based education. Background Art

[0002] Research and study is an educational method that emphasizes providing learning opportunities through field trips and practical activities. Research and study activities are designed to allow students to participate, observe and practice in person to deepen their understanding of a specific topic or field. This educational method is usually used in conjunction with classroom teaching to promote students' learning and development by bringing them to a real environment and allowing them to experience and explore in person. Therefore, many schools and education departments will arrange research and study activities specifically for students.

[0003] In the implementation of research and study practice, schools need to choose appropriate research and study courses according to teaching needs, because each research and study course must consider the type of course, the stage of study, the length of the course, the background, goals, characteristics, content, arrangement, learning tasks, emergency plans and security facilities, etc. It is also necessary to consider the distance between the location of the base of the outbound research and study and the school, and the travel distance. After selecting the base and the course, because the base’s venue and capacity are limited, the school’s itinerary cannot be aware of the registration situation of the base in real time, and a large number of school students will gather at the same base for research and study at the same time. At present, schools and research and study bases coordinate and communicate through communication tools, and there is no system within the research and study base that can effectively control the flow of people and schedule classes. All arrangements are made based on confirming the registration, and then registering for activities and experience.

[0004] The research and study practice base will accept individual and group research and study activity teaching, manage and estimate the number of people arriving each day, and more accurately calculate the venue utilization rate and admission capacity, which is a crucial task for the research and study practice base.

[0005] When schools organize research and study activities, they often do not just go to one research and study base at a time. They also have to coordinate the study schedules of multiple research and study bases, which has become a confusion for schools and research and study practice bases. Therefore, the communication efficiency is low and errors are prone to occur when selecting research and study courses and research and study bases in the existing way. The efficiency of the research and study practice bases is low, and the venues of the research and study practice bases cannot be fully and reasonably utilized, thus wasting resources. Summary of the invention

[0006] The purpose of the present invention is to propose an intelligent course selection method and system for research-based education to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0007] An intelligent course selection method for research and practice education, the method comprising the following steps:

[0008] S100: Obtain the duration of the research courses of the research education platform users;

[0009] S200: Obtain the course selection ratio of users on the research and education platform;

[0010] S300 obtains the recommended course selection value based on the peak number of students and the course selection ratio;

[0011] S400: Match the user's time with a study course that is suitable for the course based on the course selection recommendation value.

[0012] Further, in step S100, in the research and study education platform, the user's online course learning situation is obtained, and the user's online course learning situation includes the proportion of learning time and the proportion of learning progress in the course. The proportion of learning time and the proportion of learning progress are calculated to obtain the matching score of the research and study course, and the recommendation of the research and study course suitable for the user is obtained according to the matching score of the research and study course, and the recommendation is displayed in the front-end user terminal of the research and study education platform, and the user selects the research and study course independently, and the matching score of the research and study course selected by the user is recorded. If the user selects the research and study course through independent search, the matching score of the research and study course selected by the user is defined as 0;

[0013] Further, in step S200, the course selection ratio of users is counted through the research and study education platform, and the course selection ratio of users is matched with the research and study course schedule. The matching method includes:

[0014] S201: Obtain the peak number of people accepted by the entire venue of the study base. If the venue has reservations on the current date, and the study base has set a flow limit for the current date, the total number of people who can choose the study course on the current date is the flow limit set by the study base for the current date minus the number of people who have made reservations. The total number of people who can choose the study course on the current date is the peak number.

[0015] S202: Obtain a list of multiple course information selected for the research and study activity, obtain the start time and end time of each research and study course and the morning and afternoon, obtain a list of base traffic configuration dates, and find the base carrying capacity according to the start and end time of the base;

[0016] S203: Set the flow limit configuration and opening hours, and set the daily operating hours of the research and study base on daily opening days and statutory holidays;

[0017] The number of people allowed to enter is set to set the peak number of people allowed to enter the venue within the scope of the research and study base during the opening period;

[0018] Current limit configuration, set the number of people that the research and study base can accommodate within a specific date range. If the specific date range and the open day range overlap, the number of people that can be accommodated within the specific date range will be given the highest priority;

[0019] S204: Get the current limiting configuration for the day. If there is a current limiting configuration, deduct the time of the current limiting configuration from the opening time of the day; if the start time is the morning of the day and the end time is also the morning of the day, deduct the morning opening time; if the start time is the afternoon of the day and the end time is also the afternoon of the day, deduct the afternoon opening time; if the start time is the morning of the day and the end time is also the afternoon of the day, deduct the whole day opening time; when deducting, determine whether the remaining number of objects carried by the base is greater than the number of students. If it is greater than the number of students, after deduction, update and save the opening time of the day, otherwise do not update the opening time of the day and throw feedback;

[0020] Preferably, if the number of people accepted by the study base is deducted after the reservation, but the specified effective time of the activity is exceeded, the number of people that the base can accommodate will be automatically released. From the time of initiating the study activity, the travel plan for the study activity is not confirmed and submitted within 1 day, the study activity fails to pass the review and filing after initiation and is not resubmitted for more than 3 days, and one user actively cancels or terminates the activity after initiation;

[0021] In step S300, the course selection recommendation value of the intelligent course selection is obtained by obtaining the course selection ratio and the peak number of students. The method for calculating the course selection recommendation value includes the following steps:

[0022] S301: Integrate the matching degree of the corresponding user course selection ratio and the research course, define the peak number of people as R, take the peak number of people in each time period, and mark the subscript R in the order of time periods 1 , R 2 , R 3 ……R n , n is the number of the last time period recorded, and the peak number of people in each time period is used to construct a peak number set P, P = [R 1 , R 2 , R 3 ……R n ], obtain the matching degree between the user and the corresponding selected research course, and define the matching degree as Lidt, construct the weight of the corresponding time period and the user matching degree, and judge the peak number of people in adjacent time periods to schedule the weight Q1 in the peak number of people in different time periods, The R k and R k+1 are the kth and k+1th elements in the peak number set P;

[0023] S302: Construct a set D of matching degrees of research courses in each time period, D = [I 1 , I 2 , I 3 ……I n], get the subscripts of the same timestamps and time periods in sets P and D, and judge the matching weights Q2 of the research courses in different time periods by the matching degrees of the research courses in adjacent time periods. I k and I k+1 Construct the kth and k+1th elements in the set D for the matching degree of the research and study courses, assign weights according to the given Q1 and Q2, and calculate the influence weight W of the peak number of people and the matching degree of the research and study courses en , W ef ;

[0024]

[0025] Where m is the total number of time intervals in a total system operation cycle, and is also the total number of elements in set D or set P;

[0026] S303: The course selection recommendation value is obtained according to the calculation method:

[0027]

[0028] Where B is the recommended value for course selection, W en i is the peak impact weight of the number of people at time i, W ef j is the matching influence weight of the practical course at time j;

[0029] S304: The influence weight W of the matching degree between the peak number of students and the research course en , W ef Train a multi-layer neural network to combine the course recommendation value and W en , W ef , input into the neural network model for training, and output the input value X after the training is completed. The input value X is used to determine whether the recommended course selection value meets the qualified standard in the operation cycle. Each input value X needs to be trained by a separate neural network model, and the trained input data and the input value are in a mapping relationship, that is, the recommended course selection value and the input value X are in a mapping relationship;

[0030] Furthermore, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. The need to increase or decrease the number of intermediate layers or neurons is determined based on the training results. The calculation method for determining the number of intermediate layers and neurons is:

[0031] The middle layers in the neural network model are added layer by layer, and the input value X corresponding to the number of middle layers is named. When the number of middle layers is 1, the input value X is output. 1 , and so on, and according to the input value X, a nonlinear regression function is constructed, a is a constant value, and the variance D of the independent variable of the function is calculated. Where L is

[0032] `

[0033] The number of intermediate layers at output, X i The i-th input value X is output, and the X i is the input course recommendation value corresponding to the i-th input value X of the output. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance D. If D>A, the output input value X is input into the neural network model, and the number of intermediate layers is increased by 1. If D≤A, the input value X is output;

[0034] Update the weight coefficients of the neurons by gradient descent:

[0035]

[0036] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0037] Furthermore, in step S400, the calculated course selection recommendation value is matched with the corresponding time of the user to obtain a more reasonable research and study course selection. For course projects that require research and study practice, according to the operation of each link of the system, the efficiency of school course selection can be effectively improved, the efficiency of admission and management of the research and study base can be improved, the efficiency of the research and study base can be improved, the venue of the research and study practice base can be fully and reasonably utilized, and the gathering of research and study practice activities can be avoided.

[0038] A smart course selection system for research-based practical education. The system provides a client of the smart course selection system for research-based practical education through a third aspect, including a processor and a computer-readable storage medium. The computer-readable storage medium is the computer-readable storage medium as described above. The processor can execute the computer program in the computer-readable storage medium to implement the method of the smart course selection system for research-based practical education as described above.

[0039] The beneficial effects of the present invention are: an intelligent course selection method and system for research and study practice education, for course projects that require research and study practice, can effectively improve the efficiency of school course selection, improve the efficiency of admission and management of research and study bases, improve the efficiency of the floor space of research and study bases, make full and reasonable use of the venues of research and study practice bases, and avoid the gathering of research and study practice activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to further explain the present invention in detail, the present invention is provided with the following drawings and specific embodiments for explanation, and the drawings and embodiments do not limit the present invention. The drawings are as follows:

[0041] Figure 1 It is a flow chart of an intelligent course selection method for research and practice education;

[0042] Figure 2 It is a module diagram of an intelligent course selection system for research and practice education;

[0043] Figure 3 It is a course acceptance diagram of an intelligent course selection system for research and practice education. DETAILED DESCRIPTION

[0044] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0045] like Figure 1 and Figure 2 , an intelligent course selection method for research and practice education is shown, the method comprising the following steps:

[0046] S100: Obtain the duration of the research courses of the research education platform users;

[0047] S200: Obtain the course selection ratio of users on the research and education platform;

[0048] S300 obtains the recommended course selection value based on the peak number of students and the course selection ratio;

[0049] S400: Match the user's time with a study course that is suitable for the course based on the course selection recommendation value.

[0050] Furthermore, in step S100, in the research and study education platform, the user's online course learning situation is obtained, and the user's online course learning situation includes the proportion of learning time and the proportion of learning progress in the course. The proportion of learning time and the proportion of learning progress are calculated to obtain a matching score of the research and study course. According to the matching score of the research and study course, a recommendation of research and study courses suitable for the user is obtained, and the recommendation is displayed on the front-end user end of the research and study education platform. The user independently selects the research and study course, and the matching score of the research and study course independently selected by the user is recorded. If the user selects the research and study course through independent search, the matching score of the research and study course independently selected by the user is defined as 0.

[0051] Further, in step S200, the course selection ratio of users is counted through the research and study education platform, and the course selection ratio of users is matched with the research and study course schedule. The matching method includes:

[0052] S201: Obtain the peak number of people accepted by the entire venue of the study base. If the venue has reservations on the current date, and the study base has set a flow limit for the current date, the total number of people who can choose the study course on the current date is the flow limit set by the study base for the current date minus the number of people who have made reservations. The total number of people who can choose the study course on the current date is the peak number.

[0053] S202: Obtain a list of multiple course information selected for the research and study activity, obtain the start time and end time of each research and study course and the morning and afternoon, obtain a list of base traffic configuration dates, and find the base carrying capacity according to the start and end time of the base;

[0054] S203: Set the flow limit configuration and opening hours, and set the daily operating hours of the research and study base on daily opening days and statutory holidays;

[0055] The number of people allowed to enter is set to set the peak number of people allowed to enter the venue within the scope of the research and study base during the opening period;

[0056] Current limit configuration, set the number of people that the research and study base can accommodate within a specific date range. If the specific date range and the open day range overlap, the number of people that can be accommodated within the specific date range will be given the highest priority;

[0057] S204: Get the current limiting configuration for the day. If there is a current limiting configuration, deduct the time of the current limiting configuration from the opening time of the day; if the start time is the morning of the day and the end time is also the morning of the day, deduct the morning opening time; if the start time is the afternoon of the day and the end time is also the afternoon of the day, deduct the afternoon opening time; if the start time is the morning of the day and the end time is also the afternoon of the day, deduct the whole day opening time; when deducting, determine whether the remaining number of objects carried by the base is greater than the number of students. If it is greater than the number of students, after deduction, update and save the opening time of the day, otherwise do not update the opening time of the day and throw feedback;

[0058] Preferably, if the number of people accepted by the study base is deducted after the reservation, but the specified effective time of the activity is exceeded, the number of people that the base can accommodate will be automatically released. From the time of initiating the study activity, the travel plan for the study activity is not confirmed and submitted within 1 day, the study activity fails to pass the review and filing after initiation and is not resubmitted for more than 3 days, and one user actively cancels or terminates the activity after initiation;

[0059] In step S300, the course selection recommendation value of the intelligent course selection is obtained by obtaining the course selection ratio and the peak number of students. The method for calculating the course selection recommendation value includes the following steps:

[0060] S301: Integrate the matching degree of the corresponding user course selection ratio and the research course, define the peak number of people as R, take the peak number of people in each time period, and mark the subscript R in the order of time periods 1 , R 2 , R 3 ……R n , n is the number of the last time period recorded, and the peak number of people in each time period is used to construct a peak number set P, P = [R 1 , R 2 , R 3 ……R n ], obtain the matching degree between the user and the corresponding selected research course, and define the matching degree as Lidt, construct the weight of the corresponding time period and the user matching degree, and judge the peak number of people in adjacent time periods to schedule the weight Q1 in the peak number of people in different time periods, The R k and R k+1 are the kth and k+1th elements in the peak number set P;

[0061] S302: Construct a set D of matching degrees of research courses in each time period, D = [I 1 , I 2 , I 3 ……I n], get the subscripts of the same timestamps and time periods in sets P and D, and judge the matching weights Q2 of the research courses in different time periods by the matching degrees of the research courses in adjacent time periods. I k and I k+1 Construct the kth and k+1th elements in the set D for the matching degree of the research and study courses, assign weights according to the given Q1 and Q2, and calculate the influence weight W of the peak number of people and the matching degree of the research and study courses en , W ef ;

[0062]

[0063] Where m is the total number of time intervals in a total system operation cycle, and is also the total number of elements in set D or set P;

[0064] S303: The course selection recommendation value is obtained according to the calculation method:

[0065]

[0066] Where B is the recommended value for course selection, W en i is the peak impact weight of the number of people at time i, W ef j is the matching influence weight of the practical course at time j;

[0067] S304: The influence weight W of the matching degree between the peak number of students and the research course en , W ef Train a multi-layer neural network to combine the course recommendation value and W en , W ef , input into the neural network model for training, and output the input value X after the training is completed. The input value X is used to determine whether the recommended course selection value meets the qualified standard in the operation cycle. Each input value X needs to be trained by a separate neural network model, and the trained input data and the input value are in a mapping relationship, that is, the recommended course selection value and the input value X are in a mapping relationship;

[0068] Furthermore, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. The need to increase or decrease the number of intermediate layers or neurons is determined based on the training results. The calculation method for determining the number of intermediate layers and neurons is:

[0069] The middle layers in the neural network model are added layer by layer, and the input value X corresponding to the number of middle layers is named. When the number of middle layers is 1, the input value X is output. 1 , and so on, and according to the input value X, a nonlinear regression function is constructed, a is a constant value, and the variance D of the independent variable of the function is calculated. Where L is

[0070] `

[0071] The number of intermediate layers at output, X i The i-th input value X is output, and the X i is the input course recommendation value corresponding to the i-th input value X of the output. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance D. If D>A, the output input value X is input into the neural network model, and the number of intermediate layers is increased by 1. If D≤A, the input value X is output;

[0072] Update the weight coefficients of the neurons by gradient descent:

[0073]

[0074] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0075] Furthermore, in step S400, the calculated course selection recommendation value is matched with the corresponding time of the user to obtain a more reasonable research and study course selection. For course projects that require research and study practice, according to the operation of each link of the system, the efficiency of school course selection can be effectively improved, the efficiency of admission and management of the research and study base can be improved, the efficiency of the research and study base can be improved, the venue of the research and study practice base can be fully and reasonably utilized, and the gathering of research and study practice activities can be avoided.

[0076] like Figure 3 As shown, in the embodiment: a school or a research and study base is provided with a client, and a research and study course information database is pre-established in the computer-readable storage medium of the client, and the information database stores the accounts, research and study courses, class information, student information, etc. of each role.

[0077] School 1 developed R1 for 300 students in the third grade in the above system and chose the course of visiting two research and study bases, Lingnan Impression Garden and Guangdong Provincial Museum.

[0078] School 2 developed R2 for a total of 200 students in the fourth grade based on the above system and chose the course of visiting two research and study bases, Lingnan Impression Garden and Guangdong Provincial Museum.

[0079] If the research and study base A sets the venue capacity peak value of 1000 people J1 in the above system;

[0080] Research and study base B sets the venue acceptance peak value of 1000 people J2 in the above system;

[0081] The school teachers will use the research course 1 and research course 2 in the above two bases as the courses for this research activity, and select the above courses in the system when the research activity is initiated;

[0082] The client first obtains the study duration T1 and T2 of the two study courses, Study Course 1 and Study Course 2, from the study course information database: T1 = 1 day, T2 = 1 day, and then determines whether the study practice bases where the two study courses are located have a sufficient number of students. If the number of students is sufficient, the study practice base closest to the start time of the activity is given the highest priority. The travel order of the above two study courses is determined to be Study Course 1 to Study Course 2.

[0083] After the above sorting is completed, the study tour itinerary is determined, and then the number of people and date selected by the school are written into the admission data table of the study base on that date. The number of participants Rtotal is calculated as follows: Rtotal=R1+R2=300+200=500 people.

[0084] Then calculate the total number of people Y that will be accepted by Research and Study Base A on that date, Y=J1-Rtotal=1000-500=500.

[0085] The activity schedule in the above system is that Research and Study Course 1 starts in the morning and ends in the afternoon of September 15, 2023, and Research and Study Course 2 starts in the morning and ends in the afternoon of September 16, 2023; the system generates an activity from September 15 to 16, 2023 based on the above activity schedule, and completes the creation of the research and study practice activity.

Claims

1. An intelligent course selection method for research and practice education, characterized in that: The method comprises the following steps: S100: Obtain the duration of the research courses of the research education platform users; S200: Obtain the course selection ratio of users on the research and education platform; S300: Obtaining the course selection recommendation value based on the peak number of students and the course selection ratio; S400: Match the user's time with the appropriate research courses according to the course selection recommendation value; In step S100, the online course learning situation of the user is obtained in the re-study education platform, and the proportion of learning time and learning progress of the course type are calculated through an algorithm to obtain the matching score of the study practice course. According to the matching score of the study course, the recommendation of the study course suitable for the user is obtained, and the recommendation is displayed in the front-end user terminal of the study education platform. The user selects the study course independently, and the matching score of the study course selected by the user is recorded. If the user selects the study course through independent search, the matching score of the study course selected by the user is defined as 0; In step S200, the course selection ratio of users is counted through the research and study education platform, and the course selection ratio of users is matched with the research and study course schedule. The matching method includes: S201: Obtain the peak number of people accepted by the entire venue of the study base. If the venue has reservations on the current date, and the study base has set a flow limit for the current date, the total number of people who can choose the study course on the current date is the flow limit set by the study base for the current date minus the number of people who have made reservations. The total number of people who can choose the study course on the current date is the peak number. S202: Obtain a list of multiple course information selected for the research and study activity, obtain the start time and end time of each research and study course and the morning and afternoon, obtain a list of base traffic configuration dates, and find the peak number of base people according to the start and end time of the base; S203: Set the flow limit configuration and opening hours, and set the daily operating hours of the research and study base on daily opening days and statutory holidays; The number of people admitted is set to set the peak number of people that the research and study base can admit during the period when the venue is open; Current limit configuration, set the number of people that the research and study base can accommodate within a specific date range. If the specific date range and the open day range overlap, the number of people that can be accommodated within the specific date range will be given the highest priority; S204: Get the current limiting configuration for the day. If there is a current limiting configuration, deduct the time of the current limiting configuration from the opening time of the day; if the start time is the morning of the day and the end time is also the morning of the day, deduct the morning opening time; if the start time is the afternoon of the day and the end time is also the afternoon of the day, deduct the afternoon opening time; if the start time is the morning of the day and the end time is also the afternoon of the day, deduct the whole day opening time; when deducting, determine whether the remaining number of objects carried by the base is greater than the number of students. If it is greater than the number of students, after deduction, update and save the opening time of the day, otherwise do not update the opening time of the day and throw feedback; In step S300, the course selection recommendation value of the intelligent course selection is obtained by obtaining the course selection ratio and the peak number of students. The method for calculating the course selection recommendation value includes the following steps: S301: Integrate the matching degree of the corresponding user course selection ratio and the research course, define the peak number of people as R, take the peak number of people in each time period, and mark the subscripts in the order of time periods , , … , n is the number of the last time period recorded, and the peak number of people in each time period is used to construct the peak number set P, P=[ , , … ], obtain the matching degree between the user and the corresponding selected research course, and define the matching degree as Lidt, construct the weight of the corresponding time period and the user matching degree, and judge the peak number of people in adjacent time periods to schedule the weight Q1 in the peak number of people in different time periods, = , and are the kth and k+1th elements in the peak number set P; S302: Construct a set D of matching degrees of research courses in each time period, D=[ , , … ], get the subscripts of the same timestamps and time periods in sets P and D, and judge the matching weights Q2 of the research courses in different time periods by the matching degrees of the research courses in adjacent time periods. = , and Construct the kth and k+1th elements in set D for the matching degree of the research course, according to the given and Assign weights and calculate the impact weight of the peak number of people and the matching degree of the research and study courses , ; = ， = ； Where m is the total number of time intervals in a total system operation cycle, and is also the total number of elements in set D or set P; S303: The course selection recommendation value is obtained according to the calculation method: ； Where B is the recommended value for course selection. To record the peak impact weight of the number of people at time i, is the matching impact weight of the practical course at time j; S304: Influence weight based on the matching degree between the peak number of students and the research course , Train a multi-layer neural network to combine the course recommendation value and , , input into the neural network model for training, and output the input value X after the training is completed. The input value X is used to determine whether the recommended course selection value meets the qualified standard in the operation cycle. Each input value X needs to be trained by a separate neural network model, and the trained input data and the input value are in a mapping relationship, that is, the recommended course selection value and the input value X are in a mapping relationship; In the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. The training results are used to determine whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for determining the number of intermediate layers and neurons is as follows: The middle layers in the neural network model are added layer by layer, and the input value X corresponding to the number of middle layers is named. When the number of middle layers is 1, the input value is output. , and so on, and based on the input value X, construct a nonlinear regression function, f(X) = , a is a constant value, calculate the variance D of the function's independent variable, D= , where L is the number of intermediate layers at output, is the i-th input value X of the output, the The input course recommendation value corresponding to the i-th input value X of the output; Update the weight coefficients of neurons by gradient descent: （l）=- ； in, () is the partial derivative of the function, P and Q are the system output error and neuron weight increment, both are constant values. (l) is the neuron weight coefficient, (l) is the updated neuron weight coefficient, Represents the neuron learning rate, which is (l) Determine the number of neurons to be set in the intermediate layer.

2. An intelligent course selection system for research and practice education, characterized in that: The system provides a client of an intelligent course selection system for research-based practical education, including a processor and a computer-readable storage medium. The processor can execute a computer program in the computer-readable storage medium to implement a method of an intelligent course selection system for research-based practical education according to claim 1.

Images