College multi-dimensional automatic course arrangement method and system based on computational intelligence

Abstract

The invention discloses a college multi-dimensional automatic course arrangement method and system based on computational intelligence. The method comprises the steps of obtaining dynamic data and static data for automatic course arrangement; according to different curriculum type attributes in the curriculum schedule information, setting labels for the type attributes of different curriculums to endow the curriculum schedule information with fixed attributes; obtaining a constraint condition for constructing dynamic data and a constraint condition for constructing static data; and according to the constraint condition of the dynamic data and the constraint condition of the static data, establishing a calculation intelligent model and carrying out iterative calculation to obtain a final course arrangement scheme. According to the invention, by introducing an advanced computing intelligence technology, the efficiency, flexibility and intelligent level of course arrangement of colleges and universities are improved, the defects in a traditional course arrangement method are overcome, and a more efficient, intelligent and highly adaptive course arrangement solution is provided for the colleges and universities.

Description

Technical Field

[0001] The present invention belongs to the field of computational intelligence technology, and specifically relates to a multi-dimensional automatic course scheduling method and system for colleges and universities based on computational intelligence, which can solve the problem of multi-dimensional and multi-conflicting automatic course scheduling systems. Background Art

[0002] In the research field of college course scheduling system, there have been many research programs that have proposed different solutions to the course scheduling problem. These solutions can be roughly divided into the following categories: rule-based methods, optimization algorithm-based methods, and artificial intelligence-based methods.

[0003] First, rule-based methods were the first to appear. Such methods usually rely on a set of preset rules and constraints, such as teachers' teaching time, classroom capacity limits, students' course choices, etc. Rule-based scheduling systems generate schedules by setting priorities and constraints. For example, some systems set rules to ensure that each teacher's teaching time does not conflict with their personal schedules while maximizing classroom use. However, these rules may not be flexible enough to cope with complex and changing scheduling needs, and it is difficult to adapt to changing teaching environments and needs.

[0004] Secondly, methods based on optimization algorithms have been widely used in recent years. This type of method transforms the scheduling problem into an optimization problem and uses mathematical optimization techniques to find the optimal solution. These algorithms gradually improve the scheduling plan by simulating mechanisms such as natural selection and physical processes until an optimal solution that meets various constraints is found. For example, genetic algorithms can simulate the natural selection process and gradually screen out better solutions. However, these optimization algorithms usually require a long computing time, especially in large-scale and highly complex scheduling problems.

[0005] Third, artificial intelligence-based methods have also received increasing attention in recent years. Artificial intelligence technologies, such as machine learning and deep learning, have been applied to class scheduling systems to improve the level of intelligence in class scheduling. Machine learning algorithms can predict the needs of students and teachers and automatically generate class scheduling plans by analyzing historical class scheduling data. For example, deep learning models are able to handle complex nonlinear relationships and train large-scale data through neural networks to mine potential patterns. However, the implementation of artificial intelligence methods usually requires a large amount of labeled data and high computing resources, which places higher requirements on the accuracy and stability of the system.

[0006] In summary, existing research solutions have their own advantages and disadvantages when dealing with the multi-dimensional automatic course scheduling problem in colleges and universities. The rule-based method provides a simple solution, but lacks flexibility; the optimization algorithm-based method can handle complex course scheduling problems, but the calculation time is long; the artificial intelligence-based method has strong intelligent characteristics, but has high requirements for data and computing resources. Summary of the invention

[0007] In order to solve the above problems existing in the prior art, the present invention provides a multi-dimensional automatic course scheduling method and system for colleges and universities based on computational intelligence. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0008] One aspect of the present invention provides a multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence, comprising:

[0009] S1: Acquire dynamic data and static data for automatic class scheduling, wherein the static data includes teacher information, classroom information and class information, and the dynamic data includes course schedule information;

[0010] S2: according to different attributes of the course type in the course schedule information, setting annotations for different course type attributes to assign fixed attributes to the course schedule information;

[0011] S3: constructing the constraint conditions of the dynamic data and the constraint conditions of the static data;

[0012] S4: According to the constraints of the dynamic data and the constraints of the static data, a computational intelligence model is established and iterative calculations are performed to obtain a final class scheduling plan.

[0013] Another aspect of the present invention provides a multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence, the system comprising:

[0014] Data acquisition module, used to obtain dynamic data and static data required for automatic class scheduling and perform preprocessing;

[0015] A marking module, for setting markings for type attributes of different courses according to different course type attributes in the course schedule information so as to give fixed attributes to the course schedule information;

[0016] A function modeling module, used for constructing the constraint conditions of the dynamic data and the constraint conditions of the static data;

[0017] An evolutionary computing module, used to establish a computing intelligence model and perform iterative computing according to the constraints of the dynamic data and the constraints of the static data;

[0018] The termination module is used to determine whether the iteration termination condition has been reached and output the final course scheduling result.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. The multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence of the present invention can generate a course scheduling plan that meets multiple constraints in a shorter time by introducing advanced computational intelligence technology, significantly improving the efficiency, flexibility and intelligence level of course scheduling, solving the shortcomings of traditional course scheduling methods, and providing colleges and universities with a more efficient, intelligent and adaptable course scheduling solution. The present invention not only helps to optimize resource allocation and improve teaching quality, but also can provide strong decision-making support for education managers, thereby promoting the modernization of teaching management in colleges and universities.

[0021] 2. The present invention can automatically adjust the class scheduling plan through real-time data analysis and learning to cope with sudden changes, such as temporary leave of teachers or adjustment of course requirements. This high adaptability enables the system to maintain the optimal state of class scheduling under various changing circumstances, thereby improving the stability and reliability of class scheduling. The system of the present invention also has a high degree of scalability and compatibility. The modular design of computational intelligence technology enables the system to be easily integrated into the existing university management system and customized and expanded according to the specific needs of different universities. This flexible design not only reduces the complexity of system implementation, but also enhances the long-term adaptability of the system, enabling it to continue to evolve with the development of the educational environment and technology.

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of a multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence provided by an embodiment of the present invention;

[0024] Figure 2 It is a schematic diagram of visual representation of fitness value update provided by an embodiment of the present invention;

[0025] Figure 3 is a flow chart of an evolutionary computing algorithm provided by an embodiment of the present invention;

[0026] Figure 4 It is a structural diagram of a multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence provided by an embodiment of the present invention;

[0027] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the invention. DETAILED DESCRIPTION

[0028] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of a multi-dimensional automatic course scheduling method and system for colleges and universities based on computational intelligence proposed by the present invention, in combination with the accompanying drawings and specific implementation methods.

[0029] The above and other technical contents, features and effects of the present invention are clearly presented in the following detailed description of the specific implementation modes in conjunction with the accompanying drawings. Through the description of the specific implementation modes, the technical means and effects adopted by the present invention to achieve the predetermined purpose can be more deeply and specifically understood. However, the attached drawings are only for reference and explanation purposes and are not used to limit the technical solutions of the present invention.

[0030] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants are intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements that are not explicitly listed. In the absence of more restrictions, the elements defined by the statement "including one..." do not exclude the existence of other identical elements in the article or device including the elements.

[0031] Embodiment 1

[0032] This embodiment proposes a multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence, which is applied to the process of teaching plan arrangement in colleges and universities. It aims to use advanced computational intelligence technology to provide a fast and efficient optimal course scheduling solution for a large number of conflicts and constraints that arise in the course scheduling process. This automatic course scheduling method not only considers the limitations of traditional course scheduling methods, but also incorporates the latest progress in computational intelligence to achieve multi-dimensional optimization and intelligent decision-making. See Figure 1 , Figure 1 1 is a flowchart of a multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence provided by an embodiment of the present invention, the method comprising:

[0033] S1: Acquire dynamic data and static data for automatic class scheduling, the static data includes teacher information, classroom information and class information, and the dynamic data includes course schedule information.

[0034] In this step, it is necessary to obtain dynamic data and static data related to class scheduling as effectively as possible, upload dynamic data and static data through automatic import and manual editing, and perform data preprocessing. Data acquisition refers to an effective storage process, including collecting classroom information, teacher information, class information, and curriculum information, which can be achieved through automatic generation, manual entry, and self-editing. The purpose of data preprocessing is to store the entered or uploaded tabular data (including teacher information, classroom information, class information, and curriculum information) as class data of multiple dimensions in the form of a structure.

[0035] Data acquisition is the premise of the multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence. Data acquisition is to obtain all the elements related to course scheduling. Without basic data collection, other subsequent stages of work will not be possible.

[0036] Specifically, this embodiment is obtained by separating dynamic data from static data. According to the characteristics of the data, the data is divided into two categories: static data and dynamic data. Static data mainly includes teacher information, classroom information and class information. Such information has fixed attributes and does not change easily. Among them, teacher information specifically includes teacher name, title, position, teaching and research group, subjects taught, main activity campus, etc.; classroom information specifically includes the number of students that can be accommodated, whether it is a multimedia classroom, whether it is a laboratory, whether it supports group discussions, etc.; class information specifically includes the number of students in the class, grade, whether it is an experimental class, whether it is a bilingual teaching class, etc. Dynamic data mainly includes curriculum information, which specifically includes the class time of this class, classroom, teacher, course type attribute, whether experiments are required, whether it is required to go online, whether online teaching is required, etc. Further, the course type attributes of this embodiment include required courses, school elective courses, college elective courses, etc., and dynamic data will change with the course scheduling process until all dynamic information in the course plan is free of conflict or the maximum iteration round is reached.

[0037] S2: According to different attributes of the course type in the course schedule information, annotations are set for different course type attributes to assign fixed attributes to the course schedule information.

[0038] Step S2 of this embodiment specifically includes:

[0039] S2.1: Mark the course type attributes for each subject according to the course schedule information, and enter the scheduling constraints related to the course type attributes. The course type attributes include required courses, elective courses, basic courses, professional courses, freshman seminars or subject introduction courses.

[0040] The teaching plan is classified according to the different course type attributes in the course schedule information. The course type attributes can be divided into required courses, elective courses, basic courses, professional courses, etc. according to school differences and college differences, but are not limited to these. This annotation information is intended to give fixed attributes to the course schedule information. In other words, if the course type attribute of the current course is a required course, then the subsequent scheduling restrictions of the required course need to be met.

[0041] Furthermore, courses marked with course type attributes may have certain scheduling constraints, such as basic courses and professional courses must be scheduled in the morning and no more than 4 classes per day, and physical education classes and limited elective courses must be scheduled in the afternoon. Such scheduling constraints need to be edited during the data entry and classification stages. If omitted, they can be returned and modified in the subsequent process. The scheduling constraints can be described as:

[0042] maximize F(x)=(f1(x),...f m (x)) T ,

[0043] stx∈Ω,

[0044] Among them, f(x) is the function modeled by constraints, F(x) is the set of constraints, Ω is the decision space, and maximize means maximizing the set of constraints.

[0045] S2.2: If the course attribute type of the current course is not in the saved optional attribute list, add the corresponding course type attribute, and assign corresponding scheduling constraints to the newly added course type attribute.

[0046] Specifically, if the course attribute type of the required course is not in the list of optional attributes that have been stored, you can add a new course attribute type, such as freshman seminar, subject introduction course, etc.; accordingly, you need to assign corresponding scheduling constraints to the newly added course attribute types, such as the course requires a multimedia classroom, the course must start after the midterm of this semester, etc.

[0047] S2.3: Save the newly added course type attributes and the corresponding scheduling constraints to the optional attribute list.

[0048] Specifically, the newly added course type attributes and the corresponding scheduling constraints are saved in the system's optional attribute list for subsequent direct use.

[0049] S3: Construct constraints on dynamic data and constraints on static data.

[0050] In the actual course scheduling process, due to the involvement of a large number of teachers, classes, classrooms, courses and other factors, such static information will lead to a large number of conflicts in the course scheduling process. The global optimal solution may not be obtained with the help of traditional algorithms, so it is necessary to use evolutionary computing algorithms for large-scale optimization search. In the course scheduling process, several static data constraints must first be met, otherwise the course scheduling results will be invalid. The static data constraints specifically include: at the same time, a classroom can only offer one course; at the same time, a teacher can only teach one course; at the same time, a class can only participate in one course; the number of seats in the classroom is greater than or equal to the total number of students.

[0051] Furthermore, in the process of course arrangement, the constraints of dynamic data need to be met, otherwise the course arrangement results will be unreasonable, including: classrooms are divided into multimedia classrooms and ordinary classrooms, and the classroom configuration needs to meet the course requirements; basic courses, compulsory courses, etc. need to be scheduled in the morning; physical education classes, school-wide elective courses, etc. need to be scheduled in the afternoon; two identical courses cannot be taken on the same day.

[0052] In other embodiments, other constraints may be established according to actual conditions, which are not listed here one by one.

[0053] S4: Based on the constraints of dynamic data and static data, a computational intelligence model is established and iterative calculations are performed to obtain the final class scheduling plan.

[0054] Computational intelligence model is a general term for a variety of evolutionary computing algorithms. Evolutionary computing is an artificial intelligence model based on the natural selection process that simulates biological evolution. It is a method that can search for the optimal solution generation by generation in the simulated biological evolution process. With the help of coding rules, evolutionary computing abstracts the core elements of the problem into genes on chromosomes (i.e., individuals in the population), and through processes such as gene crossover and mutation, it iteratively selects excellent genes for reproduction and generates the next generation until the optimal solution or a satisfactory optimization solution is obtained.

[0055] The core task of evolutionary computing is to give the chromosome expression rules of the solution through the coding system. First, a certain number of populations need to be randomly initialized, and the population is composed of a certain number of individuals. Each individual is actually a chromosome, and the fitness value can be calculated through the rules. After the first generation of population is generated, according to the evolutionary principle of survival of the fittest, it evolves generation by generation to produce excellent offspring.

[0056] In each generation of evolution, individuals are selected according to their fitness values, and crossover and mutation are performed with the help of genetic operators of natural genetics to generate new populations. The best individuals in the last generation of populations are decoded and can be used as approximate optimal solutions to the problem. The exit condition is generally to reach the maximum number of iterations, such as 10,000, or the fitness value meets the requirements, such as reaching 0.99.

[0057] See also Figure 4 , step S4 of this embodiment specifically includes:

[0058] S4.1: Construct an initial population including multiple individuals, initialize the fitness value of each individual and set the maximum number of iterations of the algorithm. Each individual includes a random set of class scheduling plans, and a set of class scheduling plans includes class scheduling plans for multiple different courses. The class scheduling plans are represented by real number codes.

[0059] This step generates a population composed of multiple individuals, each of which represents a set of possible course scheduling schemes, and a set of possible course scheduling schemes includes a possible course arrangement for each course to be scheduled. Each initial population includes multiple initial individuals, and each initial individual includes a set of random course scheduling schemes. It should be noted that a set of course scheduling schemes includes the course schedule information of multiple courses to be scheduled. The representation of individuals can be represented by real number coding. For example, a specific arrangement in the random course scheduling scheme represented by an original individual in the original population is: Course 3 is taught in classroom 16 on Monday at the 34th period and by teacher 7, then the individual can be represented by real number coding as 003034016007. The initial individuals of the population are randomly generated to ensure the diversity of the population and the quality of the initial solution. The population initialization step of the present invention provides a wide exploration space for the subsequent search process through a diverse initial solution set, avoiding premature convergence to a local optimal solution.

[0060] Subsequently, the fitness value of each individual is initialized. In this embodiment, the fitness values ​​of the individuals in each initial population are initialized to the same value, for example, all are 0.

[0061] S4.2: A fitness function established based on the constraints of static data and the constraints of dynamic data.

[0062] In the second step of evolutionary computation, fitness evaluation plays a key role. The fitness of each individual is determined by a set fitness function. The present invention provides an adaptive fitness evaluation mechanism that can adjust the evaluation criteria according to the characteristics of the problem. The accuracy of the fitness function directly affects the convergence speed of the evolutionary computation algorithm and the quality of the solution. In this step, the present invention supports dynamic adjustment of the fitness function to cope with changes in the environment or the problem. In addition, the present invention accelerates the fitness evaluation process through parallel computing technology, thereby maintaining efficient operation in large-scale populations or high-complexity fitness functions. The evaluated fitness value provides a basis for subsequent selection operations, ensuring the priority delivery of high-quality solutions.

[0063] Specifically, according to the fitness function established based on the constraints of static data and dynamic data, individuals are selected according to their fitness in each generation of evolution.

[0064] Among them, the fitness function constructed according to the constraints of static data is:

[0065] For the class schedules of different courses included in the current individual in the population, if there is no overlap of multiple courses in the same classroom at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the same teacher does not teach in multiple classrooms at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the same class does not study in multiple courses at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the total number of students participating in the current course is less than or equal to the number of seats in the classroom, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged.

[0066] The fitness function constructed according to the constraints of dynamic data is: for the course schedules of different courses included in the current individual in the population, if the multimedia requirements of the current course match the classroom, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the course type matches the scheduled time period, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if no repeated courses are scheduled on the same day, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged. In this embodiment, the preset value can be 1. In other embodiments, different weights can also be set for the changes in the fitness value of each constraint condition.

[0067] S4.3: Update the fitness value of each individual in the population according to the constructed fitness function.

[0068] Specifically, the class schedule of each course contained in the current individual (each individual represents a chromosome expression) is traversed. If there is no overlap of multiple courses in the same classroom at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the same teacher does not teach in multiple classrooms at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the same class does not study in multiple courses at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the total number of students participating in the current course is less than or equal to the number of seats in the classroom, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged.

[0069] Traverse the schedule of each course contained in the current individual (each individual represents a chromosome expression). If the multimedia requirements of the current course match the classroom, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the course type matches the scheduled time period of noon or afternoon, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if there are no repeated courses scheduled on the same day, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged.

[0070] Similarly, according to the above processing, each individual in the population is traversed to obtain the updated fitness value of each individual.

[0071] Furthermore, during the fitness value update process, the change in fitness value can be represented by a visual rule representation, see Figure 2 , Figure 2 It is a visualization representation of fitness update provided by an embodiment of the present invention. As shown in the figure, for the current individual, if the classrooms of the courses overlap, it is indicated by red S, and if they do not overlap, it is indicated by green S. Whether there are enough seats in the classroom, if not, it is indicated by red G, otherwise it is indicated by green G. Whether there are overlapping teachers in the course class, overlap is indicated by red P, otherwise it is indicated by green P. Whether there are overlapping classes in the course class, overlap is indicated by red R, otherwise it is indicated by green R. If the course meets the multimedia requirements, it is indicated by red G if it does not meet the requirements, otherwise it is indicated by green G.

[0072] For the scenario of the embodiment of the present invention, the higher the fitness score, the better the solution provided. Its function can be expressed as:

[0073]

[0074] Among them, f i (x) represents the fitness value corresponding to each constraint condition, θ iIndicates the weight of each constraint condition. In this embodiment, the weight of each constraint condition is set to be the same. In other embodiments, the weight of each constraint condition can also be adjusted according to specific circumstances or cases.

[0075] S4.4: Based on the updated fitness value, selection and crossover operations are performed to generate new offspring individuals.

[0076] Based on the fitness value obtained in step S4.3, this step ensures that individuals with higher fitness have a greater probability of being selected, imitating the "survival of the fittest" principle in nature. In the selection operation step of this embodiment, based on the fitness evaluation result, that is, the updated fitness value of each individual, a roulette selection algorithm is used to select better individuals for reproduction. The selection operation simulates the "survival of the fittest" principle in natural selection, so that individuals with higher fitness have a greater probability of being selected to improve the quality of the solution. The present invention optimizes the traditional selection algorithm so that it can adaptively adjust the selection pressure to prevent the population diversity from being lost too quickly. In the roulette selection, the probability of an individual being selected is proportional to its fitness, thereby ensuring the transmission of high-quality individuals. In this embodiment, a roulette selection algorithm is used to select 50% of the individuals for subsequent reproduction. However, in order to avoid excessive concentration of selection leading to a decrease in population diversity, the present invention introduces an entropy-based control mechanism to ensure that the diversity of the population is maintained during the selection process, thereby increasing the breadth of the search space and improving the effect of global search.

[0077] Subsequently, the selected individuals are recombined through the crossover operation to generate new offspring individuals. The crossover operation simulates the exchange of genetic information to introduce new gene combinations and increase the diversity of solutions. Crossover methods include single-point crossover and multi-point crossover.

[0078] In the crossover operation step of the present embodiment, the selected individuals generate new offspring through the crossover process in the genetic operation. The purpose of the crossover operation is to recombine the genetic information of two or more individuals to simulate the recombination of genetic material in nature to produce new individuals with different characteristics. The present invention adopts a multi-point crossover strategy, which dynamically adjusts the crossover probability and adjusts the crossover strength of individual genes according to the fitness value to maximize the search efficiency. Multi-point crossover allows multiple segmentation points to generate more diverse offspring. The present invention proposes an adaptive crossover mechanism based on historical search information, which uses current population information to judge the effectiveness of the crossover strategy, thereby achieving a more effective exploration and development balance in the search process. It should be noted that performing a crossover operation does not change the number of individuals.

[0079] S4.5: Perform mutation operation on the new population consisting of new offspring individuals and individuals not selected in the selection operation.

[0080] In order to further improve the diversity of the population and prevent the algorithm from falling into the local optimal solution, the present invention introduces a random mutation process in the mutation operation step. The mutation operation generates a new solution space exploration direction by randomly modifying some genes of the individual. The present invention innovatively proposes an adaptive mutation rate adjustment method based on the evolutionary state, which increases the mutation rate when the population is close to convergence to enhance the search ability for unexplored areas; and reduces the mutation rate in the early exploration stage to maintain the accuracy of local search. Through this dynamic adjustment, the mutation operation can effectively balance the breadth and depth of the search and avoid premature convergence of the algorithm. Traditional mutation operations will cause the iteration time to be too long and fail to reach the global optimum. The method of the present invention greatly reduces the running time and improves the efficiency of the algorithm by improving the mutation operator.

[0081] Step S4.5 of this embodiment specifically includes:

[0082] S4.51: Set the mutation probability. In this embodiment, for the first 50% of individuals with better performance (i.e., higher fitness values) in the population, the mutation probability is set to 10%, and for the last 50% of individuals with poorer performance, the mutation probability is set to 50%.

[0083] S4.52: Randomly select mutant individuals from the new population composed of new offspring individuals and individuals not selected in the selection operation. That is to say, during the mutation process, the mutation probability of the first 50% individuals with better performance in the new population is 10%, and the mutation probability of the last 50% individuals with worse performance is 50%.

[0084] S4.53: Determine the mutation point of the mutant individual, which can be one of the class time, class classroom, and teacher in the dynamic data, and there is only one at most.

[0085] S4.54: All dynamic data in the mutation point are mutated sequentially according to the mutation probability to obtain the next generation population after mutation.

[0086] The population update step in the present invention forms the next generation population by adding new individuals generated by selection, crossover and mutation to the existing population. This step adopts a mixed replacement strategy to flexibly select the replacement ratio according to the individual fitness value and the population diversity status, thereby maximizing the optimization effect. The early termination strategy is also considered in the population update process, that is, when the fitness of individuals in the population changes little, the algorithm operation can be terminated in advance. In addition, the present invention combines an elite retention strategy to ensure that the best individuals in each generation directly enter the next generation population to ensure that the quality of the solution will not degrade during the update process. This step ensures the robustness and convergence speed of the algorithm.

[0087] S4.6: Obtain the fitness value of each individual in the current population. If the conditions for the end of iteration are met, the final class scheduling result is obtained.

[0088] The condition for the end of iteration, that is, in the termination condition setting of the present invention, the algorithm will terminate the iteration process according to the preset condition. The present invention proposes an adaptive termination judgment mechanism, which can dynamically adjust the number of iterations according to the population diversity and fitness progress, thereby avoiding unnecessary computational overhead. When the fitness of individuals in the population tends to be stable and the population diversity drops sharply, the algorithm will end early to improve computational efficiency. In addition, the present invention also supports the setting of a variety of termination condition combinations based on user needs, thereby providing a flexible optimization scheme. Through reasonable termination condition setting, the algorithm can improve the overall operating efficiency and avoid redundant calculations while ensuring the quality of the solution.

[0089] Specifically, when the current number of iterations reaches the set maximum number of iterations, the course schedule corresponding to the individual with the largest fitness value in the current population is used as the final class schedule; or when the fitness value of each individual in the current population has not changed before and after this iteration of the genetic algorithm, the course schedule corresponding to the individual with the largest fitness value in the current population is used as the final class schedule; if neither of the above two termination conditions is met, return to step S4.3 to continue iterative updating until one of the above conditions is met.

[0090] In summary, this embodiment improves the efficiency, flexibility and intelligence level of college class scheduling by introducing advanced computing intelligence technology. Its core goal is to solve the deficiencies in traditional class scheduling methods and provide colleges and universities with a more efficient, intelligent and adaptable class scheduling solution. At the same time, in view of the problem of automatic class scheduling of the present invention, evolutionary computing is optimized in a targeted manner. Through dynamically adjusted crossover operations, adaptive selection operations, and improved mutation operations, the efficiency of the algorithm is greatly increased. This not only helps to optimize resource allocation and improve teaching quality, but also can provide strong decision-making support for education managers, thereby promoting the modernization of college teaching management.

[0091] Embodiment 2

[0092] Based on the first embodiment, this embodiment provides a multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence. Figure 3 , Figure 3The present invention is a module diagram of a multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence provided by an embodiment of the present invention. The system is applied to the process of arranging teaching plans for colleges and universities. In view of a large number of conflicts and constraints that arise, computational intelligence methods are used to provide a fast and efficient optimal course scheduling solution. The automatic course scheduling system includes: a data acquisition module 201: used to obtain dynamic data and static data required for automatic course scheduling and perform preprocessing; a marking module 202: used to set markings for type attributes of different courses according to different course type attributes in the course schedule information so as to assign fixed attributes to the course schedule information; a function modeling module 203: used to construct constraints for dynamic data and constraints for static data; an evolutionary computing module 204: used to establish a computational intelligence model and perform iterative calculations according to the constraints for dynamic data and constraints for static data; a termination module 205: used to determine whether the iteration termination condition is reached and output the final course scheduling result.

[0093] Optionally, the data acquisition module 201 is specifically used for:

[0094] Data collection: Collect classroom information, teacher information, class information (static data) and class schedule (dynamic data) through automatic generation, manual entry, self-editing and other methods; Data preprocessing: Store the entered or uploaded tabular data (including teacher information, classroom information, class information and course schedule information) as multi-dimensional class data in the form of a structure.

[0095] The marking module 202 is specifically used to: mark the course type attribute for each subject according to the course schedule information, and enter the course scheduling constraints related to the course type attribute, the course type attribute includes required courses, elective courses, basic courses, professional courses, freshman seminars or subject introduction courses; if the course attribute type of the current course is not in the saved optional attribute list, add the corresponding course type attribute, and assign the corresponding course scheduling constraints to the newly added course type attribute; save the newly added course type attribute and the corresponding course scheduling constraints to the optional attribute list.

[0096] The function modeling module 203 is specifically used to construct constraints on dynamic data and constraints on static data.

[0097] The constraints of static data include: a classroom can only offer one course at the same time; a teacher can only teach one course at the same time; a class can only participate in one course at the same time; the number of seats in the classroom is greater than or equal to the total number of students.

[0098] Furthermore, the constraints that meet the dynamic data specifically include: classrooms are divided into multimedia classrooms and ordinary classrooms, and the classroom configuration needs to meet the course requirements; basic courses and compulsory courses need to be scheduled in the morning; physical education classes and school-wide elective courses need to be scheduled in the afternoon; two identical courses cannot be taken on the same day. ;

[0099] The evolutionary computing module 204 is specifically used for: constructing an initial population including multiple individuals, initializing the fitness value of each individual and setting the maximum number of iterations, each individual including a random set of class scheduling plans, a set of class scheduling plans including class scheduling plans for multiple different courses, and the class scheduling plans are represented by real number coding; establishing a fitness function according to the constraints of static data and the constraints of dynamic data; updating the fitness value of each individual in the population according to the constructed fitness function; performing selection operations and crossover operations based on the updated fitness values ​​to generate new offspring individuals; performing mutation operations on a new population consisting of new offspring individuals and individuals not selected in the selection operation to obtain a mutated next generation population.

[0100] Furthermore, the termination module 205 is used to terminate the iteration when the fitness of individuals in the population tends to be stable or reaches the maximum iteration round.

[0101] The multi-dimensional automatic class scheduling system for colleges and universities based on computational intelligence of the present invention can automatically adjust the class scheduling plan through real-time data analysis and learning to cope with sudden changes, such as temporary leave of teachers or adjustment of course requirements. This high adaptability enables the system to maintain the optimal state of class scheduling under various changing conditions, thereby improving the stability and reliability of class scheduling. The system of the present invention also has a high degree of scalability and compatibility. The modular design of computational intelligence technology enables the system to be easily integrated into the existing college management system and customized and expanded according to the specific needs of different colleges and universities. This flexible design not only reduces the complexity of system implementation, but also enhances the long-term adaptability of the system, enabling it to continue to evolve with the development of educational environment and technology.

[0102] Embodiment 3

[0103] The embodiment of the present invention further provides an electronic device, such as Figure 5As shown, it includes a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304, wherein the memory 303 is used to store computer programs; when the processor 301 is used to execute the program stored in the memory 303, the following steps are implemented: dynamic data and static data for automatic class scheduling are obtained, wherein the dynamic data includes teacher information, classroom information and class information, and the dynamic data includes course schedule information; according to different course type attributes in the course schedule information, annotations are set for type attributes of different courses to assign fixed attributes to the course schedule information; constraints on dynamic data and constraints on static data are constructed; according to the constraints on dynamic data and constraints on static data, a computational intelligence model is established and iterative calculations are performed to obtain a final course scheduling plan.

[0104] The processor 301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 301 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state.

[0105] The memory 303 may include one or more computer-readable storage media, which may be non-transitory. The memory 303 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 303 is at least used to store a computer program, wherein, after the computer program is loaded and executed by the processor 301, the relevant steps of the multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence disclosed in any of the aforementioned embodiments can be implemented. In addition, the resources stored in the memory 303 may also include an operating system and data, etc., and the storage method may be short-term storage or permanent storage. Among them, the operating system may include Windows, Unix, Linux, etc. The data may include, but is not limited to, data involved in the multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence, etc.

[0106] In the several embodiments provided by the present invention, it should be understood that the apparatus and method disclosed by the present invention can be implemented in other ways. For example, the apparatus embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0107] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0108] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence, characterized in that: include: S1: Acquire dynamic data and static data for automatic class scheduling, wherein the static data includes teacher information, classroom information and class information, and the dynamic data includes course schedule information; S2: according to different attributes of the course type in the course schedule information, setting annotations for different course type attributes to assign fixed attributes to the course schedule information; S3: constructing the constraint conditions of the dynamic data and the constraint conditions of the static data; S4: According to the constraints of the dynamic data and the constraints of the static data, a computational intelligence model is established and iterative calculations are performed to obtain a final class scheduling plan.

2. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 1 is characterized in that: The S1 includes: S1.1: Upload teacher information, classroom information, class information and course schedule information for automatic class scheduling through automatic import and manual editing, where the teacher information includes the teacher's name, title, position, teaching and research group, subjects taught and main activity campus; the classroom information includes the number of students that can be accommodated, whether it is a multimedia classroom, whether it is a laboratory and whether it supports group discussions; the class information includes the number of students in the class, grade, whether it is an experimental class and whether it is a bilingual teaching class; the course schedule information includes the class time of the current course, the classroom, the teacher, the course type attributes, whether experiments are required, whether computer use is required and whether online teaching is required; S1.2: The teacher information, the classroom information, the class information and the course schedule information are stored as class data of multiple dimensions in the form of a structure.

3. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 1 is characterized in that: The S2 includes: S2.1: marking a course type attribute for each subject according to the course schedule information, and entering the course scheduling constraints related to the course type attribute, wherein the course type attribute includes a required course, an elective course, a basic course, a professional course, a freshman seminar, or a subject introduction course; S2.2: If the course attribute type of the current course is not in the saved optional attribute list, add the corresponding course type attribute, and assign corresponding scheduling constraints to the newly added course type attribute; S2.3: Save the newly added course type attribute and the corresponding course scheduling constraint conditions to the optional attribute list.

4. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 1 is characterized in that: The constraints of the static data include: at the same time, a classroom can only offer one course; at the same time, a teacher can only teach one course; at the same time, a class can only participate in one course; the number of seats in the classroom is greater than or equal to the total number of students; The constraints of the dynamic data include: Classrooms are divided into multimedia classrooms and ordinary classrooms. The classroom configuration needs to meet the course requirements; basic courses and compulsory courses need to be scheduled in the morning; physical education classes and school-wide elective courses need to be scheduled in the afternoon; two identical courses cannot be scheduled on the same day.

5. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 1 is characterized in that: The S4 includes: S4.1: construct an initial population including multiple individuals, initialize the fitness value of each individual and set the maximum number of iterations of the algorithm, each individual includes a random set of class scheduling plans, a set of class scheduling plans includes class scheduling plans for multiple different courses, and the class scheduling plans are represented by real number coding; S4.2: Establishing a fitness function according to the constraint conditions of the static data and the constraint conditions of the dynamic data; S4.3: Update the fitness value of each individual in the population according to the constructed fitness function; S4.4: Based on the updated fitness value, selection and crossover operations are performed to generate new offspring individuals; S4.5: performing a mutation operation on the new population consisting of the new offspring individuals and the individuals not selected in the selection operation to obtain a mutated next generation population; S4.6: Obtain the fitness value of each individual in the current population. If the conditions for the end of iteration are met, the final class scheduling result is obtained.

6. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 5 is characterized in that: The S4.2 includes: A fitness function is constructed according to the constraint conditions of the static data: for the course scheduling plans of different courses included in the current individual in the population, if there is no overlap of multiple courses in the same classroom at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the same teacher does not teach in multiple classrooms at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the same class does not study in multiple courses at the same time, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the total number of students participating in the current course is less than or equal to the number of seats in the classroom, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; A fitness function is constructed according to the constraints of the dynamic data: for the class schedules of different courses included in the current individual in the population, if the multimedia requirements of the current course match those of the classroom, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if the course type matches the scheduled time period, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged; if no repeated courses are scheduled on the same day, the fitness score of the current individual increases by a preset value, otherwise the fitness value remains unchanged.

7. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 6 is characterized in that: The S4.3 includes: Traverse the schedule of each course contained in the current individual. If there is no overlap of multiple courses in the same classroom at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the same teacher does not teach in multiple classrooms at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the same class does not study in multiple courses at the same time, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the total number of students participating in the current course is less than or equal to the number of seats in the classroom, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the multimedia requirements of the current course match the classroom, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if the course type matches the scheduled time period, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; if there are no repeated courses scheduled on the same day, the fitness score of the current individual is increased by 1, otherwise the fitness value remains unchanged; According to the fitness function, each individual in the population is traversed to obtain an updated fitness value of each individual.

8. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 6 is characterized in that: The S4.5 includes: S4.51: setting a mutation probability for a new population consisting of the new offspring individuals and the individuals not selected in the selection operation; S4.52: randomly selecting mutant individuals in the new population; S4.53: Determine a mutation point of the mutant individual, where the mutation point is one of the dynamic data; S4.54: The dynamic data in the mutation points are mutated sequentially according to the mutation probability.

9. A multi-dimensional automatic course scheduling system for colleges and universities based on computational intelligence, characterized in that: The system is used to implement the multi-dimensional automatic course scheduling method for colleges and universities according to any one of claims 1 to 8, and comprises: Data acquisition module, used to obtain dynamic data and static data required for automatic class scheduling and perform preprocessing; A marking module, used for setting marks for different course type attributes according to different course type attributes in the course schedule information so as to give fixed attributes to the course schedule information; A function modeling module, used for constructing the constraint conditions of the dynamic data and the constraint conditions of the static data; An evolutionary computing module, used to establish a computing intelligence model and perform iterative computing according to the constraints of the dynamic data and the constraints of the static data; The termination module is used to determine whether the iteration termination condition has been reached and output the final course scheduling result.

10. The multi-dimensional automatic course scheduling method for colleges and universities based on computational intelligence according to claim 9 is characterized in that: The evolutionary computing module is specifically used for: Constructing an initial population including a plurality of individuals, initializing the fitness value of each individual and setting a maximum number of iterations, wherein the class scheduling scheme is represented by real number coding; Establishing a fitness function according to the constraint conditions of the static data and the constraint conditions of the dynamic data; According to the constructed fitness function, the fitness value of each individual in the population is updated; Based on the updated fitness value, selection and crossover operations are performed to generate new offspring individuals; A mutation operation is performed on a new population consisting of the new offspring individuals and the individuals not selected in the selection operation to obtain a mutated next generation population.

Images