Educational administration task implementation method and system based on automatic class splitting and combining rules

Abstract

The invention relates to an educational administration task implementation method and system based on an automatic class splitting and combining rule. The method comprises the steps of obtaining teacher course qualification data; dynamically adjusting the upper limit of the class number of teachers based on the teacher teaching quality evaluation data, and generating a teacher configuration data set; obtaining course teaching demand data, dynamically adjusting the capacity to be a preset multiple integer according to the capacity expansion mark, and generating a parameter configuration data set; constructing an automatic class splitting and combining rule set containing campus isolation, professional consistency and people number balanced distribution constraints; acquiring student class data, and performing clustering analysis on the student class data by applying a rule set to generate a class grouping set; conflict detection is carried out on the grouping set, and grouping conflict identifiers are generated according to the upper limit of the class number of teachers and historical teaching data; and optimizing the grouping set based on the identification, distributing teacher tasks and generating a task implementation scheme. According to the method, automatic class splitting and combining are achieved through parameterization rules and intelligent clustering, and the task allocation efficiency and the course arrangement success rate are improved.

Description

A method and system for implementing academic affairs tasks based on automatic class splitting and merging rules Technical Field

[0001] This application relates to the field of educational informatization technology, and in particular to a method and system for implementing teaching affairs tasks based on automatic class splitting and merging rules. Background Technology

[0002] With the continuous expansion of higher education and the increasing demands for refined teaching management, university academic affairs management faces unprecedented complexity and heavy pressure. The implementation of teaching tasks, especially the division of course classes (i.e., splitting and merging classes) and the allocation of instructors at the beginning of each semester for each major and class, is a core aspect of the academic affairs process. Traditional methods heavily rely on the personal experience and manual operation of academic affairs management personnel, requiring comprehensive consideration of multiple constraints such as course attributes, student majors and classes, teacher resources, classroom capacity, and campus distribution. Solutions are developed through repeated trial and error and coordination. This process not only consumes a significant amount of manpower and time, but also struggles to guarantee the overall rationality and optimality of the solution when dealing with complex scenarios involving large scale, multiple campuses, and multiple majors, and is unable to quickly respond to temporary adjustment needs.

[0003] Currently, while common academic affairs management systems have achieved partial digital management of information, their level of intelligence remains insufficient in the crucial decision-making process of assigning teaching tasks. These systems often only provide basic data recording and simple conflict checking functions; the core logic of class splitting and merging and teacher allocation strategies still require manual intervention. However, manual operation is inefficient and prone to errors, making it difficult to generate high-quality, feasible allocation plans that consider multiple optimization goals in a short period. Furthermore, the lack of effective mechanisms for anticipating and avoiding potential conflicts that may arise during the scheduling phase often leads to obstacles in subsequent scheduling stages, necessitating repeated backtracking and modifications, severely impacting the efficiency and stability of overall academic affairs management. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a method and system for implementing academic affairs tasks based on automatic class splitting and merging rules.

[0005] Firstly, this application provides a method for implementing academic affairs tasks based on automatic class splitting and merging rules, employing the following technical solution:

[0006] A method for implementing academic affairs tasks based on automatic class splitting and merging rules, the method comprising:

[0007] Obtain teacher qualification data, including the teachers for each course and the number of classes each teacher can teach;

[0008] Based on the input teacher teaching quality evaluation data, the upper limit of the number of classes that each course instructor can teach is dynamically adjusted to generate a teacher configuration dataset.

[0009] Obtain course teaching demand data for a specified academic year and semester, including class capacity, number of classes, expansion flag, and campus distribution information. Based on the expansion flag, dynamically adjust the class capacity to an integer multiple of a preset number, and generate a parameter configuration dataset.

[0010] Based on the teacher configuration dataset and parameter configuration dataset, an automatic class splitting and merging rule set is constructed, which includes campus isolation constraint rules, professional consistency constraint rules, and student equalization allocation constraint rules.

[0011] Obtain student class data, apply the automatic class splitting and merging rule set to perform cluster analysis on the student class data, and generate a class grouping set;

[0012] Perform group conflict detection on the class group set, and generate group conflict identifiers based on the maximum number of classes that a teacher can teach and the teacher's teaching history data in the teacher configuration dataset;

[0013] The class group set is optimized based on the group conflict identifier, teacher tasks are assigned, and task implementation plans are generated.

[0014] By adopting the above technical solution, a flexible and standardized input benchmark is first constructed using dynamic teacher allocation and standardized course parameters, thereby abstracting the complex academic management logic into a computable set of constraint rules. Then, a clustering analysis algorithm is used to automatically generate an initial student class assignment plan under multiple constraints. Furthermore, a conflict detection and optimization mechanism ensures the rationality of the plan at more granular levels, such as teacher workload and teaching experience. Ultimately, a high-quality, balanced task implementation plan that conforms to multiple constraints is generated, freeing academic staff from extremely tedious manual verification and arrangement work. This not only improves work efficiency but also ensures the scientific and balanced nature of class assignment through intelligent algorithms, laying a solid foundation for subsequent automatic course scheduling. This systematically improves the intelligence level and overall efficiency of academic management.

[0015] Secondly, this application provides a system for assigning academic affairs tasks based on automatic class splitting and merging rules, employing the following technical solution:

[0016] A system for assigning academic tasks based on automatic class splitting and merging rules, the system comprising:

[0017] The course qualification data acquisition module is used to acquire teacher qualification data, including the teachers for each course and the number of classes that each teacher can teach;

[0018] The teacher data configuration module is used to dynamically adjust the maximum number of classes that each course instructor can teach based on the input teacher teaching quality evaluation data, and generate a teacher configuration dataset.

[0019] The teaching parameter configuration module is used to obtain course teaching demand data for a specified academic year and semester, which includes teaching class capacity, number of teaching classes, expansion flag and campus distribution information. Based on the expansion flag, the teaching class capacity is dynamically adjusted to an integer multiple of a preset number, and a parameter configuration dataset is generated.

[0020] The automatic class splitting and merging rule construction module is used to construct an automatic class splitting and merging rule set based on the teacher configuration dataset and parameter configuration dataset, including campus isolation constraint rules, professional consistency constraint rules and student equalization allocation constraint rules;

[0021] The clustering analysis module is used to obtain student class data, apply the automatic class splitting and merging rule set to perform clustering analysis on the student class data, and generate a class grouping set;

[0022] The group conflict detection module is used to perform group conflict detection on the class group set and generate group conflict identifiers based on the maximum number of classes that a teacher can teach and the teacher's teaching history data in the teacher configuration dataset.

[0023] The task implementation module is used to optimize the class group set based on the group conflict identifier, assign tasks to teachers, and generate task implementation plans.

[0024] In summary, this application includes at least one of the following beneficial technical effects: by dynamically adjusting teacher resource allocation and standardized course parameters, the complex logic of class assignment in academic affairs is transformed into computable automated rules. By utilizing cluster analysis and conflict detection optimization mechanisms, scientific and balanced class grouping and teacher task schemes are generated efficiently, thereby completely liberating academic affairs staff from tedious manual verification and arrangement work, improving work efficiency and class assignment quality, and laying a solid foundation for subsequent automatic course scheduling. This systematically promotes the intelligent upgrading and overall efficiency optimization of academic affairs management. Attached Figure Description

[0025] Figure 1 is a schematic diagram of the first process of a method for implementing academic affairs tasks according to one embodiment of this application.

[0026] Figure 2 is a schematic diagram of the second process of the method for implementing teaching tasks according to one embodiment of this application.

[0027] Figure 3 is a schematic diagram of the third process of the method for implementing teaching affairs tasks according to one embodiment of this application.

[0028] Figure 4 is a schematic diagram of the fourth process of the method for implementing teaching affairs tasks according to one embodiment of this application.

[0029] Figure 5 is a schematic diagram of the fifth process of the teaching affairs task implementation method according to one embodiment of this application.

[0030] Figure 6 is a schematic diagram of the sixth process of the method for implementing teaching affairs tasks according to one embodiment of this application.

[0031] Figure 7 is a schematic diagram of the seventh process of the method for implementing academic affairs tasks according to one embodiment of this application.

[0032] Figure 8 is a schematic diagram of the eighth process of the method for implementing teaching tasks according to one embodiment of this application.

[0033] Figure 9 is a schematic diagram of the ninth process of the method for implementing teaching tasks according to one embodiment of this application. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to Figures 1-9 and embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0035] This application discloses a method for implementing academic affairs tasks based on automatic class splitting and merging rules.

[0036] Referring to Figure 1, a method for implementing academic affairs tasks based on automatic class splitting and merging rules specifically includes:

[0037] Step S101: Obtain teacher qualification data, including the teachers of each course and the number of classes each teacher can teach;

[0038] Among them, teachers are the core resource for the execution of teaching tasks. Their teaching qualifications (i.e., which courses they can teach) and the number of classes they can lead (i.e., the upper limit of teaching tasks that each teacher can undertake in the same period) are the basic constraints for the automated allocation in the embodiments of this application.

[0039] Step S102: Based on the input teacher teaching quality evaluation data, dynamically adjust the upper limit of the number of classes that each course instructor can teach, and generate a teacher configuration dataset.

[0040] Specifically, when the teaching quality evaluation result is higher than the preset threshold, the upper limit of the number of classes that teacher can teach will be increased; when the teaching quality evaluation result is lower than the preset threshold, the upper limit of the number of classes that teacher can teach will be decreased.

[0041] Specifically, by introducing external feedback signals such as "teacher teaching quality evaluation data," the system can optimize the static resource pool in the first step in real time. For example, a teacher who consistently receives excellent evaluations indicates good teaching effectiveness and strong workload capacity, and the system can automatically increase the upper limit of the number of classes they can teach, thereby making fuller use of high-quality teacher resources; conversely, for teachers with poor evaluations, the upper limit is appropriately lowered, which is both a guarantee of teaching quality and a protection mechanism for the teacher.

[0042] Ultimately, the generated teacher allocation dataset is a more rational resource model that has undergone preliminary intelligent processing, enabling the system to evolve from simple "allocation by qualification" to "allocation by effectiveness".

[0043] Step S103: Obtain course teaching demand data for the specified academic year and semester, which includes teaching class capacity, number of teaching classes, expansion flag and campus distribution information. Based on the expansion flag, dynamically adjust the teaching class capacity to an integer multiple of the preset number and generate a parameter configuration dataset.

[0044] The teaching requirements of the courses (such as the planned number of students in the teaching class, the expected number of teaching classes, and whether expansion is allowed) define the specific tasks that need to be assigned, while the campus distribution information defines the geographical distribution of different teaching resources (such as classrooms and laboratories) and student groups.

[0045] In some embodiments, when the expansion flag is set to allow expansion, the teaching class capacity value is rounded to a multiple of 5 or 10.

[0046] Understandably, normalizing the capacity to multiples of 5 or 10 can greatly simplify subsequent scheduling algorithms. This is because scheduling needs to consider factors such as classroom capacity (usually a fixed number of seats) and time blocks; the normalized capacity value can better match these resources, reducing resource fragmentation and thus fundamentally improving the success rate and efficiency of subsequent scheduling algorithms.

[0047] Step S104: Based on the teacher configuration dataset and parameter configuration dataset, construct an automatic class splitting and merging rule set, which includes campus isolation constraint rules, professional consistency constraint rules, and student equalization allocation constraint rules;

[0048] This involves abstracting complex management experience and constraints into computable and executable rule algorithms.

[0049] Specifically, the campus isolation constraint rule prohibits combining classes across campuses. By identifying the campus identifier in the student class data, it prevents classes from different campuses from being merged into the same teaching class. Since it is usually impractical for students from different campuses to travel to another campus for class at the same time, this constraint is a hard rule to ensure that all grouping operations are physically feasible.

[0050] The professional consistency constraint rule refers to grouping classes of the same course according to their professional identifier, generating a professional matching queue, and prioritizing the allocation of classes with the same professional major to the same teaching class. Grouping students with the same professional major into the same class helps maintain consistency in teaching progress and depth, simplifies teaching management, and facilitates the scheduling of subsequent professional courses.

[0051] The student size distribution constraint calculates the number of students in each class within each major matching queue. If the difference between the number of students in a class and the average exceeds a preset threshold, that class is merged into the major matching queue with the closest student size. This avoids situations where some classes are overcrowded and under-stressed, while others are undercrowded and underutilized. This ensures that the utilization rate of teaching resources is reasonable and that the teaching quality of each class remains within a controllable range.

[0052] These three constraints together constitute a multi-objective optimization rule set, which guides the system on how to achieve management optimization (professional constraints) and resource utilization optimization (population balance) as much as possible while satisfying basic feasibility (campus constraints).

[0053] Step S105: Obtain student class data, apply the automatic class splitting and merging rule set to perform cluster analysis on the student class data, and generate class grouping set;

[0054] The system obtains student class data containing attributes such as the student's campus, major, and number of students in the class. Instead of grouping the data randomly or in a simple order, it is driven by a set of rules.

[0055] Specifically, the processing flow is typically layered: First, a "campus isolation constraint" is applied to divide all students into several independent subsets based on their campuses—this is a coarse-grained division. Then, within each campus, a "major consistency constraint" is applied to cluster students by major, forming initial groups based on major. Finally, within major groups or across similar majors, a "balanced distribution constraint" is applied. Algorithms (such as calculating standard deviation or heuristic search) are used to fine-tune the initial groups, ensuring that the final "class group set" reaches a relatively optimal equilibrium state while satisfying all constraints.

[0056] Step S106: Perform group conflict detection on the class group set. Generate group conflict identifiers based on the maximum number of classes a teacher can teach and the teacher's teaching history data in the teacher configuration dataset.

[0057] Among them, group conflict detection includes detection based on the upper limit of the number of classes that can be taught and detection based on the teacher's teaching history data.

[0058] In one embodiment of this application, the detection based on the maximum number of classes a teacher can teach is a resource overload detection method, checking whether any teacher has been assigned more classes than their dynamic limit. The detection based on teachers' teaching history data is an experience-based optimization detection method; that is, combining teachers' teaching history data to check whether the same teacher is continuously assigned to the same subject's teaching classes, which may lead to teachers' knowledge structure becoming rigid or students experiencing evaluation fatigue. Ultimately, the generated group conflict identifier accurately points out the risk points in the grouping scheme, providing clear targets for further optimization.

[0059] Step S107: Optimize the class group set based on the group conflict identifier, assign teacher tasks, and generate task implementation plan.

[0060] In this process, the grouping scheme is iteratively optimized based on the grouping conflict identifiers output in the previous step. For example, for teacher workload conflicts, the system will reallocate suitable teachers with available workload to overloaded classes; for predicted major scheduling conflicts, it may fine-tune class combinations or assign tasks to another teacher with more flexible teaching arrangements, provided that the "major consistency constraint" is met. This process is an automated, rule-based fine-tuning loop until a stable grouping scheme that satisfies all hard constraints and soft optimization objectives without significant conflicts is obtained, ultimately forming an executable task implementation plan.

[0061] In the above implementation, a flexible and standardized input benchmark is first constructed using dynamic teacher allocation and standardized course parameters, thereby abstracting the complex academic management logic into a computable set of constraint rules. Then, a clustering analysis algorithm is used to automatically generate an initial student class assignment plan under multiple constraints. Furthermore, a conflict detection and optimization mechanism ensures the rationality of the plan at more granular levels, such as teacher workload and teaching experience. Ultimately, a high-quality, balanced task implementation plan that conforms to multiple constraints is generated, freeing academic staff from extremely tedious manual verification and arrangement work. This not only improves work efficiency but also ensures the scientific and balanced nature of class assignment through intelligent algorithms, laying a solid foundation for subsequent automatic course scheduling. This systematically improves the intelligence level and overall efficiency of academic management.

[0062] Referring to Figure 2, as one implementation of step S104, the step of constructing an automatic class splitting and merging rule set based on the teacher configuration dataset and the parameter configuration dataset includes:

[0063] Step S201: Parse the campus distribution information in the parameter configuration dataset and generate campus isolation constraint rules;

[0064] Among them, the campus distribution information defines the geographical distribution of different teaching resources (such as classrooms and laboratories) and student groups. The system analyzes this information and does not simply store the data, but extracts a fundamental business rule, namely the campus isolation constraint rule.

[0065] Specifically, the rule explicitly prohibits merging classes across campuses during the class grouping process. Due to the physical distance between different campuses, forcibly merging classes would lead to students frequently traveling between campuses for classes, which is impractical in terms of time management, transportation costs, and administrative complexity. It could even fundamentally conflict with campus-based resource allocation (such as binding classrooms to specific campuses). Therefore, this rule is the premise and foundation for all subsequent optimization operations, ensuring that any generated grouping scheme is built on a physically feasible foundation.

[0066] Step S202: Extract the preset professional identifier mapping relationship and generate professional consistency constraint rules;

[0067] The professional identifier mapping relationship defines the professional field to which each student's class belongs. The system extracts this relationship not for simple classification, but to generate professional consistency constraint rules. The core objective of these rules is to prioritize assigning classes with the same professional identifier to the same teaching class.

[0068] Understandably, firstly, students in the same major share highly similar curriculum systems, knowledge backgrounds, and teaching progress requirements. Grouping classes ensures the relevance and depth of the teaching content, improving teaching effectiveness. Secondly, from a management perspective, this greatly simplifies subsequent scheduling, examination administration, and grade management, as classes within the same major can be managed as a whole. This rule is a soft constraint geared towards optimization; it instructs the system, after meeting the hard condition of campus isolation, to prioritize professional attributes as the primary basis for clustering, thereby promoting more organized and efficient teaching from the outset.

[0069] Step S203: Based on the upper limit of the number of classes that a teacher can teach in the teacher configuration dataset and the teaching class capacity in the parameter configuration dataset, calculate the average class load coefficient per person and generate a constraint rule for balanced allocation of students.

[0070] The underlying principle of this step is to introduce a dynamic equilibrium and adaptive adjustment mechanism to optimize resource utilization. By linking the upper limit of the number of classes a teacher can teach with the class capacity, and calculating the average class load coefficient per student (which can be understood as a quantitative indicator that distributes the teacher's teaching workload to each student's seat), a reasonable preset capacity threshold is dynamically derived.

[0071] For example, the average class load coefficient can be used as a benchmark, combined with a configurable deviation rate (Δ), to determine a dynamically adjustable preset capacity threshold range (e.g., [benchmark - Δ, benchmark + Δ]). The resulting student allocation constraint rules (e.g., ensuring that the actual number of students in each class falls within the preset capacity threshold range) can prevent excessively large class sizes, which would lead to excessive teacher workload, decreased teaching quality, or classroom resource shortages; and can also prevent under-enrollment, which would result in the idle waste of teaching resources (teacher hours, classroom seats).

[0072] In some embodiments, calculating the average class load coefficient includes: obtaining the upper limit N of the number of classes that the current course instructor can teach and the teaching class capacity C; calculating the average class load coefficient α=N / C; generating a dynamic adjustment threshold range [α-Δ, α+Δ] based on α within a preset interval range, where Δ is the system configuration deviation rate; and the constraint rule for balanced allocation of students is defined as follows: the actual number of students in each teaching class in the class group set must satisfy: the actual number of students ∈ [C×(α-Δ), C×(α+Δ)].

[0073] Step S204: Integrate the campus isolation constraint rules, the professional consistency constraint rules, and the student equalization allocation constraint rules to construct an automatic class splitting and merging rule set.

[0074] The system generates three interrelated rules with different focuses from different dimensions (space, logic, and quantity). The task of this step is to integrate and encapsulate these scattered, logical constraints into a unified, structured rule set object.

[0075] Specifically, the rule set object explicitly defines the application priority (e.g., campus isolation is the highest priority) and interaction relationships of each rule. Through this structured encapsulation, the system transforms complex management concepts and optimization goals into standardized input parameters that can be directly invoked by the algorithm. This allows the clustering analysis engine to focus solely on the logic defined in the rule set object, without needing to concern itself with the specific source or business meaning of the rules. This significantly reduces the coupling between system modules and improves the modularity, maintainability, and scalability of the entire method.

[0076] In the above implementation, the three core dimensions of space constraints, professional logic, and resource balance are processed in sequence, and the processing results are systematically constructed into a structured rule set. This transforms the decision-making process of splitting and merging classes, which relies on human experience in the field of academic affairs management, into an automated intelligent decision-making mechanism with clear priorities, quantifiability, and calculability.

[0077] In practical applications, the built-in optimization rules (such as professional consistency and student balance) and dynamic calculations (such as load coefficients) make the final task implementation plan significantly better than the traditional manual method in terms of feasibility, rationality and resource utilization efficiency. This lays a solid foundation for subsequent scheduling and other processes, and realizes the intelligent upgrade of the core links of academic affairs management.

[0078] Referring to Figure 3, as one implementation of step S105, the steps of acquiring student class data, applying an automatic class splitting and merging rule set to perform cluster analysis on the student class data, and generating a class grouping set include:

[0079] Step S301: Based on student class data, which includes the student's campus identifier, major identifier, and class size, and combining the campus isolation constraint rule, major consistency constraint rule, and class size distribution constraint rule of the automatic class splitting and merging rule set, perform a cluster analysis process.

[0080] The underlying logic of this step lies in associating and binding the data objects to be processed with preset intelligent decision-making rules, thus setting a complete input context for subsequent automated calculations. Student class data includes three basic dimensions for grouping: geographical location (campus identifier), academic attributes (major identifier), and size quantification (class size). The automatic class splitting and combining rule set acts as an intelligent recipe for processing this data. Executing this process means that the system is not performing unguided, arbitrary data grouping, but rather initiating a controlled, goal-oriented, multi-step calculation process.

[0081] Step S302: Apply campus isolation constraint rules, perform partitioning operation on all student class data according to campus identifier, and generate multiple independent campus datasets. Classes in each independent dataset share the same campus identifier.

[0082] The logical principle behind this step is to first divide the problem into domains based on the rigid constraints of physical space, in order to reduce the complexity of the problem and ensure the underlying feasibility of the solution.

[0083] In this embodiment, the campus isolation constraint, as the highest priority rule, is applied to the partitioning operation. This is not a simple data filtering, but an algorithmic strategy. Based on the key attribute of campus identifier, the system divides the massive list of all classes in the school into several smaller, independent datasets for each campus.

[0084] Understandably, merging classes across different campuses is generally physically infeasible, thus any cross-campus class merging solution is fundamentally invalid. By first partitioning the campuses, the system decomposes a complex global optimization problem into multiple subproblems that can be solved independently within each campus. This not only significantly reduces the amount of data and complexity of subsequent calculations, but more importantly, it avoids all invalid solutions caused by geographical isolation, ensuring that all subsequent optimization operations are based on a realistic and feasible foundation.

[0085] Step S303: For each campus's independent dataset, apply the major consistency constraint rule to group classes into preliminary major group queues based on major identifiers, with each class in each queue having the same major identifier;

[0086] After spatial segmentation, the data is further aggregated according to the inherent logic of teaching management and academic attributes to optimize teaching organization and management. Within each campus's independent dataset, the system applies professional consistency constraint rules.

[0087] Specifically, the core of this step is grouping, with the goal of creating an initial professional grouping queue, that is, grouping students of the same major into the same logical set as much as possible.

[0088] Understandably, a major is the core organizational unit of teaching activities, and students in the same major usually have a unified training program, curriculum system, and teaching schedule. Prioritizing the merging of classes with the same major can maximize the relevance of teaching content, maintain the synchronization of teaching progress, and greatly simplify subsequent academic affairs such as course scheduling, examination administration, and student registration management. The formation of this initial major grouping queue signifies that the system has constructed logical units with clear teaching significance from the original data list, laying the structural foundation for the final formation of teaching classes.

[0089] Step S304: On the initial professional grouping queue, apply the constraint rule of balanced distribution of the number of students, analyze the distribution of the number of students in the class within the queue and perform the balanced optimization operation, regroup the classes to match the preset capacity threshold, and generate the optimized grouping queue set.

[0090] The underlying logic of this step is to optimize the initial grouping at a quantitative level while satisfying the aforementioned structural constraints, so as to achieve balanced and efficient utilization of teaching resources (teacher workload, classroom capacity).

[0091] Specifically, after professional grouping, the initial professional group queues may vary significantly in size. Some queues may have too many students, exceeding the reasonable capacity of a teaching class; others may have too few students, resulting in idle resources. At this point, the system applies the constraint rule of balanced student allocation and initiates a balanced optimization operation. The core of this operation is to quantitatively analyze the distribution of students in each class within the queue and intelligently adjust it based on a preset capacity threshold (a dynamic and reasonable range calculated based on factors such as teaching class capacity and teacher workload).

[0092] For example, the system might split an excessively large professional queue into several appropriately sized teaching classes based on a threshold; or it might merge several small queues with similar majors to achieve economies of scale. This process is no longer a simple classification but involves complex decisions based on load balancing algorithms. The goal is to ensure that each optimized grouped queue meets both the requirements for teaching effectiveness and the principles of optimal resource allocation in terms of the number of participants, making the generated grouping scheme not only structurally sound but also economically feasible.

[0093] Step S305: Integrate the optimized grouping queue set from all independent datasets of all campuses to generate the class grouping set.

[0094] The logical principle behind this step is to globally summarize and encapsulate the results of distributed processing, providing standardized input for the next stage of verification and optimization.

[0095] Specifically, after independent optimization within each campus, the system obtains multiple optimized group queue sets. By integrating these scattered and local optimization results into a unified and global final class group set, it ensures that the final output is a data object with a complete structure and standardized format, clearly defining which original classes (based on campus, major, and student balance rules) will be combined for each planned teaching class in the new semester.

[0096] In the above implementation, the physical feasibility of the plan is first ensured by campus zoning; then, the rationality and orderliness of teaching management are guaranteed by professional aggregation; and finally, the efficiency of resource utilization is achieved through balanced student enrollment. This technical solution transforms the traditional model, which relies on personal experience and manual operation, into a structured, automated, and quantifiable intelligent computing process. It not only improves the efficiency of task implementation but also generates grouping schemes that are superior in terms of feasibility, rationality, and economy through built-in multiple optimization rules, providing core algorithmic support for building an intelligent academic affairs management system.

[0097] Referring to Figure 4, as one implementation of step S106, performing group conflict detection on the class group set, and generating group conflict identifiers based on the maximum number of classes a teacher can teach in the teacher configuration dataset and the teacher's teaching history data, includes the following steps:

[0098] Step S401: Traverse each teaching class in the class group set and extract the associated teacher identifier and the number of classes assigned to it;

[0099] The class grouping set is the output of cluster analysis, defining a preliminary teaching task allocation scheme. This step involves traversing this operation to thoroughly examine each teaching class unit in the scheme. Its core action is to extract two crucial resource allocation attributes from each teaching class unit: "teacher identification" (identifying the task executor) and "number of classes allocated" (identifying the teacher's total workload in this scheme). This process essentially deconstructs the complete scheme into a teacher-centric load view, transforming the macro-level grouping problem into a micro-level examination of whether the workload of each individual teacher is reasonable and whether the task combination is scientific.

[0100] Step S402: Compare the number of assigned classes with the maximum number of classes that the corresponding teacher can teach in the teacher configuration dataset. If the maximum number of classes exceeds the limit, generate a teacher over-limit conflict flag.

[0101] The underlying logic of this step is to perform a basic resource capability compliance check to ensure that the task allocation plan does not exceed the hard capability limit of the core resource (teachers).

[0102] Specifically, the system compares the extracted number of assigned classes with the predefined or dynamically adjusted maximum number of classes a teacher can teach in the teacher configuration dataset to verify whether the plan exceeds the teaching load limit that the teacher can bear in the current cycle. If the number of assigned classes exceeds the limit, a teacher overload conflict flag is immediately generated.

[0103] Understandably, when a teacher is overburdened with teaching duties, their energy will inevitably be scattered, potentially reducing the depth of lesson preparation, the quality of instruction, and the time spent interacting with students. Therefore, this rule serves as a firewall to ensure basic teaching quality. The identified conflicts exceeding the limit represent fundamental resource misallocation issues that must be addressed first. Subsequent optimization efforts must resolve these conflicts before any solution is fundamentally feasible.

[0104] Step S403: For teaching classes associated with teacher over-limit conflict markers, retrieve historical teaching major identification records from the teacher's teaching history data, and detect the duplication rate between the current teaching class's major identification and the historical teaching major identification.

[0105] In addition to identifying resource overload, historical data is further introduced for correlation analysis to identify potential risks that may exacerbate the load problem or where there is room for optimization.

[0106] Specifically, detecting the duplication of professional identifiers includes: obtaining the set of professional identifiers for the most recent N semesters from the teacher's teaching history data; calculating the frequency of the professional identifier of the current teaching class in the set; if the frequency exceeds a preset frequency threshold, it is determined to be a professional conflict.

[0107] Understandably, the system does not perform this check on all classes, but rather specifically on classes "related to teacher overload conflicts." This indicates that the system believes the reasonableness of the quality of tasks assigned to teachers already under overload needs to be examined. Operationally, the system "retrieves historical teaching specialty identification records from the teacher's teaching history data" and "detects the degree of duplication between the current teaching class's specialty identification and historical teaching specialty identification."

[0108] The principle behind this step is to use historical data to predict future workload: if a teacher is assigned to teach the same or highly similar subject for several consecutive semesters, although they may become proficient due to accumulated experience, they are also more likely to experience teaching burnout due to a lack of variation, which limits the breadth of their knowledge application. It may also make the allocation of faculty in the department less flexible. Detecting duplication is a quantitative assessment of this potential risk.

[0109] Step S404: If the repetition exceeds the preset repetition threshold, generate a professional conflict marker and associate it with the corresponding teaching class;

[0110] The preset repetition threshold enables the system to distinguish between "reasonable continuation of experience" and "repetition that may lead to negative effects." This threshold can be flexibly configured according to academic management strategies.

[0111] In this embodiment, the preset repetition threshold can be set to "teaching the same major three times within the last three semesters." When the calculated repetition exceeds this threshold, the system generates a major conflict flag. This flag has a deeper meaning than "excessive faculty resources"; it no longer indicates a simple excess of "quantity," but rather an unreasonable task combination structure. This flag serves as a prompt for the optimization algorithm. When reducing the workload of this teacher (addressing the quantity problem), priority should be given to adjusting these highly repetitive teaching tasks to promote the diversity of teacher tasks and the rational flow of faculty resources within departments. This represents a deeper level of optimization strategy.

[0112] Step S405: Verify whether the actual number of students in each class in the class grouping set exceeds the dynamic adjustment capacity limit in the parameter configuration dataset, and generate a capacity conflict flag for the excess portion.

[0113] The underlying logic of this step is to check the physical capacity of the teaching tasks to ensure that the grouping scheme matches the actual teaching resources (such as classroom capacity and teaching facilities).

[0114] Specifically, the dynamic adjustment of the capacity limit is derived from the normalized capacity (such as a multiple of the number of classroom seats) previously calculated based on the "expansion flag". This step checks the one-to-one matching relationship: whether the number of students allocated to each teaching class exceeds the maximum capacity that the preset physical space or resources can accommodate. If it does, a capacity conflict flag is generated. The fundamental principle is to ensure the normal conduct of teaching activities: a teaching class exceeding the classroom capacity will not be able to conduct face-to-face teaching normally. Therefore, this is a hard constraint check on the physical feasibility of the solution, identifying conflicts between the solution and the infrastructure.

[0115] Step S406: Aggregate the teacher over-limit conflict markers, major conflict markers, and capacity conflict markers to construct group conflict identifiers.

[0116] Specifically, scattered and different types of conflict information are systematically integrated and encapsulated to generate a comprehensive, clear, and easily automated diagnostic report for subsequent processing. By creating a structured data object containing information such as conflict type, associated class identifier, and severity, this object provides highly precise input for the next stage of optimizing the class grouping set.

[0117] Next, the optimization algorithm can clearly know what problems exist in the current solution, the location of the problems, the nature of the problems, and their priorities (for example, prioritizing the resolution of hard conflicts such as teacher overcapacity and capacity conflicts) based on this object, thereby carrying out targeted and efficient iterative optimization.

[0118] In the above implementation method, the first step is to ensure that task allocation does not exceed the absolute workload capacity of teachers. Then, the historical rationality of task combinations is thoroughly evaluated to mitigate potential risks, while simultaneously verifying the matching degree between the plan and physical resources. Finally, this technical solution integrates all identified problems into a structured diagnostic report, transforming the traditional plan review process, which relies on manual experience and subjective judgment, into an automated, systematic, and deeply integrated intelligent pre-inspection process based on historical data. This not only identifies superficial resource overload issues but also uncovers deeper configuration risks, thereby improving the feasibility, rationality, and robustness of the final task implementation plan. It is a core guarantee for achieving intelligent decision-making in academic affairs management.

[0119] Referring to Figure 5, as one implementation of step S107, the steps of optimizing the class group set based on the group conflict identifier, assigning teacher tasks, and generating task implementation plans include:

[0120] Step S501: parse the teacher over-limit conflict marker in the group conflict identifier and identify the associated teaching class and teacher identifier;

[0121] In particular, after identifying the teacher overload conflict marker, the system further extracts the teaching class and teacher identifier associated with this marker. This means that the system can accurately understand which teaching class (teaching class identifier) ​​allocation scheme caused which teacher (teacher identifier) ​​workload to exceed their personal capacity limit.

[0122] Step S502: Based on the maximum number of classes that a teacher can teach in the teacher configuration dataset, perform a teacher reallocation operation for teaching classes with overstaffing. Select alternative teachers from the teacher configuration dataset whose number of classes they can teach has not reached the maximum, and generate an updated class grouping set.

[0123] The process of reassigning teachers includes: obtaining the course identifiers of classes with excess teachers; querying the set of teachers in the teacher configuration dataset who can teach the course and whose current number of assigned classes has not reached the limit; sorting the teacher set based on teacher teaching quality evaluation data, and selecting the teacher with the highest evaluation as the replacement teacher.

[0124] Understandably, the logic behind this step is to perform optimal or suboptimal matching within a constrained resource pool to address the problem of uneven resource allocation and ensure the feasibility of the solution. For the problematic teaching class identified in the previous step, the system performs a teacher reassignment operation. The system searches the entire teacher pool for teachers who are qualified to teach the course (qualification information exists in the teacher configuration dataset) and whose current workload is not yet saturated. This is essentially a constraint-based optimization search algorithm whose goal is to rebalance the workload of all teachers and eliminate overload for individual teachers.

[0125] Step S503: For the professional conflict marker in the group conflict identifier, retrieve the candidate teachers in the teacher's teaching history data whose professional identifier matches the current teaching class with a lower than the preset matching threshold, and reassign the candidate teachers to the corresponding teaching classes.

[0126] The system retrieves teachers' teaching history data and calculates the overall match between the candidate teachers' past teaching specialties and the specialties of their current classes. Then, the system proactively selects teachers whose overall match is below a preset threshold for assignment. This prevents a teacher from being confined to a single professional field for an extended period, promoting knowledge expansion and updating of teaching content. Furthermore, this also represents a structural optimization of the teaching team, making faculty allocation more flexible.

[0127] Step S504: Verify the capacity conflict flag in the group conflict identifier. For teaching classes where the actual number of students exceeds the dynamic adjustment capacity limit, adjust the class capacity or split the class based on the expansion flag in the parameter configuration dataset, and update the class group set.

[0128] Among them, the system verifies capacity conflict flags, that is, it handles teaching classes that have more students than the physical space such as classrooms can accommodate.

[0129] In some embodiments, there are two basic strategies to resolve this conflict. The system will intelligently select the expansion flag in the parameter configuration dataset: one is "adjust class capacity", that is, if there is sufficient physical space and expansion is allowed, the capacity of the teaching class will be appropriately increased; the other is "split class", that is, when expansion is not possible, the excessively large teaching class will be split into two or more new teaching classes that meet the capacity limit according to certain rules (such as prioritizing the consistency of majors).

[0130] Step S505: Based on the optimized class grouping set, traverse each teaching class and assign teacher identifiers to generate a teacher task allocation scheme;

[0131] The system first optimizes and stabilizes the grouping structure, then binds it to specific responsible individuals (teachers) to generate a clear responsibility allocation list. After completing all the conflict resolution and structural adjustments, the system obtains an optimized set of class groups, which achieves a favorable state in terms of load, structure, and capacity. At this point, the system needs to traverse each class and assign a teacher identifier, meaning that the system explicitly designates a unique responsible teacher for each finalized class. The final teacher task allocation scheme is a complete mapping table that clearly defines "who (teacher) teaches what (course) in which class (teaching class)."

[0132] Step S506: Integrate the class group set and teacher task allocation plan to construct a task implementation plan.

[0133] Specifically, the system integrates the class group set (which defines which student classes make up each teaching class) and the teacher task allocation scheme (which defines which teacher is responsible for each teaching class) to create a data entity containing all necessary information (such as teaching class ID, student class list, teacher ID, teaching campus, class capacity, etc.) with a clear structure (such as using JSON or XML format).

[0134] In the above implementation, precise signals from the conflict detection stage are received, and core conflicts such as resource overload, unreasonable task structure, and physical capacity mismatch are resolved sequentially according to priority. Through dynamic teacher reallocation, optimization adjustments based on historical similarity, and flexible capacity handling strategies, a preliminary grouping scheme with potential defects is iteratively optimized into a high-quality, executable scheme that achieves a balance in terms of manpower load, professional configuration, and physical resources. The final output structured task implementation scheme not only automates task allocation but also improves the feasibility and rationality of the allocation results through built-in multiple optimization logics.

[0135] Referring to Figure 6, as one implementation of step S503, the step of reassigning candidate teachers to corresponding teaching classes based on the professional conflict markers in the group conflict identifiers includes:

[0136] Step S601: Parse the professional conflict markers and extract the professional identifier and course qualification requirement code of the current teaching class;

[0137] The current teaching class's professional identifier indicates the subject area where the conflict occurs (e.g., "Software Engineering"), which is the target direction for finding alternative teachers; while the "Course Qualification Requirements Code" clarifies the legal or school-mandated qualification thresholds that must be possessed to undertake the teaching of this course (e.g., "Senior Programmer Qualification Certificate" or "University Teacher Qualification Certificate - Computer Science").

[0138] Step S602: Traverse the teacher configuration dataset and construct a teaching ability tag set for each teacher, including the professional identifier of historical teaching and the subject coverage code of the teacher qualification certificate;

[0139] Among them, the history teaching specialty identifier is an objective record of a teacher's practical experience, reflecting the professional field in which they have actually engaged in teaching activities; while the subject coverage code of the teacher qualification certificate is the officially certified capability boundary, representing the scope of teaching that is recognized by authoritative institutions. Combining these two into a tag set is equivalent to creating a standardized profile for each teacher that includes "practical experience" and "certification scope," allowing the system to move away from subjective descriptions and instead search and evaluate teacher capabilities based on these two types of objective, structured data tags.

[0140] Step S603: Select teachers whose subject coverage codes for teacher qualification certificates include course qualification requirement codes as candidate teachers;

[0141] The extracted course qualification requirement code is used as a filter applied to all constructed teacher competency tag sets. Only teachers whose subject coverage codes in their teacher qualification certificates include the required code are allowed to proceed to the next round of screening. This step ensures that any teacher considered for reassignment first meets the minimum legal or regulatory qualification requirements for teaching the course, fundamentally eliminating the risk of unqualified teachers and guaranteeing that all subsequent optimization operations regarding experience matching are conducted within a compliant and secure framework.

[0142] Step S604: Generate a matching degree based on the correlation between the professional identifiers of historical teaching and the professional identifiers of the current teaching class, and filter out candidate teacher identifiers with matching degrees lower than the dynamic matching degree threshold.

[0143] For candidate teachers who pass the qualification screening, the system no longer searches for the best match. Instead, it calculates the correlation (i.e., matching degree) between the "historical teaching major identifier" and the "current teaching class major identifier" and actively filters out teachers whose matching degree is lower than a certain preset threshold. The matching degree can be based on a simple identifier consistency judgment (such as a high matching degree if they are completely identical) or on a more complex major association graph (such as a medium matching degree for similar majors).

[0144] For example, a low threshold (such as 0.3) can be preset, meaning the system deliberately selects teachers who are qualified to teach but lack recent teaching experience in this major. This design directly echoes the original intention of this application to break the rigidity of teaching staff, and forcibly introduces the diversity of teaching experience through algorithms. It aims to avoid the mindset that teachers may develop due to teaching a fixed major for a long time, and promote the exchange of teaching methods between different majors.

[0145] In some embodiments, the matching degree calculation includes: if the historical teaching major identifier is completely consistent with the target major identifier, the matching degree is 1.0; if there is a related major defined by the course knowledge graph, the matching degree is 0.5-0.8; if there is no related major, the matching degree is 0.0.

[0146] Based on this, dynamic matching thresholds can be set according to the differences in teaching attributes and risk levels of courses: a high threshold (such as 0.7) can be set for "professional core courses" with high requirements for knowledge system coherence to control the teaching risks caused by excessive experience span; a low threshold (such as 0.4) can be set for "general education courses" with high requirements for knowledge breadth to encourage broader experience crossover; and an intermediate threshold (such as 0.6) can be set for "practical courses" that require a certain professional foundation to achieve a balance between safety and innovation.

[0147] Furthermore, a fixed safety threshold (e.g., 0.2) can be set. Once the matching is too low (i.e., experience is almost completely irrelevant), manual review will be triggered regardless of the course type. This establishes a clear decision boundary and intervention mechanism between pursuing teacher rotation and ensuring basic teaching quality. This application's embodiment transforms qualitative management experience (such as "avoiding rigidity," "appropriate rotation," and "adapting to the course") into a set of calculable, configurable, and graded quantitative rules, achieving refined, intelligent, and risk-controllable optimization of teacher allocation.

[0148] Step S605: Based on the maximum number of classes an instructor can teach and the matching degree, reassign candidate instructors to the corresponding teaching classes.

[0149] Specifically, there may be multiple candidate teachers with low matching degree after screening. Therefore, the system needs to consider two key factors when making the final allocation: one is the "maximum number of classes that a teacher can teach", which is a rigid constraint on the teacher's workload to ensure that the total number of classes taught by the teacher does not exceed the limit after allocation; the other is the "matching degree ranking", which is the core objective of this optimization scheme.

[0150] In some embodiments, the allocation logic is as follows: Provided the maximum number of classes a teacher can teach is met, tasks are preferentially allocated to teachers with the lowest matching degree (i.e., the greatest experience gap) and who are not overloaded. If multiple teachers have the same matching degree, fairness principles such as prioritizing those with the fewest classes can be applied. Through this step, the system ultimately outputs a solution that both adheres to the hard constraints of teacher resources and workload and maximizes the optimization goal of "low-matching degree allocation" (i.e., professional rotation).

[0151] The above implementation transforms the complex problem of relying on human experience for judgment and coordination into an automated decision-making process based on data (qualification codes, historical records) and clear rules (qualification filtering, matching degree thresholds, load constraints). This technical solution not only ensures the compliance of teaching arrangements through qualification verification but also proactively breaks down long-term fixed teaching patterns for teachers through a unique "low matching degree" screening and allocation mechanism, promoting the optimal allocation of teaching resources and the cross-disciplinary development of the teaching staff. Ultimately, while efficiently resolving professional conflicts, it enhances the foresight and scientific nature of academic management, enabling the automated task allocation system to not only correctly allocate tasks but also further optimize the faculty structure.

[0152] Referring to Figure 7, as a further implementation of the method for implementing teaching tasks, after the steps of optimizing the class group set based on group conflict identifiers, assigning teacher tasks, and generating task implementation plans, the method further includes:

[0153] Step S701: Collect the scheduling conflict records after the implementation of the task plan in the academic affairs system, and extract the scheduling conflict rate as the first feedback indicator;

[0154] Among these, scheduling conflicts (such as multiple teachers vying for the same classroom at the same time, or one teacher being assigned to teach two classes at the same time) serve as a litmus test for the rationality of the task implementation plan. The "scheduling conflict rate" is a highly condensed and quantitative expression of this complex phenomenon, which can be defined as the proportion of the number of conflicting classes to the total number of classes.

[0155] Understandably, a high scheduling conflict rate directly indicates that the grouping scheme generated in the early stage based on the rule set has potential irrationality or is too idealistic in the time or space dimensions, providing a clear signal that the system needs to be optimized.

[0156] Step S702: Obtain the dynamically adjusted capacity limit from the parameter configuration dataset, and calculate the class capacity utilization rate as the second feedback indicator by combining it with the actual number of students in the class.

[0157] The system collects data on "actual number of students in each class" (the number of students in each class determined by the grouping scheme) and "dynamically adjustable capacity limit in the parameter configuration dataset" (the maximum allowable capacity preset by the system for that class). It then calculates the "class capacity utilization rate" (usually expressed as actual number of students / capacity limit). The class capacity utilization rate reflects the system's focus on resource efficiency. A low utilization rate means that teaching resources (such as classroom space) are idle, resulting in waste. A consistently high utilization rate, even reaching saturation, may indicate that the current capacity settings cannot meet actual needs, posing a bottleneck risk and potentially indirectly leading to scheduling difficulties (because classroom resources that meet capacity requirements are even more scarce).

[0158] Step S703: Analyze the correlation between the first feedback indicator and the second feedback indicator;

[0159] Step S704: When the scheduling conflict rate exceeds the preset conflict threshold and the class capacity utilization rate is lower than the preset utilization rate threshold, generate a teacher configuration optimization instruction, and recalculate the upper limit of the number of classes that the teacher can teach based on the teacher teaching quality evaluation data, and update the teacher configuration dataset.

[0160] A high conflict rate usually indicates scheduling difficulties, but if class capacity utilization is also low, it largely rules out the possibility of insufficient classroom resources, thus pointing to an unreasonable allocation structure of teacher tasks. For example, it might lead to excessive competition for specific types of classrooms (such as specific laboratories), or overly rigid teacher schedules. Through this correlation analysis, the system can make a more accurate judgment: the current problem is more likely to be solved by optimizing teacher resource allocation rather than simply increasing classroom capacity.

[0161] Next, when the system determines that teacher allocation needs optimization, it can recalculate the maximum number of classes a teacher can teach based on teacher teaching quality evaluation data. This means that a teacher's workload capacity is not a static value, but a variable dynamically linked to their teaching performance. Teachers with high teaching evaluations indicate that their teaching effectiveness and workload capacity may be stronger, and the system will appropriately increase the maximum number of classes they can teach, thereby making fuller use of high-quality resources; conversely, it may decrease the maximum number of classes they can teach to ensure teaching quality.

[0162] Subsequently, the teacher configuration dataset is updated, and this calibration result is solidified into the core configuration of the system. Based on the actual operating results and teacher performance, the system continuously and adaptively optimizes its internal resource model so that its next decision can be based on more accurate data.

[0163] Step S705: When the class capacity utilization rate continues to be higher than the preset utilization rate upper limit threshold, generate parameter configuration optimization instructions, dynamically increase the teaching class capacity adjustment multiple according to the expansion flag, and update the parameter configuration dataset;

[0164] In some embodiments, when the class capacity utilization rate is above 90% for three consecutive semesters, the expansion factor is adjusted from 5 times to 10 times.

[0165] Understandably, when constructing new task implementation plans, the system tends to set up slightly larger but fewer teaching classes, thereby reducing the demand for classrooms and alleviating scheduling pressure within the overall classroom resource constraint. This is the system's flexible adjustment of resource supply parameters to adapt to actual demand levels.

[0166] Step S706: Input the updated teacher configuration dataset and parameter configuration dataset into the next round of task implementation process.

[0167] The system replaces the original datasets with updated teacher configuration and parameter configuration datasets. When the task implementation process for the next academic year begins, the system will directly use these new parameters, which have been tested in practice and calibrated with optimized algorithms, to make decisions. Thus, the completion of each task implementation marks the beginning of the next, even better decision. Through this continuous iterative optimization, the system can constantly approach the optimal resource allocation state, demonstrating strong adaptability and intelligence.

[0168] In the above implementation, real-world data (scheduling conflict rate, capacity utilization rate) after the plan is executed is collected as feedback signals. Intelligent correlation analysis is then used to accurately diagnose the root causes of problems, allowing for targeted adjustments to the system's core parameters (teacher workload capacity, class capacity baseline). This closed-loop process enables the system to transcend static automation and become a dynamically adaptable intelligent agent that changes in the actual teaching environment. In practical applications, this technical solution not only solves problems in the current cycle but also, through iterative parameter updates, makes subsequent decisions more accurate and efficient, ultimately achieving continuous synergistic optimization of the task implementation plan in terms of feasibility, resource utilization efficiency, and teaching quality assurance.

[0169] Referring to Figure 8, as a further implementation of the method for implementing academic affairs tasks, after the step of generating the task implementation plan, the method further includes:

[0170] Step S801: Obtain the class grouping set, teacher task allocation scheme, and site constraint information from the parameter configuration dataset in the task implementation plan;

[0171] Among them, the class group set defines the organizational form of the student group. This data structure contains the student size and professional attribute distribution when the course is executed in the future; the teacher task allocation scheme binds specific teacher resources with the aforementioned student group and defines the subject of the teaching task execution; the site constraint information specifies the specific requirements of each teaching class for physical space (such as classrooms and laboratories), including hard conditions such as site type, capacity, and special equipment.

[0172] Step S802: Extract historical course scheduling data from the academic affairs system database, and construct a conflict prediction model based on the course scheduling conflict records of similar course combinations in the same semester.

[0173] In this embodiment of the application, the construction of the conflict prediction model includes: extracting feature vectors of course combinations from the historical course scheduling database, including the number of teaching classes, the average daily teaching hours of teachers, and the density of venue usage; labeling historical course scheduling conflict events as training labels, and training a multi-class conflict predictor using the random forest algorithm; adjusting the model parameters through cross-validation, and outputting a probability distribution matrix of conflict types.

[0174] Specifically, in traditional academic affairs management, scheduling conflicts often only emerge during the actual scheduling process. This step aims to transform this posterior problem into a priori prediction. First, by leveraging the wealth of historical scheduling data, we can uncover actual conflict events and their associated characteristics that occurred during the final scheduling stage for similar course combinations in the same semester. These characteristics may include the tightness of teachers' time windows, the concentration of demand for specific types of venues, and the intensity of competition for booking special equipment.

[0175] Then, using these historical feature data labeled "conflict" or "non-conflict", a conflict prediction model is trained through supervised learning using classification algorithms such as random forest. The essence of this model is to learn the complex mapping relationship between resource constraint patterns in historical data and the final conflict outcome.

[0176] Understandably, the conflict prediction model constructed in this way does not rely on simple rules (such as "a teacher cannot teach two classes at the same time") to make judgments. Instead, it can comprehensively assess the systemic risks formed by the interweaving of multiple factors such as the spatiotemporal distribution of teacher workload, the aggregation effect of venue resource demand, and the continuity requirements of equipment use, thereby achieving a quantitative assessment of the conflict probability of the implementation plan for the new task.

[0177] Step S803: Input the task implementation plan into the conflict prediction model and output the virtual scheduling conflict probability value and conflict type distribution;

[0178] Specifically, the virtual scheduling conflict probability value output by the model is a comprehensive risk assessment index. It quantifies the likelihood that the proposed solution will encounter obstacles in future scheduling phases under current resource constraints. Simultaneously, the model also outputs a conflict type distribution, indicating whether the conflict risk primarily stems from teacher time conflicts, venue resource conflicts, or special equipment conflicts. This distribution information has crucial diagnostic value; it not only tells decision-makers that the solution is high-risk but also precisely identifies the main sources and nature of the risk, thus providing clear and targeted guidance for subsequent optimization measures.

[0179] In this embodiment, the model maps the feature vector of the current task implementation plan to the interval [0,1]. For example, a calculated value of 0.85 indicates an 85% probability of conflict. The core principle of this model is to simulate the intensity of resource competition in a scheduling environment. For example, if a teacher is assigned to four campuses, insufficient commuting time triggers a time conflict; if three classes requiring projectors are concentrated in one location, insufficient equipment turnover triggers an equipment conflict. The resulting conflict type distribution, i.e., the output probability vector, is [0.6, 0.3, 0.1], corresponding to the weights of teacher / venue / equipment conflicts, respectively.

[0180] Step S804: When the probability value of virtual scheduling conflict exceeds the preset risk threshold, retrieve the set of available venues in the teaching affairs system based on the venue constraint information, and retrieve the available time slots of teachers to be assigned in the teaching affairs system based on the teacher task allocation scheme.

[0181] Among them, the preset risk threshold is based on the retrospective analysis of historical operational data or management strategy settings. It is a control parameter that balances the efficiency of plan formulation and the success rate of the plan. Once the predicted probability of conflict exceeds the threshold, the system will immediately and proactively explore the possibility of resource adjustment.

[0182] Specifically, if the probability value of virtual scheduling conflict exceeds the preset risk threshold, the idle resource pool data of the academic affairs system will be retrieved. The idle resource pool data includes the set of available time periods, the set of available venues, and the idle time periods of teachers to be assigned. The set of available venues refers to the idle venues that meet various teaching requirements, and the idle time periods of teachers to be assigned refer to the working time blocks of each teacher that have not yet been occupied.

[0183] Step S805: Perform preprocessing operations according to the conflict type distribution matching optimization strategy library to obtain the preprocessed task implementation plan and resource consumption prediction report;

[0184] Specifically, based on the diagnosis results (type distribution) of conflict risks, the system calls preset, targeted treatment strategies to fine-tune the original plan in order to eliminate or reduce the main conflict risks. The system maintains an optimization strategy library, which contains processing logic for different conflict types.

[0185] For example, when the conflict type distribution shows that teacher time conflicts are the main issue, the system will invoke the corresponding strategy. The core of this strategy is to use scheduling algorithms such as time slot filling algorithms, based on the retrieved available teacher time slots, to find and reassign the conflicting courses to other suitable available time slots for the teacher, thus alleviating congestion on their schedules. For venue conflicts, the strategy involves dynamically replacing the conflicting course with an alternative venue from the available venue set that meets the requirements in terms of type, capacity, and equipment, while reducing overall venue competition pressure. For conflicts related to specific equipment needs, the strategy may involve adjusting the grouping priority or scheduling of relevant practical teaching classes to distribute their demand for scarce equipment resources more evenly across the timeline.

[0186] Ultimately, the pre-processed task implementation plan is a new version based on the original plan, incorporating various optimization operations (such as adjusting time slots and changing venues). It inherits the advantages of the original plan while reducing the identified conflict risks. Furthermore, the system simultaneously generates a resource occupancy prediction report. This report, based on the pre-processed plan, simulates and extrapolates the potential resource occupancy after future scheduling, such as predicting peak utilization rates for each venue at different time periods and the tight schedules of teachers. This resource occupancy prediction report generates a resource load heatmap or timeline view by simulating and aggregating the trajectories of all resource scheduling intentions in the optimized plan. This prediction report not only proves the theoretical feasibility of the optimized plan but also provides valuable, forward-looking decision support information for academic administrators, enabling them to anticipate potential resource bottlenecks after the plan's implementation.

[0187] Step S806: Feed back the preprocessed task implementation plan and resource usage forecast report to the parameter configuration dataset update process to update the expansion flag and campus distribution information.

[0188] In this embodiment, the system feeds this data back to the parameter configuration dataset update process, signifying the initiation of a higher-level parameter tuning process. For example, if the resource occupancy forecast report consistently shows that a certain type of venue (such as a large computer room) remains under high load after preprocessing, and the preprocessing scheme frequently avoids conflicts by dynamically replacing venues, the system can infer that the current resource configuration of this type of venue may be insufficient. In this case, the system can generate an optimized expansion flag, allowing courses requiring the use of this type of venue to be offered at a larger capacity (e.g., from 60 students to 80 students) in future task implementations, thereby reducing the batches of demand for scarce venue resources at the source.

[0189] Similarly, by analyzing cross-campus class rescheduling records, campus distribution information can be optimized, and the course capacity ratio of each campus can be planned more rationally. This closed-loop mechanism enables the system to learn from each plan formulation and optimization, continuously adjusting its underlying parameters and constraints, thus making the implementation of the next round of tasks more scientific and efficient from the starting point.

[0190] In the above implementation, the resource conflict risks that may be exposed during the scheduling stage are significantly mitigated and quantitatively assessed during the task allocation stage, thereby transforming academic management from a passive conflict response model to a proactive conflict prevention model. This technical solution not only provides an objective measure of solution feasibility through virtual conflict probability assessment, but also achieves automated and precise correction of high-risk solutions through a matching optimization strategy library. Finally, through a feedback mechanism, optimization experience is solidified into the system's adaptive capabilities, improving the first-time pass rate of task implementation plans and the success rate of scheduling, achieving better allocation and utilization of teaching resources in both time and space.

[0191] Referring to Figure 9, as a further implementation of the method for implementing academic affairs tasks, the steps of performing preprocessing operations based on the conflict type distribution matching optimization strategy library include:

[0192] Step S901: Based on the type of teacher time conflict, and in conjunction with the teacher assignment identifier in the teacher task allocation scheme and the available time slots of the teachers to be assigned, the teaching time slots are reallocated.

[0193] When the conflict prediction model identifies a high risk of teacher time conflict, it essentially predicts that a teacher (uniquely identified by the "teacher identifier") will have an excessively concentrated or overlapping teaching task in the future scheduling phase, exceeding the reasonable time window.

[0194] In this embodiment, the response strategy does not modify "who teaches" (teacher assignment), but rather optimizes "when to teach" (teaching time slots). The system accesses a dataset of "unassigned teacher free time slots," which essentially represents unoccupied time blocks on the teacher's schedule that can be used to schedule new teaching tasks. The process of reallocating teaching time slots is a local search and scheduling optimization within the teacher's individual timeline. The system attempts to find a suitable new time slot within the teacher's free time slots for teaching tasks with a high risk of conflict, avoiding conflicts with other already scheduled or predicted scheduled tasks. This ensures that, while maintaining the teacher's task affiliation, the distribution of tasks on the timeline is adjusted to dilute time density and eliminate potential conflicts. Its core principle is to guarantee the feasibility and balance of each teacher's work schedule.

[0195] Step S902: Based on the type of site conflict, dynamically replace the site identifier by combining the site constraint information and the set of available sites.

[0196] The steps for dynamically replacing site identifiers include: calculating the spatial distance weights and equipment matching degrees between the original site and the set of available sites; and selecting the site with the highest product of weights and matching degrees as the replacement target.

[0197] Specifically, spatial distance weighting measures the physical location relationship between the alternative and original sites. A higher weight usually means closer proximity or more convenient transportation. The aim is to minimize the additional commuting costs and time losses for students and teachers caused by changing sites, and to maintain the spatial continuity of teaching activities. Equipment matching assesses the degree to which the alternative site meets the functional constraints of the original site, such as whether it also has specific equipment such as projectors and lab benches.

[0198] In this embodiment, the algorithm selects the site with the highest product of "spatial distance weight" and "equipment matching degree" as the replacement target, reflecting the principle of maximizing comprehensive utility in multi-attribute decision-making. This algorithm not only requires the new site to meet basic constraints but also seeks proximity and high functional matching, thereby resolving resource conflicts while minimizing disruption to the original teaching arrangements and improving the acceptability and implementation efficiency of the alternative.

[0199] Step S903: Based on the practical teaching class identifier in the class grouping set, adjust the priority of the practical teaching class grouping according to the equipment demand conflict type.

[0200] Understandably, equipment demand conflicts often arise from multiple "practical teaching classes" (identified by "practical teaching class identifiers") simultaneously or consecutively requesting the same type of scarce equipment (such as large experimental instruments or dedicated computer rooms), exceeding the actual supply capacity of that equipment. A traditional solution might be to increase the number of devices, but this embodiment employs a more intelligent time resource reallocation strategy.

[0201] Specifically, the system needs to accurately identify practical teaching classes requiring special equipment, and then adjust their grouping priority. This involves shifting their allocated time slots to the time intervals with the highest equipment availability in the idle resource pool. The idle resource pool refers to the set of all available teaching time slots, while the time interval with the highest equipment availability refers to the time interval when the target equipment is least idle and under the least competitive pressure. By proactively scheduling teaching tasks with high equipment demand to the off-peak usage period of equipment resources, the equipment load curve can be effectively smoothed, avoiding congestion during peak periods, thereby improving the overall utilization and turnover efficiency of equipment resources.

[0202] In the above implementation, for teacher time conflicts, the workload of individuals is balanced on the time axis by reallocating teaching time slots; for venue conflicts, venue resources are intelligently and smoothly replaced by comprehensively evaluating distance and functional matching; and for equipment demand conflicts, the load balance of scarce equipment resources is achieved by staggered scheduling of time slots.

[0203] In practical applications, this technical solution transforms the resolution of discrete and reactive conflicts that may arise during the scheduling phase into a systematic, proactive, and prediction-based conflict prevention and resource pre-optimization process by integrating them into the task implementation plan generation stage. This not only improves the first-time success rate of the final scheduling plan and reduces the workload of subsequent manual coordination, but also achieves better allocation of teaching resources (teacher time, venue, and equipment) in the spatiotemporal dimensions through algorithms, thereby improving the overall operational efficiency and intelligence level of the academic affairs management system.

[0204] This application also discloses an administrative task implementation system based on automatic class splitting and merging rules.

[0205] A system for assigning academic affairs tasks based on automatic class splitting and merging rules, specifically including:

[0206] The course qualification data acquisition module is used to acquire teacher qualification data, including the teachers for each course and the number of classes that each teacher can teach;

[0207] The teacher data configuration module is used to dynamically adjust the maximum number of classes that each course instructor can teach based on the input teacher teaching quality evaluation data, and generate a teacher configuration dataset.

[0208] The teaching parameter configuration module is used to obtain course teaching demand data for a specified academic year and semester, which includes teaching class capacity, number of teaching classes, expansion flag and campus distribution information. Based on the expansion flag, the teaching class capacity is dynamically adjusted to an integer multiple of a preset number, and a parameter configuration dataset is generated.

[0209] The automatic class splitting and merging rule building module is used to build an automatic class splitting and merging rule set based on the teacher configuration dataset and parameter configuration dataset. This set includes campus isolation constraint rules, professional consistency constraint rules, and student equalization allocation constraint rules.

[0210] The clustering analysis module is used to obtain student class data, apply the automatic class splitting and merging rule set to perform clustering analysis on the student class data, and generate class group sets;

[0211] The group conflict detection module is used to perform group conflict detection on the class group set. Based on the maximum number of classes that a teacher can teach and the teacher's teaching history data in the teacher configuration dataset, it generates group conflict identifiers.

[0212] The task implementation module is used to optimize the class group set based on the group conflict identifier, assign tasks to teachers, and generate task implementation plans.

[0213] The teaching affairs task implementation system based on automatic class splitting and merging rules in this application embodiment can implement any of the above methods, and the specific working process of each module in the system can refer to the corresponding process in the above method embodiments.

[0214] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the division of a certain module is merely a logical functional division, and in actual implementation there may be other division methods, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0215] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for implementing academic affairs tasks based on automatic class splitting and merging rules, characterized in that, The method includes: acquiring teacher qualification data, including the teachers for each course and the number of classes each teacher can teach; dynamically adjusting the upper limit of the number of classes each teacher can teach based on input teacher teaching quality evaluation data, generating a teacher configuration dataset; acquiring course teaching demand data for a specified academic year and semester, including teaching class capacity, number of teaching classes, expansion flags, and campus distribution information, dynamically adjusting the teaching class capacity to an integer multiple of a preset number according to the expansion flags, generating a parameter configuration dataset; constructing an automatic class splitting and merging rule set based on the teacher configuration dataset and the parameter configuration dataset, including campus isolation constraint rules, professional consistency constraint rules, and student balanced allocation constraint rules; acquiring student class data, applying the automatic class splitting and merging rule set to perform cluster analysis on the student class data, generating a class group set; performing group conflict detection on the class group set, generating group conflict identifiers based on the upper limit of the number of classes each teacher can teach and teacher teaching history data in the teacher configuration dataset; optimizing the class group set based on the group conflict identifiers, assigning teacher tasks, and generating task implementation plans.

2. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 1, characterized in that, The steps for constructing an automatic class splitting and merging rule set based on the teacher configuration dataset and parameter configuration dataset include: parsing the campus distribution information in the parameter configuration dataset to generate campus isolation constraint rules; extracting the preset professional identifier mapping relationship to generate professional consistency constraint rules; calculating the average class load coefficient per teacher based on the upper limit of the number of classes a teacher can teach in the teacher configuration dataset and the teaching class capacity in the parameter configuration dataset to generate a student equalization constraint rule; and integrating the campus isolation constraint rule, professional consistency constraint rule, and student equalization constraint rule to construct an automatic class splitting and merging rule set.

3. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 1, characterized in that, The steps of acquiring student class data and applying the automatic class splitting and merging rule set to perform cluster analysis on the student class data to generate a class group set include: based on the student class data, which includes the student's campus identifier, major identifier, and class size, performing a cluster analysis process in conjunction with the campus isolation constraint rule, major consistency constraint rule, and student size distribution constraint rule of the automatic class splitting and merging rule set; applying the campus isolation constraint rule to perform a partitioning operation on all student class data according to the campus identifier, generating multiple campus-independent datasets, with classes in each independent dataset sharing the same campus identifier; for each campus-independent dataset, applying the major consistency constraint rule to group classes into preliminary major group queues according to the major identifier, with classes in each queue having the same major identifier; applying the student size distribution constraint rule to the preliminary major group queues to analyze the distribution of class sizes within the queues and performing a balance optimization operation, regrouping classes to match a preset capacity threshold, generating an optimized group queue set; and integrating the optimized group queue sets from all campus-independent datasets to generate a class group set.

4. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 3, characterized in that, The steps for performing group conflict detection on the class group set, and generating group conflict identifiers based on the maximum number of classes a teacher can teach in the teacher configuration dataset and the teacher's teaching history data, include: traversing each teaching class in the class group set, extracting the associated teacher identifier and the number of classes assigned; comparing the number of assigned classes with the maximum number of classes a corresponding teacher identifier can teach in the teacher configuration dataset, and generating a teacher over-limit conflict marker if the limit is exceeded; for the teaching class associated with the teacher over-limit conflict marker, retrieving historical teaching major identifier records from the teacher's teaching history data, and detecting the duplication rate between the current teaching class's major identifier and the historical teaching major identifier; if the duplication rate exceeds a preset duplication rate threshold, generating a major conflict marker and associating it with the corresponding teaching class; verifying whether the actual number of students in each teaching class in the class group set exceeds the dynamic adjustment capacity limit in the parameter configuration dataset, and generating a capacity conflict marker for the excess portion; and aggregating the teacher over-limit conflict marker, major conflict marker, and capacity conflict marker to construct a group conflict identifier.

5. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 4, characterized in that, The steps of optimizing the class group set based on the group conflict identifier, allocating teacher tasks, and generating task implementation plans include: parsing the teacher over-limit conflict marker in the group conflict identifier, identifying the associated teaching classes and teacher identifiers; performing teacher reassignment operations on teaching classes with teacher over-limits based on the maximum number of classes a teacher can teach in the teacher configuration dataset, selecting alternative teachers whose number of classes they can teach is below the maximum from the teacher configuration dataset, and generating an updated class group set; for the professional conflict marker in the group conflict identifier, retrieving candidate teachers from the teacher's teaching history data whose professional identifier matches the current teaching class with a matching degree lower than a preset matching degree threshold, and reassigning the candidate teachers to the corresponding teaching classes; verifying the capacity conflict marker in the group conflict identifier, adjusting the class capacity or splitting the classes for teaching classes whose actual number of students exceeds the dynamic adjustment capacity limit based on the expansion flag in the parameter configuration dataset, and updating the class group set; based on the optimized class group set, traversing each teaching class and assigning teacher identifiers to generate a teacher task allocation plan; and integrating the class group set and the teacher task allocation plan to construct a task implementation plan.

6. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 5, characterized in that, The steps for reassigning candidate teachers to corresponding classes based on the professional conflict markers in the group conflict identifiers include: parsing the professional conflict markers to extract the professional identifiers and course qualification requirement codes of the current class; traversing the teacher configuration dataset to construct a teaching ability tag set for each teacher, including historical teaching professional identifiers and subject coverage codes of teacher qualification certificates; selecting teachers whose subject coverage codes of teacher qualification certificates contain the course qualification requirement codes as candidate teachers; generating a matching degree based on the correlation between historical teaching professional identifiers and the professional identifiers of the current class, and selecting candidate teacher identifiers with matching degrees lower than a dynamic matching degree threshold; and reassigning candidate teachers to corresponding classes based on the maximum number of classes a teacher can teach and the matching degree ranking.

7. A method for implementing academic affairs tasks based on automatic class splitting and merging rules according to any one of claims 1 to 6, characterized in that, After optimizing the class group set based on the group conflict identifier, assigning teacher tasks, and generating a task implementation plan, the process further includes: collecting scheduling conflict records from the academic affairs system after executing the task implementation plan, and extracting the scheduling conflict rate as a first feedback indicator; obtaining the dynamic adjustment capacity limit from the parameter configuration dataset, and calculating the class capacity utilization rate as a second feedback indicator based on the actual number of students in the class; analyzing the correlation between the first and second feedback indicators; when the scheduling conflict rate exceeds a preset conflict threshold and the class capacity utilization rate is lower than a preset utilization rate threshold, generating a teacher configuration optimization instruction, and recalculating the upper limit of the number of classes that a teacher can teach based on the teacher teaching quality evaluation data, and updating the teacher configuration dataset; when the class capacity utilization rate is consistently higher than the preset utilization rate upper limit threshold, generating a parameter configuration optimization instruction, dynamically increasing the teaching class capacity adjustment multiple according to the expansion flag, and updating the parameter configuration dataset; and inputting the updated teacher configuration dataset and parameter configuration dataset into the next round of task implementation process.

8. The method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 5, characterized in that, Following the step of generating the task implementation plan, the process further includes: obtaining the class grouping set, teacher task allocation plan, and site constraint information from the parameter configuration dataset in the task implementation plan; extracting historical scheduling data from the academic affairs system database, and constructing a conflict prediction model based on scheduling conflict records of similar course combinations in the same semester; inputting the task implementation plan into the conflict prediction model, and outputting a virtual scheduling conflict probability value and conflict type distribution; when the virtual scheduling conflict probability value exceeds a preset risk threshold, retrieving the available site set in the academic affairs system based on the site constraint information, and retrieving the available time slots of teachers to be assigned in the academic affairs system based on the teacher task allocation plan; performing preprocessing operations according to the conflict type distribution matching optimization strategy library to obtain a preprocessed task implementation plan and resource occupancy prediction report; and feeding the preprocessed task implementation plan and resource occupancy prediction report back to the parameter configuration dataset update process to update the expansion flag and campus distribution information.

9. A method for implementing academic affairs tasks based on automatic class splitting and merging rules according to claim 8, characterized in that, The steps for performing preprocessing operations based on the conflict type distribution matching optimization strategy library include: for teacher time conflict types, reallocating teaching time slots by combining the teacher identifier in the teacher task allocation scheme and the available time slots of the teachers to be assigned; for venue conflict types, dynamically replacing venue identifiers by combining the venue constraint information and the set of available venues; and for equipment demand conflict types, adjusting the priority of practical teaching class groups based on the practical teaching class identifier in the class grouping set.

10. A system for implementing academic affairs tasks based on automatic class splitting and merging rules, characterized in that, The system includes: a teacher qualification data acquisition module, used to acquire teacher qualification data, including the teachers for each course and the number of classes each teacher can teach; a teacher data configuration module, used to dynamically adjust the upper limit of the number of classes each teacher can teach for each course based on input teacher teaching quality evaluation data, generating a teacher configuration dataset; a teaching parameter configuration module, used to acquire course teaching demand data for a specified academic year and semester, including teaching class capacity, number of teaching classes, expansion flags, and campus distribution information, and dynamically adjust the teaching class capacity to an integer multiple of a preset number based on the expansion flags, generating a parameter configuration dataset; and an automatic class splitting and merging rule construction module, used to construct rules based on the teacher configuration data. The system comprises: a dataset of parameter configurations; a set of rules for automatically splitting and merging classes; a clustering analysis module; a grouping conflict detection module; a task assignment module; and a task implementation module. The system includes a campus isolation constraint rule, a major consistency constraint rule, and a student enrollment balance allocation constraint rule. The system also includes a clustering analysis module to acquire student class data, apply the automatic class splitting and merging rule set to perform clustering analysis on the student class data, and generate class group sets.

Images