Multi-task intelligent scheduling management platform

Abstract

The invention relates to the technical field of task scheduling management, and particularly discloses a multi-task intelligent scheduling management platform which analyzes each data sample and a to-be-processed task in a historical task data set by adopting a data processing and natural language processing algorithm based on artificial intelligence and deep learning. In this way, the embedded semantic features of the to-be-processed task and the aggregated coding features among the semantics of the multiple data samples are captured, and then the task priority is allocated to the to-be-processed task through the task semantic matching technology, so that efficient task scheduling and resource allocation are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of task scheduling management, and more specifically, to a multi-task intelligent scheduling management platform. Background Art

[0002] With the rapid development of e-commerce and retail industries, logistics and delivery efficiency has become one of the key factors affecting customer satisfaction. Especially during peak periods, such as promotional events, logistics companies are faced with the pressure of processing massive orders. How to efficiently dispatch resources, optimize delivery routes, and increase task processing priorities have become urgent issues that logistics companies need to solve.

[0003] Traditional dispatch management methods mainly rely on manual experience or simple rule engines, which are difficult to cope with complex multi-task scenarios. For example, faced with multiple orders, different delivery destinations, time requirements, and resource constraints (such as vehicles, personnel, etc.), traditional methods often cannot achieve the global optimal resource allocation, resulting in low delivery efficiency, increased costs, and even affecting customer experience.

[0004] Therefore, a multi-task intelligent scheduling management platform is desired. Summary of the invention

[0005] The present application provides a multi-task intelligent scheduling management platform, which can achieve efficient task scheduling and resource allocation.

[0006] According to one aspect of the present application, a multi-task intelligent scheduling management platform is provided, including: a pending task extraction module, which is used to extract pending tasks from a task queue; a historical task data set extraction module, which is used to extract a historical task data set from a background database, wherein each data sample in the historical task data set includes a task content and a task priority; a data set splitting module, which is used to perform data set splitting on the historical task data set based on the task priority to obtain a plurality of historical task data subsets; a historical task data semantic aggregation coding module, which is used to perform semantic embedding coding and semantic aggregation analysis on each data sample in each historical task data subset to obtain a plurality of historical task data subset semantic aggregation coding vectors; a pending task semantic embedding coding module, which is used to perform semantic embedding coding on the pending task to obtain a pending task semantic embedding coding vector; a task semantic matching calculation module, which is used to respectively calculate the task semantic matching degree between the pending task semantic embedding coding vector and each of the multiple historical task data subset semantic aggregation coding vectors to obtain a plurality of task semantic matching degrees; a task priority designation module, which is used to designate the task priority of the pending task based on the task priority corresponding to the largest of the multiple task semantic matching degrees.

[0007] In the above-mentioned multi-task intelligent scheduling management platform, the historical task data semantic aggregation coding module includes: a historical task semantic embedding coding unit, which is used to perform semantic embedding coding on each data sample in each historical task data subset to obtain a subset of historical task semantic embedding coding vectors; a feature aggregation analysis unit, which is used to perform feature aggregation analysis based on information core convergence on the subset of the historical task semantic embedding coding vectors to obtain the semantic aggregation coding vector of the historical task data subset.

[0008] In the above-mentioned multi-task intelligent scheduling management platform, the feature aggregation analysis unit includes: a historical task semantic coarse-grained aggregation coding subunit, which is used to input a subset of the historical task semantic embedded coding vector into an information core coarse-grained aggregation network to obtain a historical task semantic coarse-grained aggregation coding vector; a core aggregation compensation weight factor determination subunit, which is used to determine the core aggregation compensation weight of each historical task semantic embedded coding vector in the subset of the historical task semantic embedded coding vector based on the historical task semantic coarse-grained aggregation coding vector to obtain a set of core aggregation compensation weight factors; a dynamic compensation aggregation analysis subunit, which is used to perform dynamic compensation aggregation analysis based on node fine-grained features on the historical task semantic coarse-grained aggregation coding vector and the subset of the historical task semantic embedded coding vector based on the set of core aggregation compensation weight factors to obtain a historical task semantic fine-grained compensation aggregation coding vector; a historical task data aggregation subunit, which is used to input the historical task semantic fine-grained compensation aggregation coding vector and the historical task semantic coarse-grained aggregation coding vector into a residual unit to obtain the historical task data subset semantic aggregation coding vector.

[0009] In the above-mentioned multi-task intelligent scheduling management platform, the core convergence compensation weight factor determination subunit includes: a core convergence compensation factor calculation secondary subunit, which is used to calculate the core convergence compensation factor of each historical task semantic embedding coding vector in a subset of the historical task semantic embedding coding vector relative to the historical task semantic coarse-grained convergence coding vector to obtain a set of core convergence compensation factors; a compensation explicit modeling processing secondary subunit, which is used to perform compensation explicit modeling based on a gating function on the set of core convergence compensation factors to obtain the set of core convergence compensation weight factors.

[0010] In the above-mentioned multi-task intelligent scheduling management platform, the core convergence compensation factor calculation secondary sub-unit includes: a vector modulation tertiary sub-unit, which is used to modulate the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector to obtain the historical task semantic embedding coding modulation vector and the historical task semantic coarse-grained convergence coding modulation vector; a compensation information calculation tertiary sub-unit, which is used to calculate the compensation information between the historical task semantic embedding coding modulation vector and the historical task semantic coarse-grained convergence coding modulation vector to obtain the historical task semantic core convergence compensation weight vector; a historical task semantic deviation compensation factor calculation tertiary sub-unit, which is used to calculate the historical task semantic deviation compensation factor based on the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector; a core convergence compensation factor calculation tertiary sub-unit, which is used to calculate the core convergence compensation factor based on the historical task semantic core convergence compensation weight vector and the deviation compensation factor.

[0011] In the above-mentioned multi-task intelligent scheduling management platform, the three-level sub-unit for calculating the historical task semantic deviation compensation factor is used to: in response to the binary norm of the historical task semantic embedded coding vector being less than the binary norm of the historical task semantic coarse-grained converged coding vector, calculate the binary norm of the historical task semantic embedded coding vector divided by the binary norm of the historical task semantic coarse-grained converged coding vector and then add it to the constant one, and calculate the logarithmic function value with base 2 of the numerical value; in response to the binary norm of the historical task semantic embedded coding vector being greater than or equal to the binary norm of the historical task semantic coarse-grained converged coding vector, calculate the binary norm of the historical task semantic embedded coding vector divided by the binary norm of the historical task semantic coarse-grained converged coding vector.

[0012] In the above-mentioned multi-task intelligent scheduling management platform, the compensation explicit modeling processing secondary sub-unit is used to: in response to the core convergence compensation factor in the set of core convergence compensation factors being greater than or equal to a predetermined threshold, input the core convergence compensation factor into the sigmoid function; in response to the core convergence compensation factor in the set of core convergence compensation factors being less than a predetermined threshold, set the core convergence compensation factor to 0.

[0013] In the above-mentioned multi-task intelligent scheduling management platform, the dynamic compensation convergence analysis subunit is used to: input the set of the core convergence compensation weight factors, the historical task semantic coarse-grained convergence coding vector and a subset of the historical task semantic embedded coding vector into the node fine-grained dynamic compensation convergence network to obtain the historical task semantic fine-grained compensation convergence coding vector.

[0014] In the above-mentioned multi-task intelligent scheduling management platform, the set of core convergence compensation weight factors, the historical task semantic coarse-grained convergence coding vector and the subset of the historical task semantic embedded coding vector are input into the node fine-grained dynamic compensation convergence network to obtain the historical task semantic fine-grained compensation convergence coding vector, including: calculating the positional difference between each historical task semantic embedded coding vector in the subset of the historical task semantic embedded coding vector and the historical task semantic coarse-grained convergence coding vector respectively to obtain a set of historical task semantic convergence deviation vectors; using the set of core convergence compensation weight factors as weights, calculating the weighted sum between each historical task semantic convergence deviation vector in the set of historical task semantic convergence deviation vectors to obtain the historical task semantic fine-grained compensation convergence coding vector.

[0015] In the above-mentioned multi-task intelligent scheduling management platform, the task semantic matching calculation module is used to: input the semantic embedding coding vector of the task to be processed and each of the historical task data subset semantic aggregation coding vectors in the multiple historical task data subset semantic aggregation coding vectors into the task semantic matching network respectively to obtain the semantic matching degrees of the multiple tasks.

[0016] The present application provides a multi-task intelligent scheduling management platform, which analyzes each data sample in the historical task data set and the tasks to be processed by adopting data processing and natural language processing algorithms based on artificial intelligence and deep learning, so as to capture the embedded semantic features of the tasks to be processed and the aggregated coding features between the semantics of multiple data samples, and then assigns task priorities to the tasks to be processed through task semantic matching technology, thereby helping to achieve efficient task scheduling and resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0018] Figure 1 This is a schematic block diagram of a multi-task intelligent scheduling management platform according to an embodiment of the present application.

[0019] Figure 2 This is a data flow diagram of the multi-task intelligent scheduling management platform of an embodiment of the present application.

[0020] Figure 3 This is a schematic block diagram of a historical task data semantic aggregation encoding module in a multi-task intelligent scheduling management platform according to an embodiment of the present application.

[0021] Figure 4This is a schematic block diagram of a feature aggregation analysis unit in a multi-task intelligent scheduling management platform according to an embodiment of the present application.

[0022] Figure 5 This is a schematic block diagram of a subunit for determining a core convergence compensation weight factor in a multi-task intelligent scheduling management platform according to an embodiment of the present application.

[0023] Figure 6 It is a schematic block diagram of a secondary subunit for calculating a core convergence compensation factor in a multi-task intelligent scheduling management platform according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0025] In recent years, with the rapid development of artificial intelligence and machine learning technologies, intelligent scheduling systems have gradually become an effective tool to solve this problem. By introducing technologies such as natural language processing, semantic analysis, and deep learning, intelligent scheduling systems can learn task priority patterns from historical task data and make dynamic scheduling decisions based on real-time task characteristics. This data-driven scheduling method can not only improve the efficiency of task processing, but also significantly reduce the cost of manual intervention.

[0026] Specifically, the technical concept of the present application is to analyze each data sample in the historical task data set and the task to be processed by adopting data processing and natural language processing algorithms based on artificial intelligence and deep learning, so as to capture the embedded semantic features of the task to be processed and the aggregated coding features between the semantics of multiple data samples, and then assign task priorities to the tasks to be processed through task semantic matching technology, thereby helping to achieve efficient task scheduling and resource allocation.

[0027] More specifically, if Figure 1 and Figure 2As shown, the multi-task intelligent scheduling management platform 1 includes: a pending task extraction module 10, which is used to extract pending tasks from the task queue; a historical task data set extraction module 20, which is used to extract a historical task data set from a background database, and each data sample in the historical task data set includes a task content and a task priority; a data set splitting module 30, which is used to split the historical task data set based on the task priority to obtain multiple historical task data subsets; a historical task data semantic aggregation coding module 40, which is used to perform semantic embedding coding and semantic aggregation analysis on each data sample in each historical task data subset to obtain multiple A historical task data subset semantic aggregation coding vector; a pending task semantic embedding coding module 50, used to perform semantic embedding coding on the pending task to obtain a pending task semantic embedding coding vector; a task semantic matching calculation module 60, used to respectively calculate the task semantic matching degree between the pending task semantic embedding coding vector and each of the multiple historical task data subset semantic aggregation coding vectors to obtain multiple task semantic matching degrees; a task priority designation module 70, used to designate the task priority of the pending task based on the task priority corresponding to the largest of the multiple task semantic matching degrees.

[0028] Exemplarily, in the pending task extraction module 10, pending tasks are extracted from the task queue. It should be understood that the purpose of extracting pending tasks from the task queue is to ensure that all tasks can be identified, analyzed and reasonably allocated resources for processing according to priority by the system. This process not only helps to improve the efficiency of task processing, but also ensures the optimal task scheduling and resource allocation under limited resources. Specifically, in the e-commerce and retail industries, especially during promotional activities, companies may face the pressure of processing massive orders. These tasks are not only huge in number, but also have different urgency, time requirements and resource constraints (such as vehicles, personnel, etc.). If these tasks are not managed and scheduled in a systematic way, it is easy to cause problems such as waste of resources, delivery delays, etc., thereby affecting customer experience and service quality. Therefore, first extract the pending tasks from the task queue.

[0029] Exemplarily, in the historical task data set extraction module 20, the historical task data set is extracted from the background database, and each data sample in the historical task data set contains the task content and the task priority. It should be understood that in a complex business environment, such as e-commerce or logistics distribution industry, a large number of task requests are generated every day. These tasks are not only numerous, but also have different urgency, time requirements and resource requirements. If the priority and resource allocation of tasks are determined only by manual experience or simple rules, it is often difficult to cope with such complex and changeable situations, which may easily lead to resource waste or task delays. Therefore, by extracting the historical task data set from the background database, the system can analyze based on real historical records to discover potential patterns and laws. For example, certain specific types of tasks may always be given higher priorities because they are usually associated with high-value customers or urgent needs; while other tasks can be flexibly arranged according to their completion time windows. In this way, by learning from historical data, the system can make more intelligent decisions when facing new tasks, ensure that resources are optimally utilized, and improve overall service efficiency and quality.

[0030] Exemplarily, in the data set splitting module 30, the historical task data set is split based on the task priority to obtain multiple historical task data subsets. It should be understood that in the intelligent scheduling management platform, the historical task data set usually contains a large number of task samples, each sample has different task content and priority labels. In order to process new tasks more efficiently and assign priorities to them, the platform needs to learn the pattern of task priority from the historical task data. Based on this, in the technical solution of the present application, first, the historical task data set is split based on the task priority to obtain multiple historical task data subsets, and each data sample in each historical task data subset is semantically embedded and encoded and semantically aggregated to obtain multiple historical task data subset semantic aggregation encoding vectors. That is to say, the historical task data set is divided into multiple historical task data subsets according to priority labels (such as high, medium, and low), wherein each subset contains task samples of the same priority. This splitting method can reduce data complexity, so that subsequent semantic analysis can be performed separately for each priority category, thereby more accurately capturing the semantic features of historical task data of different priority categories.

[0031] Exemplarily, in the historical task data semantic aggregation coding module 40, semantic embedding coding and semantic aggregation analysis are performed on each data sample in each historical task data subset to obtain multiple historical task data subset semantic aggregation coding vectors. It should be understood that semantic embedding coding and semantic aggregation analysis are performed on each data sample in each historical task data subset, so as to embed and map each data sample in the historical task data subsets of different priorities into a common semantic space and convert them into high-dimensional vector representations, and then capture the aggregation features and common representations between the embedded semantics of each historical task data sample in the historical task data subset of each priority, so as to provide a basis for the subsequent priority allocation of tasks to be processed.

[0032] In one embodiment, Figure 3 As shown, the historical task data semantic aggregation coding module 40 includes: a historical task semantic embedding coding unit 41, which is used to perform semantic embedding coding on each data sample in each historical task data subset to obtain a subset of historical task semantic embedding coding vectors; a feature aggregation analysis unit 42, which is used to perform feature aggregation analysis based on information core convergence on the subset of the historical task semantic embedding coding vectors to obtain the semantic aggregation coding vector of the historical task data subset.

[0033] Specifically, in the historical task semantic embedding coding unit 41, for each priority historical task data subset, first, each data sample in each historical task data subset is semantically embedded and encoded to obtain a subset of the historical task semantic embedding coding vector. That is to say, each data sample in each historical task data subset is semantically embedded and encoded to convert the task content into a high-dimensional vector representation. For example, the delivery address is mapped to different positions in the high-dimensional vector space using the BERT or Word2Vec model, and the product information is mapped to another high-dimensional vector. The semantic embedding vector of each task data sample captures the embedded semantic information of its task content, providing a basis for subsequent aggregation analysis and priority matching and designation of tasks to be processed.

[0034] Specifically, in the feature aggregation analysis unit 42, in order to capture the common representation between the semantics of each historical task data in each task priority set, it is necessary to perform aggregation analysis on each historical task data in each task priority set, but directly performing global aggregation (such as mean pooling) on ​​a subset of the historical task semantic embedding vector will lose the personalized information of the key nodes. For example, a high-priority task may require special processing due to a special delivery address (such as a remote area), but a simple global average will dilute this feature. Based on this, in the technical solution of the present application, a subset of the historical task semantic embedding coding vector is further subjected to feature aggregation analysis based on information core convergence to obtain the historical task data subset semantic aggregation coding vector. In particular, the processing core of the feature aggregation analysis based on information core convergence is staged processing, from coarse-grained to fine-grained, combining global and local historical task data semantic association features. That is to say, global features are constructed through coarse-grained convergence, and then the core convergence compensation factor and the fine-grained dynamic compensation mechanism are used to retain local details, and finally a global and local historical task data subset semantic aggregation feature representation is generated. In this way, the balance problem of global and local features of the semantic association of each historical task data in each priority set is solved, and the advantage is multi-level feature integration and dynamic adjustment. It not only effectively captures the statistical representation of global features, but also meticulously enhances the individual feature significance of the semantics of each historical task data, thereby realizing the multi-level and multi-scale semantic feature integration of complex historical task data, providing a basis for the priority matching task of subsequent new tasks.

[0035] In one embodiment, Figure 4 As shown, the feature aggregation analysis unit 42 includes: a historical task semantic coarse-grained aggregation coding subunit 421, which is used to input a subset of the historical task semantic embedded coding vector into an information core coarse-grained aggregation network to obtain a historical task semantic coarse-grained aggregation coding vector; a core aggregation compensation weight factor determination subunit 422, which is used to determine the core aggregation compensation weight of each historical task semantic embedded coding vector in the subset of the historical task semantic embedded coding vector based on the historical task semantic coarse-grained aggregation coding vector to obtain a set of core aggregation compensation weight factors; a dynamic compensation aggregation analysis subunit 423, which is used to perform a dynamic compensation aggregation analysis based on node fine-grained features on the historical task semantic coarse-grained aggregation coding vector and the subset of the historical task semantic embedded coding vector based on the set of core aggregation compensation weight factors to obtain a historical task semantic fine-grained compensation aggregation coding vector; a historical task data aggregation subunit 424, which is used to input the historical task semantic fine-grained compensation aggregation coding vector and the historical task semantic coarse-grained aggregation coding vector into a residual unit to obtain the historical task data subset semantic aggregation coding vector.

[0036] Specifically, the calculation process of the historical task semantic coarse-grained aggregation encoding subunit 421 can be expressed as follows:

[0037] in, is a subset of the historical task semantic embedding encoding vectors, are the first, second, and third subsets of the historical task semantic embedding encoding vectors. and The historical task semantic embedding encoding vector, and It means taking the maximum and minimum values ​​in a vector. represents the global feature information nuclear factor, for function, is the global feature information kernel weight, is the number of vectors in the subset of the historical task semantic embedding encoding vectors, Coarse-grained aggregation of encoding vectors for historical task semantics.

[0038] It should be understandable that the semantic embedding encoding vectors of historical tasks in these subsets are input into the information core coarse-grained aggregation network. The role of this network is to generate a coarse-grained aggregation encoding vector representing the overall characteristics of the subset by compressing the global features of all historical tasks in each subset. Specifically, the information core coarse-grained aggregation network captures the complex similarity structure between nodes (i.e., the semantic embedding encoding vector of each historical task) through a nonlinear mapping mechanism, thereby forming a highly condensed global feature representation. This step is equivalent to providing an "overview map" for each priority category, so that the overall characteristics of a certain type of historical tasks can be quickly identified at the macro level. In this process, the concept of information core plays a key role. It not only helps to establish a node relationship model in a high-dimensional space, but also effectively captures the similarities and differences between different historical tasks. For example, in the high-priority historical task subset, certain specific types of historical tasks (such as emergency delivery for VIP customers) may always occupy an important position, and this pattern can be accurately identified by the information core coarse-grained aggregation network. In this way, we can better understand which factors determine the high priority of a certain historical task and formulate corresponding scheduling strategies accordingly.

[0039] In one embodiment, Figure 5As shown, the core convergence compensation weight factor determination subunit 422 includes: a core convergence compensation factor calculation secondary subunit 422-1, which is used to calculate the core convergence compensation factor of each historical task semantic embedding coding vector in the subset of the historical task semantic embedding coding vector relative to the historical task semantic coarse-grained convergence coding vector to obtain a set of core convergence compensation factors; a compensation explicit modeling processing secondary subunit 422-2, which is used to perform compensation explicit modeling based on a gating function on the set of core convergence compensation factors to obtain the set of core convergence compensation weight factors, In one embodiment, Figure 6 As shown, the core convergence compensation factor calculation secondary subunit 422-1 includes: a vector modulation tertiary subunit 422-11, which is used to modulate the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector to obtain a historical task semantic embedding coding modulation vector and a historical task semantic coarse-grained convergence coding modulation vector; a compensation information calculation tertiary subunit 422-12, which is used to calculate the compensation information between the historical task semantic embedding coding modulation vector and the historical task semantic coarse-grained convergence coding modulation vector to obtain a historical task semantic core convergence compensation weight vector; a historical task semantic deviation compensation factor calculation tertiary subunit 422-13, which is used to calculate the historical task semantic deviation compensation factor based on the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector; a core convergence compensation factor calculation tertiary subunit 422-14, which is used to calculate the core convergence compensation factor based on the historical task semantic core convergence compensation weight vector and the deviation compensation factor.

[0040] It should be understood that relying solely on the coarse-grained converged encoding vector is not enough to fully describe the unique characteristics of each task. In order to retain the personalized information of each task, a core convergence compensation mechanism is introduced. Specifically, based on the coarse-grained converged encoding vector of the historical task semantics, the deviation between the semantic embedding encoding vector of each historical task and the coarse-grained converged encoding vector is calculated to determine the core convergence compensation weight of each task. This process can be regarded as a correction rule for the global features, which aims to ensure that the unique information of each task is not lost due to the compression of the global features. In this process, by measuring the difference between the feature vector of each node and the coarse-grained converged encoding vector, a compensation factor for each task is dynamically generated. These compensation factors are essentially a quantitative description of the degree of deviation of each task from the overall pattern. For example, in the high-priority task subset, some specific types of tasks (such as emergency delivery for VIP customers) may be significantly different from most tasks, and these differences can be accurately captured by the compensation factors. In this way, even when faced with a large number of similar tasks, those tasks that need special attention can be identified and given corresponding resource tilt. Next, these compensation factors are used to construct a set of core convergence compensation weight factors. This set provides a personalized adjustment coefficient for each task, enabling more accurate decisions in subsequent task scheduling and resource allocation. In this way, not only can the overall pattern be maintained, but also the unique needs of each task can be flexibly responded to. This dynamic adjustment mechanism based on the core convergence compensation weight is similar to the attention mechanism, which allows the coordinated expression of features at the global and local levels. For example, when processing high-priority tasks, resources can be flexibly allocated according to the compensation weight factor to ensure that key tasks are processed in a timely manner, while the scheduling order of other tasks is reasonably arranged. In this way, not only can the overall service efficiency be improved, but also resource waste and delays can be effectively avoided. In addition, the application of the core convergence compensation weight factor also enhances the robustness and adaptability of the model. In actual operation, the demand and priority of tasks may fluctuate over time. By introducing a compensation mechanism, these changes can be responded to more quickly and adjusted accordingly.

[0041] Specifically, the calculation process of the vector modulation three-level subunit 422-11 can be expressed as follows:

[0042] in, It is the point convolution layer processing. is the first weight matrix, is the second weight matrix, for function, The first one in the set of historical task semantic embedding coding modulation vectors The historical task semantic embedding coded modulation vector, Coarse-grained aggregation of coded modulation vectors for historical task semantics.

[0043] Specifically, the calculation process of the compensation information calculation third-level subunit 422-12 can be expressed as follows:

[0044] in, For positional subtraction, To take the absolute value, Aggregate compensation weight vector for historical task semantic kernel.

[0045] In one embodiment, the historical task semantic deviation compensation factor calculation three-level sub-unit is used to: in response to the bi-norm of the historical task semantic embedding coding vector being less than the bi-norm of the historical task semantic coarse-grained converged coding vector, calculate the bi-norm of the historical task semantic embedding coding vector divided by the bi-norm of the historical task semantic coarse-grained converged coding vector and then add it to the constant one, and calculate the logarithmic function value with base 2 of the numerical value; in response to the bi-norm of the historical task semantic embedding coding vector being greater than or equal to the bi-norm of the historical task semantic coarse-grained converged coding vector, calculate the bi-norm of the historical task semantic embedding coding vector divided by the bi-norm of the historical task semantic coarse-grained converged coding vector. Specifically, the process can be expressed by the formula:

[0046] in, represents the two-norm of a vector, is the logarithmic function value with base 2, Represents the historical task semantic deviation compensation factor.

[0047] In particular, here, for the deviation compensation factor between the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector, the performance deviation of the core convergence strategy as a scenario strategy can be measured by quantifying the regret metric based on the information core compression hypothesis in the core convergence decision process, that is, the game counterfactual regret value. Specifically, the counterfactual regret value is provided with a normalized decision point loss description based on the policy action, that is, the vector norm representation of the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector through the vector norm representation, and then, for the possible vector distribution action game scenario differences, the compensation rule of node personalized information is corrected by the degree of information distribution of the regret value and the relative distribution amplitude of the regret value, so that the node personalized information is considered as the untaken action in the decision, and the biased compensation is performed based on the potential benefits of the information core convergence hypothesis.

[0048] Specifically, the calculation process of the core convergence compensation factor calculation third-level subunit 422-14 can be expressed as follows:

[0049] in, and denote the compensation weight matrix and compensation bias vector respectively, To compensate the modulation vector, for The corresponding nuclear convergence compensation factor.

[0050] In one embodiment, the compensation explicit modeling processing secondary subunit 422-2 is used to: in response to the core convergence compensation factor in the set of core convergence compensation factors being greater than or equal to a predetermined threshold, input the core convergence compensation factor into a sigmoid function; in response to the core convergence compensation factor in the set of core convergence compensation factors being less than a predetermined threshold, set the core convergence compensation factor to 0. Specifically, the process can be expressed as follows:

[0051] in, is the preset threshold, To compensate for the explicit modeling operation, for The corresponding kernel convergence compensation weight factor.

[0052] Specifically, these kernel convergence compensation factors are further optimized, namely, explicit modeling of compensation based on the gating function. In this process, the role of the gating function is to dynamically select and regulate information, ensuring that only those local features that contribute significantly to the overall model are highlighted, while irrelevant or redundant information is appropriately suppressed. In this way, the coordinated expression of features can be achieved at the global and local levels, ensuring that the unique needs of each task are fully considered. Specifically, the gating function explicitly models the compensation factors, screens out those compensation factors that have a significant impact on the final encoding results, and assigns them appropriate weights. This step not only improves the compactness and significance of feature aggregation, but also enhances the model's ability to capture complex relationships. For example, in the high-priority task subset, certain specific types of tasks (such as emergency delivery for VIP customers) may have unique attributes that can be accurately identified and emphasized through the gating function. In this way, even when faced with a large number of similar tasks, those tasks that require special attention can be identified and given corresponding resource tilts.

[0053] In one embodiment, the dynamic compensation convergence analysis subunit 423 is used to: input the set of the core convergence compensation weight factors, the historical task semantic coarse-grained converged coding vector and the subset of the historical task semantic embedded coding vector into the node fine-grained dynamic compensation convergence network to obtain the historical task semantic fine-grained compensation converged coding vector. Specifically, the process can be expressed as follows:

[0054] in, Aggregating encoding vectors for semantic fine-grained compensation for historical tasks.

[0055] It should be understood that in the process of dynamically compensating and converging the historical task semantics coarse-grained converged encoding vector and the subset of the historical task semantics embedded encoding vector based on the node fine-grained features based on the set of these kernel convergence compensation weight factors, the historical task semantics fine-grained compensated converged encoding vector is generated by dynamically adjusting each node feature vector. The core of this process is to introduce high fidelity so that the compensated features more accurately describe the local details of the node while adapting to the dynamic changes of the global feature constraints. Specifically, the positional difference between each historical task semantic embedded encoding vector and its corresponding coarse-grained converged encoding vector is calculated to form a set of historical task semantics convergence deviation vectors. Then, the weighted sum between these deviation vectors is calculated using the kernel convergence compensation weight factor as the weight to generate the historical task semantics fine-grained compensated converged encoding vector. This process is similar to the attention mechanism, allowing for coordinated expression of features at the global and local levels. For example, in the high-priority task subset, certain specific types of tasks (such as emergency delivery for VIP customers) may have unique attributes that can be accurately captured by fine-grained compensated convergence analysis. Even when faced with a large number of similar tasks, we can identify those that require special attention and allocate resources accordingly, thus maintaining an understanding of the overall pattern while being able to flexibly respond to the unique needs of each task.

[0056] In one embodiment, the set of core convergence compensation weight factors, the historical task semantic coarse-grained convergence coding vector and a subset of the historical task semantic embedded coding vector are input into a node fine-grained dynamic compensation convergence network to obtain a historical task semantic fine-grained compensation convergence coding vector, including: calculating the positional difference between each historical task semantic embedded coding vector in the subset of the historical task semantic embedded coding vector and the historical task semantic coarse-grained convergence coding vector to obtain a set of historical task semantic convergence deviation vectors; using the set of core convergence compensation weight factors as weights, calculating the weighted sum between each historical task semantic convergence deviation vector in the set of historical task semantic convergence deviation vectors to obtain the historical task semantic fine-grained compensation convergence coding vector.

[0057] Specifically, the calculation process of the historical task data aggregation subunit 424 can be expressed as follows:

[0058] in, and is the residual weight coefficient, A semantically aggregated encoding vector for the historical task data subset.

[0059] It should be understood that the historical task semantic fine-grained compensation converged encoding vector and the historical task semantic coarse-grained converged encoding vector are input into the residual unit to generate the historical task data subset semantic aggregated encoding vector. The core of this process is to fuse global and local features to ensure that both the overall pattern can be captured and the unique attributes of each task can be carefully reflected. Specifically, the design of residual units is not only widely used in deep learning, but its significance here lies in an effective mechanism for information fusion. Through residual connections, the gradient vanishing problem can be avoided, so that the fine-grained dynamically adjusted features can not only inherit the global vision of the coarse-grained features, but also retain the local significant details. This design forms a refined feature expression that combines the global and local. In actual operation, suppose we have a high-priority task subset that contains a large number of urgent delivery tasks for VIP customers. Although these tasks have certain common characteristics (such as high priority), the specific requirements of each task may be different (such as a specific delivery time window or special delivery requirements). By inputting the historical task semantics fine-grained compensation converged encoding vector and the historical task semantics coarse-grained converged encoding vector into the residual unit, it is possible to maintain the understanding of the overall characteristics of high-priority tasks at the macro level, while accurately capturing the unique requirements of each task at the micro level.

[0060] Exemplarily, in the task semantic embedding coding module 50, the task to be processed is semantically embedded to obtain a task semantic embedding coding vector. That is, similarly, the task to be processed is semantically embedded to obtain a task semantic embedding coding vector. Through semantic embedding coding, the task to be processed can be embedded and mapped into a high-dimensional space, thereby being converted into a vector representation, and the embedded semantics of the task to be processed can be captured.

[0061] Exemplarily, in the task semantic matching calculation module 60, the task semantic matching degree between the semantic embedding coding vector of the task to be processed and each of the semantic aggregation coding vectors of the historical task data subsets in the multiple semantic aggregation coding vectors of the historical task data subsets is calculated respectively to obtain multiple task semantic matching degrees. It should be understood that the semantic embedding coding vector of the task to be processed represents the semantic information of the new task (such as delivery address, product information, etc.), which is converted into a high-dimensional vector through semantic embedding coding technology (such as BERT, Word2Vec). Each of the semantic aggregation coding vectors of the historical task data subsets in the multiple semantic aggregation coding vectors of the historical task data subsets represents the representative semantic features of each priority subset (such as the high priority subset may contain features such as "urgent delivery" and "VIP customers"), which are generated through semantic aggregation analysis. In order to determine which priority subset the new task best matches, in the technical solution of the present application, the task semantic matching calculation module is used to: input the semantic embedding coding vector of the task to be processed and each of the semantic aggregation coding vectors of the historical task data subsets in the multiple semantic aggregation coding vectors of the historical task data subsets into the task semantic matching network respectively to obtain the multiple task semantic matching degrees. In a specific embodiment, the task semantic matching network calculates the task semantic matching degree between the semantic embedding coding vector of the task to be processed and each of the historical task data subset semantic aggregation coding vectors in the multiple historical task data subset semantic aggregation coding vectors through an attention mechanism (Attention Mechanism) or cosine similarity (Cosine Similarity).

[0062] Exemplarily, in the task priority designation module 70, the task priority corresponding to the largest task semantic match among the multiple tasks is designated as the task priority of the task to be processed. It should be understood that the historical task data set contains a large number of tasks that have been successfully processed in the past and their corresponding priority labels. These historical data provide a valuable reference for the priority allocation of current tasks. By matching historical task patterns, it is possible to draw on past experience and make more scientific and reasonable decisions. At the same time, matching the task to be processed with the most similar subset of historical tasks and assigning corresponding priorities can ensure that resources are optimally allocated. For example, during a promotional event, the system can reasonably arrange delivery vehicles and personnel according to the urgency and special requirements of the task to avoid waste of resources and delays.

[0063] In a specific embodiment, the maximum value is first found from a plurality of task semantic matching degrees. Then, the corresponding priority is assigned to the task to be processed according to the priority of the historical task data subset corresponding to the maximum matching degree. For example, if the maximum matching degree appears in the high priority subset, the priority of the task to be processed is set to "high". In this way, the task semantic matching technology can be used to assign task priorities to the tasks to be processed, thereby facilitating efficient task scheduling and resource allocation.

[0064] In summary, according to the embodiment of the present application, a multi-task intelligent scheduling management platform is explained, which analyzes each data sample in the historical task data set and the task to be processed by adopting data processing and natural language processing algorithms based on artificial intelligence and deep learning, so as to capture the embedded semantic features of the task to be processed and the aggregated coding features between the semantics of multiple data samples, and then assigns task priorities to the tasks to be processed through task semantic matching technology, thereby helping to achieve efficient task scheduling and resource allocation.

[0065] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0066] It should be understood that the specific examples in this article are only intended to help those skilled in the art better understand the embodiments of the present application, rather than to limit the scope of the embodiments of the present application.

[0067] It should also be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0068] It should also be understood that the various implementation methods described in this specification can be implemented individually or in combination, and the embodiments of the present application are not limited to this.

[0069] Unless otherwise stated, all technical and scientific terms used in the embodiments of the present application are the same as the meanings generally understood by those skilled in the art of the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the scope of the present application. The term "and / or" used in the present application includes any and all combinations of one or more related listed items. The singular forms of "a kind of", "above" and "the" used in the embodiments of the present application and the appended claims are also intended to include majority forms, unless the context clearly indicates other meanings. In addition, the terms "first", "second" etc. are only used for description purposes and cannot be understood as indicating or suggesting relative importance.

[0070] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0072] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0073] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0074] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-task intelligent scheduling management platform, characterized in that: include: A pending task extraction module is used to extract pending tasks from the task queue; A historical task data set extraction module is used to extract a historical task data set from a background database, wherein each data sample in the historical task data set includes task content and task priority; A data set splitting module, used for splitting the historical task data set based on task priority to obtain multiple historical task data subsets; A historical task data semantic aggregation coding module is used to perform semantic embedding coding and semantic aggregation analysis on each data sample in each historical task data subset to obtain multiple historical task data subset semantic aggregation coding vectors; A semantic embedding coding module for tasks to be processed, used for performing semantic embedding coding on the tasks to be processed to obtain semantic embedding coding vectors for the tasks to be processed; A task semantic matching calculation module is used to respectively calculate the task semantic matching degree between the semantic embedding coding vector of the task to be processed and each historical task data subset semantic aggregation coding vector in the multiple historical task data subset semantic aggregation coding vectors to obtain multiple task semantic matching degrees; The task priority designation module is used to designate the task priority corresponding to the largest task among the plurality of task semantic matching degrees as the task priority of the task to be processed.

2. The multi-task intelligent scheduling management platform according to claim 1 is characterized in that: The historical task data semantic aggregation encoding module includes: A historical task semantic embedding coding unit, used for performing semantic embedding coding on each data sample in each historical task data subset to obtain a subset of historical task semantic embedding coding vectors; A feature aggregation analysis unit is used to perform feature aggregation analysis based on information core convergence on a subset of the historical task semantic embedded coding vector to obtain the historical task data subset semantic aggregation coding vector.

3. The multi-task intelligent scheduling management platform according to claim 2 is characterized in that: The feature aggregation analysis unit comprises: A historical task semantics coarse-grained convergence encoding subunit, used for inputting a subset of the historical task semantics embedding encoding vectors into an information core coarse-grained convergence network to obtain a historical task semantics coarse-grained convergence encoding vector; A core convergence compensation weight factor determination subunit is used to determine the core convergence compensation weight of each historical task semantic embedding coding vector in the subset of the historical task semantic embedding coding vector based on the historical task semantic coarse-grained convergence coding vector to obtain a set of core convergence compensation weight factors; A dynamic compensation convergence analysis subunit is used to perform a dynamic compensation convergence analysis based on node fine-grained features on a subset of the historical task semantics coarse-grained converged coding vector and the historical task semantics embedded coding vector based on the set of the core convergence compensation weight factors to obtain a historical task semantics fine-grained compensation convergence coding vector; The historical task data aggregation subunit is used to input the historical task semantic fine-grained compensation aggregation coding vector and the historical task semantic coarse-grained aggregation coding vector into the residual unit to obtain the historical task data subset semantic aggregation coding vector.

4. The multi-task intelligent scheduling management platform according to claim 3 is characterized in that: The core convergence compensation weight factor determination subunit includes: A secondary subunit for calculating a kernel convergence compensation factor is used to calculate a kernel convergence compensation factor of each historical task semantic embedding coding vector in a subset of the historical task semantic embedding coding vector relative to the historical task semantic coarse-grained convergence coding vector to obtain a set of kernel convergence compensation factors; The compensation explicit modeling processing secondary subunit is used to perform compensation explicit modeling based on a gating function on the set of core convergence compensation factors to obtain the set of core convergence compensation weight factors.

5. The multi-task intelligent scheduling management platform according to claim 4 is characterized in that: The core convergence compensation factor calculation secondary subunit includes: A vector modulation tertiary subunit, used for modulating the historical task semantic embedding coding vector and the historical task semantic coarse-grained aggregation coding vector to obtain a historical task semantic embedding coding modulation vector and a historical task semantic coarse-grained aggregation coding modulation vector; The compensation information calculation tertiary subunit is used to calculate the compensation information between the historical task semantic embedding coding modulation vector and the historical task semantic coarse-grained convergence coding modulation vector to obtain the historical task semantic core convergence compensation weight vector; A historical task semantic deviation compensation factor calculation tertiary subunit is used to calculate the historical task semantic deviation compensation factor based on the historical task semantic embedding coding vector and the historical task semantic coarse-grained convergence coding vector; The third-level sub-unit for calculating the kernel convergence compensation factor is used to calculate the kernel convergence compensation factor based on the historical task semantic kernel convergence compensation weight vector and the deviation compensation factor.

6. The multi-task intelligent scheduling management platform according to claim 5, characterized in that: The historical task semantic deviation compensation factor calculation three-level sub-unit is used to: In response to the binary norm of the historical task semantic embedding coding vector being less than the binary norm of the historical task semantic coarse-grained converged coding vector, the binary norm of the historical task semantic embedding coding vector is calculated and divided by the binary norm of the historical task semantic coarse-grained converged coding vector, and then added to a constant 1, and a logarithmic function value with base 2 of the numerical value is calculated; In response to the binary norm of the historical task semantic embedding coding vector being greater than or equal to the binary norm of the historical task semantic coarse-grained aggregation coding vector, the binary norm of the historical task semantic embedding coding vector is calculated by dividing the binary norm of the historical task semantic coarse-grained aggregation coding vector.

7. The multi-task intelligent scheduling management platform according to claim 6, characterized in that: The compensation explicit modeling processing secondary subunit is used to: in response to the core convergence compensation factor in the set of core convergence compensation factors being greater than or equal to a predetermined threshold, input the core convergence compensation factor into a sigmoid function; in response to the core convergence compensation factor in the set of core convergence compensation factors being less than a predetermined threshold, set the core convergence compensation factor to 0.

8. The multi-task intelligent scheduling management platform according to claim 7 is characterized in that: The dynamic compensation convergence analysis subunit is used to: input the set of core convergence compensation weight factors, the historical task semantic coarse-grained convergence coding vector and a subset of the historical task semantic embedded coding vector into the node fine-grained dynamic compensation convergence network to obtain the historical task semantic fine-grained compensation convergence coding vector.

9. The multi-task intelligent scheduling management platform according to claim 8, characterized in that: Inputting the set of the core convergence compensation weight factors, the historical task semantics coarse-grained converged coding vector and a subset of the historical task semantics embedded coding vector into a node fine-grained dynamic compensation convergence network to obtain a historical task semantics fine-grained compensation converged coding vector, including: Calculate the positional difference between each historical task semantic embedding coding vector in the subset of the historical task semantic embedding coding vector and the historical task semantic coarse-grained converged coding vector to obtain a set of historical task semantic converged deviation vectors; Taking the set of core convergence compensation weight factors as weights, the weighted sum of each historical task semantic convergence deviation vector in the set of historical task semantic convergence deviation vectors is calculated to obtain the historical task semantic fine-grained compensation convergence encoding vector.

10. The multi-task intelligent scheduling management platform according to claim 9, characterized in that: The task semantic matching calculation module is used to: input the semantic embedding coding vector of the task to be processed and each historical task data subset semantic aggregation coding vector in the multiple historical task data subset semantic aggregation coding vectors into the task semantic matching network respectively to obtain the multiple task semantic matching degrees.

Images