Scheme generation method and device, electronic equipment and storage medium

Abstract

The invention discloses a scheme generation method and device, electronic equipment and a storage medium. According to the specific embodiment, the method comprises the steps of determining a structured data set; performing semantic analysis on the structured data set through a semantic analysis model to obtain key information; utilizing the structured data set, the semantic analysis model and the key information to construct a comprehensive objective function; and solving the comprehensive objective function to obtain a service execution scheme set composed of at least one service execution scheme. Semantic analysis is conducted on the structured data set through the semantic analysis model, dynamic analysis of the structured data set is achieved so as to meet the requirement for flexibly processing the to-be-executed service, the service execution scheme is obtained by constructing and solving the comprehensive objective function, the to-be-executed service can be responded in time, planning efficiency is improved, and the human error rate is reduced.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a scheme generation method, apparatus, electronic device, and storage medium. Background Technology

[0002] In existing technologies, intelligent service planning methods mainly rely on manual analysis and static planning tools. Service systems need to meet diverse requirements, involving multiple aspects such as document interpretation, business requirement analysis, resource allocation, and user services. Existing planning methods typically involve manual data parsing followed by the formulation of a planning scheme based on the analysis results. However, the manual data parsing process is time-consuming, labor-intensive, and prone to human error, leading to low planning efficiency. Existing technologies also construct rule-based planning systems, which typically pre-set fixed business rules and resource allocation schemes. However, service requirements are usually complex and changing, and fixed schemes are difficult to flexibly address the needs of different scenarios.

[0003] Existing manual analysis and static planning tools are inefficient and unable to adapt to complex service requirements. Rule-based planning systems struggle to flexibly handle complex collaborative optimization requirements and lack dynamic adjustment capabilities. Therefore, existing technologies using manual analysis and static planning methods cannot efficiently process services and achieve real-time planning, resulting in planning solutions that cannot adapt to the ever-changing needs of government and enterprise services. Summary of the Invention

[0004] This invention provides a scheme generation method, apparatus, electronic device, and storage medium to realize the dynamic generation of schemes required for services, thereby improving the efficiency of service planning.

[0005] According to one aspect of the present invention, a scheme generation method is provided, comprising:

[0006] Determine a structured dataset, the structured dataset including at least one type of data related to the service to be executed;

[0007] The structured dataset is semantically parsed using a semantic parsing model to obtain key information. The key information includes information required to execute the service to be executed, including business rules and key requirements. The business rules include the business rules contained in the structured dataset, and the key requirements include the requirements contained in the structured dataset.

[0008] Using the structured dataset, the semantic parsing model, and the key information, a comprehensive objective function is constructed. The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements.

[0009] Solving the comprehensive objective function yields a set of service execution schemes consisting of at least one service execution scheme, wherein the service execution scheme is a scheme for executing the service to be executed.

[0010] According to another aspect of the present invention, a scheme generation apparatus is provided, comprising:

[0011] A determination module is used to determine a structured dataset, the structured dataset including at least one type of data related to the service to be executed;

[0012] The parsing module is used to perform semantic parsing on the structured dataset through a semantic parsing model to obtain key information, including information required to execute the service to be executed, including business rules and key requirements;

[0013] The construction module is used to construct a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information. The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements.

[0014] The solution module is used to solve the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme, wherein the service execution scheme is a scheme for executing the service to be executed.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the scheme generation method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the scheme generation method described in any embodiment of the present invention.

[0020] The technical solution of this invention involves: determining a structured dataset; performing semantic parsing on the structured dataset using a semantic parsing model to obtain key information; constructing a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information; and solving the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme. By performing semantic parsing on the structured dataset using a semantic parsing model, dynamic parsing of the structured dataset is achieved to meet the need for flexible handling of services to be executed. By constructing and solving the comprehensive objective function to obtain service execution schemes, timely responses to services to be executed can be achieved, improving planning efficiency and reducing human error rates.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a scheme generation method provided in Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of a method for constructing a comprehensive objective function according to Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of a scheme generation device according to Embodiment 3 of the present invention;

[0026] Figure 4 This is a block diagram of an electronic device provided according to Embodiment 4 of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1 This is a flowchart of a scheme generation method according to Embodiment 1 of the present invention. This embodiment is applicable to situations where a service execution scheme is determined. This method can be executed by a scheme generation device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0031] S110. Determine the structured dataset.

[0032] The structured dataset includes at least one type of data related to the service to be executed.

[0033] In this embodiment, a structured dataset can be understood as the data required to execute the service to be executed, and the data in the structured dataset is data that can be processed by the model.

[0034] Specifically, we can first obtain text information related to the service to be executed, and then perform structuring on this text information to obtain a structured dataset. The structured dataset can include data such as rules, guidelines, requirements, resource consumption, and progress status required to execute the service.

[0035] S120. Semantic parsing is performed on the structured dataset using a semantic parsing model to obtain key information.

[0036] The key information includes the information required to execute the service to be executed. The key information includes business rules and key requirements. The business rules include the business rules contained in the structured dataset, and the key requirements include the requirements contained in the structured dataset.

[0037] In this embodiment, the semantic parsing model can be understood as a model used to perform semantic parsing on a structured dataset. The parameters of the semantic parsing model can be adaptively adjusted, and the semantic parsing model can also be used to construct functions. Key information can be understood as the information contained in the structured dataset, which may be the information required to execute the service to be executed. Business rules may be the business rules contained in the structured dataset. Key requirements include the requirements contained in the structured dataset corresponding to the service to be executed.

[0038] Specifically, a semantic parsing model can be an adaptively adjustable model that performs semantic parsing on structured datasets to extract key information, including business rules and key requirements.

[0039] For example, through a semantic parsing model For structured datasets Perform natural language processing to extract key information : ,in, Key information to be extracted, including business rules and key needs T represents the total length of the input text indicated by the structured dataset. , , Let be the query vector, key vector, and value vector of the t-th word in the structured dataset, respectively. The dimension of the key vector can be used for normalization. This represents the weight adjustment coefficient at time step t. The softmax function is used to normalize key information.

[0040] S130. Using the structured dataset, the semantic parsing model, and the key information, construct a comprehensive objective function.

[0041] The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements.

[0042] In this embodiment, the comprehensive objective function can be understood as the objective function used to determine the scheme for executing the service to be executed. The global requirement optimization function can be understood as a function related to at least one global requirement of the service to be executed, and the global requirement optimization function can be determined by the business rules in the key information. The local requirement optimization function can be understood as a function related to at least one local requirement of the service to be executed, and the local requirement optimization function can be determined by the key requirements in the key information.

[0043] Specifically, firstly, based on the business rules and key requirements in the key information, a global requirement optimization function and a local requirement optimization function are constructed using a semantic parsing model. Next, based on the importance of global and local requirements in the overall objective function, the weights corresponding to each optimization function are determined; these weights can be set according to the scenario in which the service to be executed is located. Finally, the global and local requirement optimization functions, along with their respective weights, are weighted and summed to obtain the overall objective function.

[0044] S140. Solve the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme.

[0045] The service execution plan is a plan for executing the service to be executed.

[0046] In this embodiment, a service execution scheme can be understood as a scheme for executing a service to be executed. A service execution scheme set can be formed by multiple service execution schemes. A service execution scheme can be a scheme that achieves a balance in terms of resource allocation, service efficiency, user satisfaction, and resource consumption.

[0047] Specifically, constraints are set for the comprehensive objective function. These constraints may include conditions for resource consumption corresponding to global requirements and resource consumption corresponding to local requirements. Based on these constraints, the comprehensive objective function can be solved to obtain a set of service execution schemes consisting of at least one service execution scheme.

[0048] For example, service execution scheme ;in, For the i-th service execution plan, Let be the resource allocation matrix for the i-th service execution plan, representing the allocation of resources under different demands. Let be the service efficiency metric for the i-th service execution plan. Rate the user satisfaction of the i-th service implementation plan. This represents the total resource consumption of the i-th service execution plan.

[0049] The technical solution of this invention involves: determining a structured dataset; performing semantic parsing on the structured dataset using a semantic parsing model to obtain key information; constructing a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information; and solving the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme. By performing semantic parsing on the structured dataset using a semantic parsing model, dynamic parsing of the structured dataset is achieved to meet the need for flexible handling of services to be executed. By constructing and solving the comprehensive objective function to obtain service execution schemes, timely responses to services to be executed can be achieved, improving planning efficiency and reducing human error rates.

[0050] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0051] In one embodiment, determining the structured dataset includes:

[0052] Obtain multi-source data related to the service to be executed, including execution file data, business requirement data, resource consumption data, and project progress data required when executing the service to be executed;

[0053] Construct the unstructured dataset corresponding to the multi-source data;

[0054] The unstructured dataset is converted to a new format to obtain an initial structured dataset.

[0055] The initial structured dataset is preprocessed to obtain a structured dataset. The preprocessing includes word segmentation, data cleaning, and structuring.

[0056] In this embodiment, multi-source data can be understood as at least one type of text information related to the service to be executed. An unstructured dataset can be understood as a dataset formed by combining various text information. An initial structured dataset can be understood as a dataset formed after converting the data format of the unstructured dataset composed of text information; the initial structured dataset has not yet undergone preprocessing.

[0057] Specifically, execution document data may include issued guidance documents related to the execution of the service to be performed. Business requirement data may include specific requirements for the service to be performed, involving business process optimization, technical support, project collaboration, etc. Resource consumption data includes the pre-allocated resources when executing the service to be performed. Project progress data includes regular reports on the executed project, including project progress, key milestones, problems encountered, and resource consumption.

[0058] For example, the acquired executable file data Business requirement data Resource consumption data and project progress data Combined to form an unstructured dataset For unstructured datasets The data in the dataset undergoes data format conversion, transforming it into a unified structured dataset. For structured datasets Text segmentation is performed to break down natural language text into lexical units, which are denoted as the segmented structured dataset. ; for the structured dataset after word segmentation Data cleaning is performed to remove redundant information, invalid symbols, and erroneous data, resulting in a cleaned structured dataset. ; cleaned structured dataset Perform structured processing, based on the preset database format. The data in the dataset is classified and labeled to obtain a structured dataset. .

[0059] In one embodiment, before performing semantic parsing on the structured dataset using a semantic parsing model to obtain key information, the method further includes:

[0060] Determine the set of model parameters, which is related to the complexity of the structured dataset;

[0061] The parameters in the model parameter set are determined as the parameters of the semantic parsing model.

[0062] In this embodiment, the model parameter set can be understood as the set of parameters of the semantic parsing model, which are related to the complexity of the structured dataset.

[0063] For example, based on structured datasets The complexity, such as the quantification of parsing difficulty when parsing structured datasets, and the dynamic adjustment of the semantic parsing model. parameter. ;in, This represents the set of parameters for the semantic parsing model, including the number of model layers, the number of attention heads, and the learning rate. and These are the weighting coefficients, Representing structured datasets The weights within the overall architecture are allocated based on the criticality of the data source. Representing structured datasets The amount of data, Representing structured datasets The complexity, This represents the number of dimensions in the j-th dimension of the structured dataset. Let j be the size of the feature space of dimension j. It is a small constant that avoids the denominator being zero.

[0064] In one embodiment, solving the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme includes:

[0065] Solving the comprehensive objective function yields an initial service execution scheme set, which includes at least one initial service execution scheme.

[0066] For each initial service execution scheme in the set of initial service execution schemes, a frontier analysis is performed to select at least one target service execution scheme, and the at least one target service execution scheme forms a non-dominated solution set;

[0067] For each target service execution scheme in the non-dominated solution set, a comprehensive evaluation value of the target service execution scheme is calculated. When the comprehensive evaluation value is greater than a set threshold, the target service execution scheme is determined as a service execution scheme.

[0068] The set of all the service execution schemes is defined as the service execution scheme set.

[0069] In this embodiment, the initial service execution scheme set can be understood as a set composed of multiple initial service execution schemes, which can be the result of solving the comprehensive objective function. The target service execution scheme can be understood as the scheme selected from the initial service execution schemes through frontier analysis. The non-dominated solution set can be understood as the set storing the target service execution schemes. The comprehensive evaluation value can be understood as a numerical value used to indicate the performance of the target service execution scheme; the higher the comprehensive evaluation value, the better the performance of the target service execution scheme.

[0070] For example, a multi-objective evolutionary algorithm is used to evaluate the comprehensive objective function. The solution is used to generate the initial set of service execution schemes. For each initial service execution plan in the generated set of initial service execution plans. Perform frontier analysis to select non-dominated solution sets. The initial service execution scheme of the target in the non-dominated solution set The following conditions must be met: for any two initial service execution schemes and There is no initial service execution plan. Superior to service execution solutions across all objective dimensions Execute a scheme for each target service in the non-dominated solution set. Calculate the comprehensive evaluation function : ;in, The comprehensive evaluation value of the service execution plan for the j-th target, where G is the number of targets, a total of 4. Let g be the weight of the g-th objective. The normalization function of the execution plan serving the j-th target on the g-th target is calculated as follows: ;in, The actual value of the execution plan for serving the j-th objective on the g-th objective. and These are the maximum and minimum values ​​of the g-th objective in the non-dominated solution set, respectively. Based on the comprehensive evaluation value... The target service execution schemes in the non-dominated solution set are sorted from largest to smallest, and the comprehensive evaluation value of the m-th target service execution scheme after sorting is determined as the set threshold. Finally, the target service execution schemes in the non-dominated solution set whose comprehensive evaluation value is greater than the set threshold are determined as service execution schemes. and a collection of service execution plans. This serves as the final set of service execution solutions.

[0071] Optionally, solving the comprehensive objective function to obtain the initial service execution scheme set includes:

[0072] Determine the global total resource limit and the local total resource limit. The global total resource limit is related to the resource consumption corresponding to the global demand, and the local total resource limit is related to the resource consumption corresponding to the local demand.

[0073] The global total resource limit is defined as the global demand constraint condition of the global demand optimization function in the comprehensive objective function;

[0074] The local total resource limit is defined as the local demand constraint condition of the local demand optimization function in the comprehensive objective function;

[0075] Based on the global and local requirement constraints, the comprehensive objective function is solved to obtain an initial set of service execution schemes.

[0076] In this embodiment, the global total resource limit is related to the resource consumption corresponding to the global requirements of the service to be executed; the resource consumption of the global requirements when executing the service cannot exceed the global total resource limit. The local total resource limit is related to the resource consumption corresponding to the local requirements of the service to be executed; the resource consumption of the local requirements when executing the service cannot exceed the local total resource limit. Global requirement constraints are conditions set for the resource consumption of global requirements when executing the service. Local requirement constraints are conditions set for the resource consumption of local requirements when executing the service.

[0077] For example, the comprehensive objective function The global demand optimization function in the comprehensive objective function. and local demand optimization function Multi-objective collaborative optimization is performed as follows:

[0078] ;

[0079] in, The resource consumption for the k-th global demand. It is the consumption of the j-th local demand. Due to global resource limitations, This is due to local total resource constraints. Based on global demand constraints... Optimization function for global requirements Solve according to the local demand constraints. Optimization function for local requirements By solving this problem, we can obtain the initial set of service execution schemes. .

[0080] Example 2

[0081] Figure 2 This is a flowchart of a method for constructing a comprehensive objective function according to Embodiment 2 of the present invention. This embodiment focuses on the method for constructing the comprehensive objective function in the above embodiment. Figure 2 As shown, the method includes:

[0082] S210. Determine the structured dataset.

[0083] S220. Semantic parsing is performed on the structured dataset using a semantic parsing model to obtain key information.

[0084] S230. Using the semantic parsing model and the structured dataset, generate the global requirement optimization function corresponding to the business rule and the local requirement optimization function corresponding to the key requirement.

[0085] Specifically, the semantic parsing model generates a business rule extraction model and a requirement identification model based on the extracted business rules and key requirements. These models are then used to process the structured dataset, yielding a global requirement optimization function corresponding to the business rules and a local requirement optimization function corresponding to the key requirements. The business rule extraction model corresponds to the business rules within the key information and is related to the global requirements of the service to be executed. The requirement identification model corresponds to the key requirements within the key information and is related to the local requirements of the service to be executed.

[0086] For example, a business rule extraction model and demand identification model The training and optimization process can be: building a business rule extraction model and demand identification model joint loss function The joint loss function is used to evaluate the business rule extraction model. and demand identification model Performance, This is the adjustment coefficient for the loss function. and These are the business rules and key requirements extracted from the i-th rule, respectively. and Business rule extraction model and demand identification model The prediction result on the i-th structured data item, where N is the number of items in the dataset. It is a small constant that avoids the denominator being zero.

[0087] Optionally, the step of generating the global requirement optimization function corresponding to the business rule and the local requirement optimization function corresponding to the key requirement using the semantic parsing model and the structured dataset includes:

[0088] Using the semantic parsing model, a business rule extraction model corresponding to the business rule and a requirement identification model corresponding to the key requirement are generated;

[0089] The structured dataset is optimized using the business rule extraction model to obtain at least one global requirement for the service to be executed.

[0090] The structured dataset is optimized using the demand identification model to obtain at least one local demand for the service to be executed.

[0091] For each global requirement, determine the global requirement optimization result and global requirement resource consumption corresponding to the global requirement. The global requirement optimization result includes the result obtained by the business rule extraction model processing the business rule.

[0092] For each local requirement, determine the local requirement optimization result and local requirement resource consumption, wherein the local requirement optimization result includes the result obtained by the requirement identification model processing the key requirement;

[0093] Based on the global requirement optimization results and global requirement resource consumption corresponding to each of the global requirements, construct the global requirement optimization function corresponding to the business rule;

[0094] Based on the optimization results and resource consumption of each local requirement, the local requirement optimization function corresponding to the key requirement is constructed.

[0095] In this embodiment, the global requirement optimization result includes the result obtained by the business rule extraction model processing the business rules, and the global requirement optimization result is used to indicate the optimization result of each global requirement. The global requirement resource consumption is used to indicate the resource consumption of each global requirement. The local requirement optimization result includes the result obtained by the requirement identification model processing the key requirements, and the local requirement optimization result is used to indicate the optimization result of each local requirement. The local requirement resource consumption is used to indicate the resource consumption of each local requirement.

[0096] For example, the semantic parsing model is based on business rules extracted from key information. and key needs Generate business rule extraction models respectively and demand identification model Based on business rules and key needs Models are extracted through business rules respectively. and demand identification model Optimize the structured dataset to generate at least one global requirement. and at least one local requirement .

[0097] For each global requirement, determine the corresponding global requirement optimization result. and global resource consumption Based on the optimization results and resource consumption of each global requirement, a global requirement optimization function corresponding to the business rule is constructed. ;in, Extract the global requirement optimization result of the model for the k-th global requirement for the business rules, where K is the number of dimensions of the global requirement. This represents the weight coefficient of the k-th global demand. The global resource consumption for the k-th global requirement. .

[0098] Similarly, for each local requirement, determine the corresponding local requirement optimization result. and local demand for resource consumption By analyzing the optimization results and resource consumption of each local requirement, an optimization function for the local requirements corresponding to the key requirements is constructed. ;in, Let J represent the local requirement optimization result of the requirement identification model for the j-th local requirement, where J represents the number of dimensions of the local requirement. Let j be the weight coefficient of the j-th local demand. For user satisfaction related to the j-th local requirement, This represents the resource consumption of the j-th local demand.

[0099] S240. Determine the weight adjustment coefficients, which include the global demand weight coefficient and the local demand weight coefficient.

[0100] In this embodiment, the weight adjustment coefficients can be set according to the specific scenario of the task to be executed, so as to maintain a balance between global and local requirements. The global requirement weight coefficient can be understood as the weight corresponding to the global requirements of the task to be executed. The local requirement weight coefficient can be understood as the weight corresponding to the local requirements of the task to be executed.

[0101] For example, the global demand weight coefficients can be dynamically updated using deep reinforcement learning algorithms. and local demand weighting coefficient :

[0102] ;

[0103] ;

[0104] in, and These are the global demand weight coefficient and the local demand weight coefficient at time step t+1, respectively. and These are the global demand weighting coefficients and local demand weighting coefficients at time step t; How should the learning rate be set? and These are the immediate rewards for global and local demands at time step t, respectively. and For state Next action and The Q-value function at that time; These are the parameters of the deep reinforcement learning model; and They represent respectively to and The gradient.

[0105] S250. The sum of the product of the global demand optimization function and the global demand weight coefficient, and the product of the local demand optimization function and the local demand weight coefficient, is determined as the comprehensive objective function.

[0106] For example, using global demand weighting coefficients and local demand weighting coefficient and global demand optimization function and local demand optimization function The comprehensive objective function is obtained. .

[0107] S260. Solve the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme.

[0108] The technical solution of this invention utilizes the semantic parsing model and the structured dataset to generate a global requirement optimization function corresponding to the business rules and a local requirement optimization function corresponding to the key requirements; determines weight adjustment coefficients; and sums the products of the global requirement optimization function and the global requirement weight coefficients, and the products of the local requirement optimization function and the local requirement weight coefficients, to determine the comprehensive objective function. By constructing the comprehensive objective function through the semantic parsing model, the structured dataset, and key information, dynamic parsing of the structured dataset is achieved. Compared with traditional manual parsing and static planning tools, this improves planning efficiency and eliminates the risk of human error.

[0109] Example 3

[0110] Figure 3 This is a schematic diagram of a scheme generation device according to Embodiment 3 of the present invention. Figure 3 As shown, the device includes:

[0111] The first determining module 310 is used to determine a structured dataset, the structured dataset including at least one type of data related to the service to be executed;

[0112] The parsing module 320 is used to perform semantic parsing on the structured dataset through a semantic parsing model to obtain key information, including information required to execute the service to be executed, including business rules and key requirements;

[0113] The construction module 330 is used to construct a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information. The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements.

[0114] The solver module 340 is used to solve the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme, wherein the service execution scheme is a scheme for executing the service to be executed.

[0115] The scheme generation apparatus provided in this embodiment of the invention determines a structured dataset through a first determining module; performs semantic parsing on the structured dataset using a semantic parsing model through a parsing module to obtain key information; constructs a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information through a construction module; and solves the comprehensive objective function through a solving module to obtain a set of service execution schemes consisting of at least one service execution scheme. Through the cooperation between the modules, and by performing semantic parsing on the structured dataset using the semantic parsing model, dynamic parsing of the structured dataset is achieved to meet the need for flexible handling of services to be executed. By constructing and solving the comprehensive objective function, service execution schemes are obtained, enabling timely responses to services to be executed, improving planning efficiency, and reducing human error rates.

[0116] In one embodiment, the first determining module 310 is specifically used for:

[0117] Obtain multi-source data related to the service to be executed, including execution file data, business requirement data, resource consumption data, and project progress data required when executing the service to be executed;

[0118] Construct the unstructured dataset corresponding to the multi-source data;

[0119] The unstructured dataset is converted to a new format to obtain an initial structured dataset.

[0120] The initial structured dataset is preprocessed to obtain a structured dataset. The preprocessing includes word segmentation, data cleaning, and structuring.

[0121] In one embodiment, the scheme generation apparatus further includes a second determining module, specifically used for:

[0122] Determine the set of model parameters, which is related to the complexity of the structured dataset;

[0123] The parameters in the model parameter set are determined as the parameters of the semantic parsing model.

[0124] In one embodiment, the building module 330 includes:

[0125] The generation unit is used to generate the global requirement optimization function corresponding to the business rule and the local requirement optimization function corresponding to the key requirement using the semantic parsing model and the structured dataset.

[0126] The first determining unit is used to determine the weight adjustment coefficient, which includes a global demand weight coefficient and a local demand weight coefficient.

[0127] The second determining unit is used to determine the comprehensive objective function by summing the product of the global demand optimization function and the global demand weight coefficient, and the product of the local demand optimization function and the local demand weight coefficient.

[0128] In one embodiment, the generating unit is specifically used for:

[0129] Using the semantic parsing model, a business rule extraction model corresponding to the business rule and a requirement identification model corresponding to the key requirement are generated;

[0130] The structured dataset is optimized using the business rule extraction model to obtain at least one global requirement for the service to be executed.

[0131] The structured dataset is optimized using the demand identification model to obtain at least one local demand for the service to be executed.

[0132] For each global requirement, determine the global requirement optimization result and global requirement resource consumption corresponding to the global requirement. The global requirement optimization result includes the result obtained by the business rule extraction model processing the business rule.

[0133] For each local requirement, determine the local requirement optimization result and local requirement resource consumption, wherein the local requirement optimization result includes the result obtained by the requirement identification model processing the key requirement;

[0134] Based on the global requirement optimization results and global requirement resource consumption corresponding to each of the global requirements, construct the global requirement optimization function corresponding to the business rule;

[0135] Based on the optimization results and resource consumption of each local requirement, the local requirement optimization function corresponding to the key requirement is constructed.

[0136] In one embodiment, the solving module 340 includes:

[0137] The solution unit is used to solve the comprehensive objective function to obtain an initial service execution scheme set, wherein the initial service execution scheme set includes at least one initial service execution scheme;

[0138] The analysis unit is used to perform frontier analysis on each initial service execution scheme in the initial service execution scheme set, screen out at least one target service execution scheme, and form a non-dominated solution set from the at least one target service execution scheme;

[0139] The calculation unit is used to calculate the comprehensive evaluation value of each target service execution scheme in the non-dominated solution set, and when the comprehensive evaluation value is greater than a set threshold, the target service execution scheme is determined as a service execution scheme.

[0140] The third determining unit is used to determine the set of service execution schemes as a service execution scheme set.

[0141] In one embodiment, the solving unit is specifically used for:

[0142] Determine the global total resource limit and the local total resource limit. The global total resource limit is related to the resource consumption corresponding to the global demand, and the local total resource limit is related to the resource consumption corresponding to the local demand.

[0143] The global total resource limit is defined as the global demand constraint condition of the global demand optimization function in the comprehensive objective function;

[0144] The local total resource limit is defined as the local demand constraint condition of the local demand optimization function in the comprehensive objective function;

[0145] Based on the global and local requirement constraints, the comprehensive objective function is solved to obtain an initial set of service execution schemes.

[0146] The scheme generation device provided in this embodiment of the invention can execute the scheme generation method provided in any embodiment of the invention. Through the cooperation and collaborative work between the modules, the scheme is generated, and it has the corresponding functional modules and beneficial effects of the execution method.

[0147] Example 4

[0148] According to embodiments of the present invention, the present invention also provides an electronic device and a computer-readable storage medium.

[0149] Figure 4 This is a block diagram of an electronic device according to Embodiment 4 of the present invention, which can implement the scheme generation method described in the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0150] like Figure 4 As shown, the electronic device 410 includes at least one processor 411 and a memory, such as a read-only memory (ROM) 412 or a random access memory (RAM) 413, communicatively connected to the at least one processor 411. The memory stores computer programs executable by the at least one processor. The processor 411 can perform various appropriate actions and processes based on the computer program stored in the ROM 412 or loaded from storage unit 418 into the RAM 413. The RAM 413 may also store various programs and data required for the operation of the electronic device 410. The processor 411, ROM 412, and RAM 413 are interconnected via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0151] Multiple components in the electronic device are connected to the I / O interface 415, including: an input unit 416, such as a keyboard, mouse, etc.; an output unit 417, such as various types of displays, speakers, etc.; a storage unit 418, such as a disk, optical disk, etc.; and a communication unit 419, such as a network card, modem, wireless transceiver, etc. The communication unit 419 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0152] Processor 411 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 411 performs the various methods and processes described above, such as scheme generation methods.

[0153] In some embodiments, the scheme generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 418. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 410 via ROM 412 and / or communication unit 419. When the computer program is loaded into RAM 413 and executed by processor 411, one or more steps of the scheme generation method described above may be performed. Alternatively, in other embodiments, processor 411 may be configured to perform the scheme generation method by any other suitable means (e.g., by means of firmware).

[0154] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0155] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0156] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0157] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0158] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0159] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0160] The technical solution of this invention provides a solution generation method, apparatus, electronic device, and storage medium. It involves: determining a structured dataset; performing semantic parsing on the structured dataset using a semantic parsing model to obtain key information; constructing a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information; and solving the comprehensive objective function to obtain a set of service execution solutions consisting of at least one service execution solution. By performing semantic parsing on the structured dataset using a semantic parsing model, dynamic parsing of the structured dataset is achieved to meet the need for flexible handling of services to be executed. By constructing and solving the comprehensive objective function to obtain service execution solutions, timely responses to services to be executed can be provided, improving planning efficiency and reducing human error rates.

[0161] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0162] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A scheme generation method, characterized in that, include: Determine a structured dataset, the structured dataset including at least one type of data related to the service to be executed; The structured dataset is semantically parsed using a semantic parsing model to obtain key information. The key information includes information required to execute the service to be executed, including business rules and key requirements. The business rules include the business rules contained in the structured dataset, and the key requirements include the requirements contained in the structured dataset. Using the structured dataset, the semantic parsing model, and the key information, a comprehensive objective function is constructed. The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements. Solving the comprehensive objective function yields a set of service execution schemes consisting of at least one service execution scheme, wherein the service execution scheme is a scheme for executing the service to be executed.

2. The method according to claim 1, characterized in that, The determination of the structured dataset includes: Obtain multi-source data related to the service to be executed, including execution file data, business requirement data, resource consumption data, and project progress data required when executing the service to be executed; Construct the unstructured dataset corresponding to the multi-source data; The unstructured dataset is converted to a new format to obtain an initial structured dataset. The initial structured dataset is preprocessed to obtain a structured dataset. The preprocessing includes word segmentation, data cleaning, and structuring.

3. The method according to claim 1, characterized in that, Before performing semantic parsing on the structured dataset using a semantic parsing model to obtain key information, the method further includes: Determine the set of model parameters, which is related to the complexity of the structured dataset; The parameters in the model parameter set are determined as the parameters of the semantic parsing model.

4. The method according to claim 1, characterized in that, The step of constructing a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information includes: Using the semantic parsing model and the structured dataset, a global requirement optimization function corresponding to the business rule and a local requirement optimization function corresponding to the key requirement are generated. Determine the weight adjustment coefficients, which include global demand weight coefficients and local demand weight coefficients; The sum of the product of the global demand optimization function and the global demand weight coefficient, and the product of the local demand optimization function and the local demand weight coefficient, is determined as the comprehensive objective function.

5. The method according to claim 4, characterized in that, The step of generating the global requirement optimization function corresponding to the business rule and the local requirement optimization function corresponding to the key requirement using the semantic parsing model and the structured dataset includes: Using the semantic parsing model, a business rule extraction model corresponding to the business rule and a requirement identification model corresponding to the key requirement are generated; The structured dataset is optimized using the business rule extraction model to obtain at least one global requirement for the service to be executed. The structured dataset is optimized using the demand identification model to obtain at least one local demand for the service to be executed. For each global requirement, determine the global requirement optimization result and global requirement resource consumption corresponding to the global requirement. The global requirement optimization result includes the result obtained by the business rule extraction model processing the business rule. For each local requirement, determine the local requirement optimization result and local requirement resource consumption, wherein the local requirement optimization result includes the result obtained by the requirement identification model processing the key requirement; Based on the global requirement optimization results and global requirement resource consumption corresponding to each of the aforementioned global requirements, construct the global requirement optimization function corresponding to the business rule; Based on the optimization results of each local requirement and the resource consumption of each local requirement, the local requirement optimization function corresponding to the key requirement is constructed.

6. The method according to claim 1, characterized in that, Solving the comprehensive objective function yields a set of service execution schemes consisting of at least one service execution scheme, including: Solving the comprehensive objective function yields an initial service execution scheme set, which includes at least one initial service execution scheme. For each initial service execution scheme in the set of initial service execution schemes, a frontier analysis is performed to select at least one target service execution scheme, and the at least one target service execution scheme forms a non-dominated solution set; For each target service execution scheme in the non-dominated solution set, a comprehensive evaluation value of the target service execution scheme is calculated. When the comprehensive evaluation value is greater than a set threshold, the target service execution scheme is determined as a service execution scheme. The set of all the service execution schemes is defined as the service execution scheme set.

7. The method according to claim 6, characterized in that, Solving the comprehensive objective function yields an initial set of service execution schemes, including: Determine the global total resource limit and the local total resource limit. The global total resource limit is related to the resource consumption corresponding to the global demand, and the local total resource limit is related to the resource consumption corresponding to the local demand. The global total resource limit is defined as the global demand constraint condition of the global demand optimization function in the comprehensive objective function; The local total resource limit is defined as the local demand constraint condition of the local demand optimization function in the comprehensive objective function; Based on the global and local requirement constraints, the comprehensive objective function is solved to obtain an initial set of service execution schemes.

8. A scheme generation apparatus, characterized in that, include: A determination module is used to determine a structured dataset, the structured dataset including at least one type of data related to the service to be executed; The parsing module is used to perform semantic parsing on the structured dataset through a semantic parsing model to obtain key information, including information required to execute the service to be executed, including business rules and key requirements; The construction module is used to construct a comprehensive objective function using the structured dataset, the semantic parsing model, and the key information. The comprehensive objective function consists of a global requirement optimization function and a local requirement optimization function. The global requirement optimization function is related to the global requirements of the service to be executed, and the local requirement optimization function is related to the local requirements of the service to be executed. The global requirement optimization function corresponds to the business rules, and the local requirement optimization function corresponds to the key requirements. The solution module is used to solve the comprehensive objective function to obtain a set of service execution schemes consisting of at least one service execution scheme, wherein the service execution scheme is a scheme for executing the service to be executed.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the scheme generation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the scheme generation method according to any one of claims 1-7.

Images