CN116957510A

CN116957510A - Planning scheme determining method, device and storage medium

Info

Publication number: CN116957510A
Application number: CN202310945504.XA
Authority: CN
Inventors: 蔡一欣; 刘颖慧; 刘楠; 李堃
Original assignee: China United Network Communications Group Co Ltd
Current assignee: China United Network Communications Group Co Ltd
Priority date: 2023-07-28
Filing date: 2023-07-28
Publication date: 2023-10-27

Abstract

The application provides a planning scheme determining method, a planning scheme determining device and a storage medium, relates to the technical field of computers, and is used for improving the accuracy of a planning scheme. The method comprises the following steps: acquiring parameters of a target item and first decision models of a plurality of application scenes; the first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects; the plurality of items includes a target item; the parameters comprise application scenes, service requirements and service characteristics; determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the business requirements; the second decision model is used for outputting a planning scheme of the target project; and inputting the business characteristics of the target project into a second decision model to obtain a planning scheme of the target project.

Description

Planning scheme determining method, device and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and apparatus for determining a planning scheme, and a storage medium.

Background

The scientific research enterprises filter and integrate the scientific research data to determine a planning scheme of the project so as to realize digital transformation of the scientific research enterprises. Currently, the method for determining the planning scheme by a scientific enterprise is generally as follows: and (3) building a data base or a data middle table, integrating various scientific research data, and analyzing the integrated scientific research data to obtain a planning scheme.

However, in the process of analyzing the integrated scientific data, the integrated scientific data needs to be manually analyzed, so that the actual value of the scientific data may be difficult to mine, the data analysis is not in place or inaccurate, and the planning scheme may be inaccurate.

Disclosure of Invention

The application provides a planning scheme determining method, a planning scheme determining device and a storage medium, which are used for improving the accuracy of a planning scheme.

In order to achieve the above purpose, the application adopts the following technical scheme:

in a first aspect, the present application provides a method for determining a planning scheme, the method comprising: acquiring parameters of a target item and first decision models of a plurality of application scenes; the first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects; the plurality of items includes a target item; the parameters comprise application scenes, service requirements and service characteristics; determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the business requirements; the second decision model is used for outputting a planning scheme of the target project; and inputting the business characteristics of the target project into a second decision model to obtain a planning scheme of the target project.

In one possible implementation, determining, based on the application scenario and the business requirement, a second decision model from the first decision models of the plurality of application scenarios includes: and determining the first decision model as a second decision model under the condition that the application scene of the first decision model is the same as that of the target project and the output result of the first decision model meets the service requirement.

In one possible implementation, determining, based on the application scenario, a second decision model from the first decision models of the plurality of application scenarios includes: and determining the first decision model as a second decision model under the condition that the application scene of the first decision model is the same as that of the target project.

In one possible implementation, obtaining a first decision model of a plurality of application scenarios includes: acquiring online characteristics of a plurality of application scenes, offline characteristics of the plurality of application scenes and an initial decision model of the plurality of application scenes; the online features include project completion and/or project outcome yield; the offline feature includes at least one of: item attributes, item categories, or

Project scale; matching the online characteristics of the plurality of application scenes with the offline characteristics of the plurality of application scenes to obtain online sample characteristics of the plurality of application scenes; and optimizing the initial decision models of the plurality of application scenes based on the online sample characteristics of the plurality of application scenes to obtain first decision models of the plurality of application scenes.

In one possible implementation, obtaining an initial decision model for a plurality of application scenarios includes: transforming and combining the offline characteristics of the plurality of application scenes to determine the offline sample characteristics of the plurality of application scenes; training the preset model based on the offline sample characteristics of the plurality of application scenes to obtain an initial decision model of the plurality of application scenes.

In one possible implementation, the method further includes: storing the online features to a feature library; and carrying out validity analysis on the features in the feature library to determine the availability of the features in the feature library.

In one possible implementation, the application scenario may include at least one of: the intelligent planning scene of scientific research activities, the intelligent selection of cultivation scenes by talents, the internal and external knowledge analysis application scene or the flexible infrastructure application scene.

In a second aspect, the present application provides a planning scheme determination apparatus, the apparatus comprising: a communication unit and a processing unit; the communication unit is used for acquiring parameters of a target item and first decision models of a plurality of application scenes; the first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects; the plurality of items includes a target item; the parameters comprise application scenes, service requirements and service characteristics; the processing unit is used for determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the service requirements; the second decision model is used for outputting a planning scheme of the target project; and the processing unit is also used for inputting the business characteristics of the target project into the second decision model to obtain the planning scheme of the target project.

In one possible implementation manner, the processing unit is further configured to determine the first decision model as the second decision model when the application scenario of the first decision model is the same as the application scenario of the target item, and the output result of the first decision model meets the service requirement.

In a possible implementation manner, the processing unit is further configured to determine the first decision model as the second decision model when the application scenario of the first decision model is the same as the application scenario of the target item.

In one possible implementation manner, the communication unit is further configured to obtain online features of the multiple application scenes, offline features of the multiple application scenes, and an initial decision model of the multiple application scenes; the online features include project completion and/or project outcome yield; the offline feature includes at least one of: item attributes, item categories, or item scales; the processing unit is also used for matching the online characteristics of the plurality of application scenes with the offline characteristics of the plurality of application scenes to obtain online sample characteristics of the plurality of application scenes; the processing unit is further used for optimizing the initial decision models of the plurality of application scenes based on the online sample characteristics of the plurality of application scenes to obtain first decision models of the plurality of application scenes.

In a possible implementation manner, the processing unit is further configured to transform and combine offline features of the multiple application scenarios, and determine offline sample features of the multiple application scenarios; the processing unit is further used for training the preset model based on the offline sample characteristics of the plurality of application scenes to obtain initial decision models of the plurality of application scenes.

In a possible implementation manner, the processing unit is further configured to store the online feature to a feature library; and the processing unit is also used for carrying out validity analysis on the features in the feature library and determining the availability of the features in the feature library.

In a third aspect, the present application provides a planning scheme determining apparatus, the apparatus comprising: a processor and a communication interface; the communication interface is coupled to a processor for running a computer program or instructions to implement the planning scheme determination method as described in any one of the possible implementations of the first aspect and the first aspect.

In a fourth aspect, the present application provides a computer readable storage medium having instructions stored therein which, when run on a terminal, cause the terminal to perform a method of determining a planning scheme as described in any one of the possible implementations of the first aspect and the first aspect.

In a fifth aspect, the present application provides a computer program product comprising instructions which, when run on a planning scheme determination device, cause the planning scheme determination device to perform the planning scheme determination method as described in any one of the possible implementations of the first aspect and the first aspect.

In a sixth aspect, the present application provides a chip comprising a processor and a communication interface, the communication interface and the processor being coupled, the processor being for running a computer program or instructions to implement the method of determining a planning scheme as described in any one of the possible implementations of the first aspect and the first aspect.

In particular, the chip provided in the present application further includes a memory for storing a computer program or instructions.

In the planning scheme determining method provided by the embodiment of the application, as the first decision-making models of the plurality of application scenes establish a strong association relationship between the business characteristics of the target item and the planning scheme of the target item in the plurality of application scenes, and the second decision-making model is selected and determined from the first decision-making models of the plurality of application scenes by the planning scheme determining equipment based on the application scenes of the target item and/or the business requirements of the target item, the planning scheme determining equipment inputs the business characteristics of the target item into the second decision-making model, and a more accurate planning scheme of the target item can be obtained. And the planning scheme determining device determines the second decision model from the first decision models of the plurality of application scenes based on the application scenes of the target projects and/or the business requirements of the target projects, so that the application scenes of the second decision model can be ensured to accord with the application scenes and/or the business requirements of the target projects, and the accuracy of the planning scheme is further improved.

Drawings

Fig. 1 is a schematic structural diagram of a planning scheme determining system according to an embodiment of the present application;

fig. 2 is a schematic structural diagram of a planning scheme determining device according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of another planning scheme determining apparatus according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a planning scheme determining device according to an embodiment of the present application;

FIG. 5 is a flowchart of a method for determining a planning scheme according to an embodiment of the present application;

FIG. 6 is a flowchart of another method for determining a planning scheme according to an embodiment of the present application;

FIG. 7 is a flowchart of another method for determining a planning scheme according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of another planning scheme determining apparatus according to an embodiment of the present application.

Detailed Description

The following describes a method, an apparatus and a storage medium for determining a planning scheme according to an embodiment of the present application in detail with reference to the accompanying drawings.

The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone.

The terms "first" and "second" and the like in the description and in the drawings are used for distinguishing between different objects or between different processes of the same object and not for describing a particular order of objects.

Furthermore, references to the terms "comprising" and "having" and any variations thereof in the description of the present application are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or apparatus.

It should be noted that, in the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "e.g." in an embodiment should not be taken as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.

Along with the rapid development of information technology, the digitization transformation enters a maturity stage, and scientific enterprises serve as important innovation main bodies in a scientific innovation system, so that the information system can be built to realize the digitization of scientific resource sharing, scientific management decision, management architecture and management flow (namely, the digitization of the scientific enterprises), and further, the scientific innovation can be led and the high-quality development can be promoted.

At present, scientific enterprises can safely transmit scientific data and filter and integrate the scientific data, so that the scientific data can fully play the data value, and the scientific enterprises can determine a planning scheme of a project based on the scientific data, so that the digital transformation of the scientific enterprises is realized. However, in the process of determining a planning scheme of an item based on scientific research data, a scientific research enterprise may have problems of insufficient information intercommunication, information data redundancy, information management failure and unreachable information feedback.

In order to avoid the problems of insufficient information intercommunication, redundant information data, invalid information management, unqualified information feedback and the like in the process of determining a planning scheme of an item based on scientific research data, a method for determining the planning scheme by a scientific research enterprise is generally as follows: the scientific research enterprises can build a data base or a data middle platform, integrate various scientific research data, and combine the application scenes of the projects and the business requirements of the projects to perform data energy on the integrated scientific research data. And analyzing scientific research data after data enabling by the scientific research enterprises to obtain a planning scheme.

However, the method for determining the planning scheme of the scientific research enterprise cannot directly obtain the planning scheme of the project, and the planning scheme of the project can be determined by manually analyzing scientific research data after data energization. In the process of manually analyzing the scientific research data after the data are energized, the actual value of the scientific research data can be difficult to mine, and therefore the situation that the data are not analyzed in place or are inaccurate can be caused, and the planning scheme can be inaccurate.

In view of this, in the method for determining a planning scheme provided in the embodiment of the present application, since the first decision-making models of the multiple application scenarios establish a strong association relationship between the business features of the target item and the planning scheme of the target item in the multiple application scenarios, and the second decision-making model is selected and determined by the planning scheme determining device based on the application scenario of the target item and/or the business requirement of the target item from the first decision-making models of the multiple application scenarios, the planning scheme determining device inputs the business features of the target item into the second decision-making model, so that a more accurate planning scheme of the target item can be obtained. And the planning scheme determining device determines the second decision model from the first decision models of the plurality of application scenes based on the application scenes of the target projects and/or the business requirements of the target projects, so that the application scenes of the second decision model can be ensured to accord with the application scenes and/or the business requirements of the target projects, and the accuracy of the planning scheme is further improved.

Illustratively, as shown in fig. 1, fig. 1 shows a schematic structural diagram of a planning scheme determining system according to an embodiment of the present application. The planning scheme determination system includes: the planning scheme determines the device 101 and the server 102. Fig. 1 illustrates an example in which the plan determining system includes a plan determining apparatus 101 and a server 102.

The planning scheme determining device 101 is configured to obtain parameters of a target item and first decision models of multiple application scenarios, and determine a second decision model from the first decision models of the multiple application scenarios based on the application scenarios and/or service requirements. The planning scheme determining device 101 is further configured to input the business feature of the target item into the second decision model, to obtain a planning scheme of the target item.

A server 102 for sending a first decision model of a plurality of scenarios to the planning scheme determination device 101.

The first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects. The plurality of items includes a target item. The parameters include application scenarios, business requirements, and business characteristics. The second decision model is used for outputting a planning scheme of the target project.

In one example, the planning scheme determination device 101 may be a server. The server may be a single server, or may be a server cluster formed by a plurality of servers. In some implementations, the server cluster may also be a distributed cluster.

Illustratively, as shown in fig. 2, fig. 2 shows a schematic structural diagram of a planning scheme determining apparatus provided by an embodiment of the present application. The planning scheme determination device 101 may include an application scenario module, a model library, and a processing module.

An application scenario module may be used to provide parameters of the target item for the planning scheme determination device 101.

The model library may be used to provide a first decision model of a plurality of application scenarios for the planning scheme determination device 101.

The processing module can be used for determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the service demands, and inputting the service characteristics of the target project into the second decision model to obtain the planning scheme of the target project.

Alternatively, the model library may include model version, model verification, model online, model diagnosis, and model evaluation. The model version can be used for acquiring a model uploaded by a user and storing the uploaded model. The model versions can also be used for merging models with the same application scene and/or models with similar output planning schemes to obtain multiple versions of the model. The model verification can be used for verifying the characteristics used in the model training process, and performing functional verification and performance test on the models to determine the application scenes of the models and the service requirements which can be met by the models. The model is online, can be used for deploying the combined model, and verifies the combined model based on the fort model. Model diagnostics, which may be used to diagnose features used in training a model, and to determine parameters for various versions of the model. Model evaluation may be used to evaluate the model. For example, model evaluation evaluates the model based on a model evaluation index (e.g., based on the area under the subject's working characteristic curve (AUC)). As another example, model evaluation evaluates models based on A/B testing (A/B text).

In an alternative implementation, the planning scheme determination device 101 may also include a training module, a feature library, and an intelligent decision platform.

And the training module can be used for carrying out business logic processing on the features (online features and offline features) to obtain sample features (online sample features and offline sample features) required by the training model. The training module can also be used for carrying out arithmetic logic processing on the characteristics. The business logic process may include model training, feature matching, model execution, and model embedding, among others. The algorithmic logic processes may include machine learning algorithms, deep learning algorithms, federal learning algorithms, and operational research algorithms.

The feature library may include feature design, feature extraction, feature processing, feature management, and feature quality. The feature library may be used to obtain and store online features of multiple application scenarios and offline features of multiple application scenarios, and determine a correspondence between the online features and the offline features (i.e., whether features of the same item in the same application scenario are present). The feature library can also be used for monitoring and validity analysis of the features and establishing a quality monitoring mechanism to ensure the service quality of the features in the feature library.

The intelligent decision platform can comprise a user management module, a model management module, a feature management module, a scene management module and a decision brain module.

And the user management module can be used for matching corresponding use authorities for users with different user attributes. The user attributes may include administrator users, development users, and general users, among others. The use rights of the administrator user include: and determining the user attribute and creating an application scene of the model. Developing the use rights of the user includes: features (e.g., offline features and online features) and models are obtained, and the models are iterated and optimized based on the features. The usage rights of the general user include: the model is used and scoring and opinion feedback is performed on the model.

The model management module and the feature management module can be used for carrying out version management, check management and grading management on the models in the model library and the features in the feature library, and determining functions, service conditions and application scenes of the models.

The scene management module may be configured to add and delete application scenes in the application scene module, and determine application scenes supported by the planning scheme determining device 101.

The decision brain module can be used for displaying the number and the type of the models of each application scene in a plurality of application scenes in real time and planning schemes of the problems and the output solved by the models.

As shown in fig. 3, fig. 3 is a schematic structural diagram of a planning scheme determining device according to an embodiment of the present application. The planning scheme determination device 101 may include a user operation module, a model loading module, and an execution prediction module.

And the user operation module can be used for acquiring the parameters of the target item.

The model loading module may be configured to obtain first decision models of a plurality of application scenarios, and determine a second decision model from the first decision models of the plurality of application scenarios based on the application scenarios and/or the service requirements.

And the execution prediction module can be used for inputting the business characteristics of the target project into the second decision model to obtain the planning scheme of the target project.

Optionally, the planning scheme determining device 101 may further include a feature acquisition module, a feature processing module, a model processing module, and a user access module.

And the feature acquisition module can be used for acquiring offline features of a plurality of application scenes from the offline database. The feature acquisition module is also used for acquiring online features of a plurality of application scenes and storing the offline features and the online features into a feature library.

The feature processing module can be used for carrying out transformation and combination on the offline features of the plurality of application scenes, determining the offline sample features of the plurality of application scenes, and matching the online features of the plurality of application scenes with the offline features of the plurality of application scenes to obtain the online sample features of the plurality of application scenes.

The model processing module can comprise model evaluation and model automatic optimization and can be used for evaluating and optimizing the model.

A user access module may be used to provide an application program interface (application programming interface, API) for other access parties, so that the other access parties may use the planning scheme to determine the device 101 through the API and obtain the output result of the second decision model (i.e. the planning scheme of the target item).

Optionally, during the process of using the planning scheme determining device 101 by other access parties, the planning scheme determining device 101 may allocate bandwidth to the planning scheme determining device 101 according to the network usage, so as to ensure load balancing of the planning scheme determining device 101.

In another example, the server 102 may be a single server, or may be a server cluster composed of a plurality of servers. In some implementations, the server cluster may also be a distributed cluster. In the embodiment of the present application, the device for implementing the function of the server 102 may be the server 102, or may be a device capable of supporting the server 102 to implement the function, for example, a chip or a chip system.

In addition, the planning scheme determining system described in the embodiment of the present application is for more clearly describing the technical scheme of the embodiment of the present application, and does not constitute a limitation on the technical scheme provided by the embodiment of the present application, and as a person of ordinary skill in the art can know, with evolution of the network architecture and appearance of the new planning scheme determining system, the technical scheme provided by the embodiment of the present application is equally applicable to similar technical problems.

In particular, the apparatus of fig. 1 may employ the constituent structure shown in fig. 4, or may include the components shown in fig. 4. Fig. 4 is a schematic diagram of a plan determining apparatus 400 according to an embodiment of the present application, where the plan determining apparatus 400 may be a plan determining device 101 or a chip or a system on a chip in the plan determining device 101. Alternatively, the planning scheme determination apparatus 400 may be a chip or a system-on-chip therein. As shown in fig. 4, the planning scheme determination apparatus 400 may include a processor 401 and a communication line 402.

Further, the planning scheme determination device 400 may further include a communication interface 403 and a memory 404. The processor 401, the memory 404, and the communication interface 403 may be connected by a communication line 402.

Processor 401 is, among other things, a CPU, general-purpose processor, network processor (network processor, NP), digital signal processor (digital signal processing, DSP), microprocessor, microcontroller, programmable logic device (programmable logic device, PLD), or any combination thereof. The processor 401 may also be any other device having a processing function, such as a circuit, a device, or a software module, without limitation.

A communication line 402 for communicating information between the components included in the planning scheme determination apparatus 400.

A communication interface 403 for communicating with other devices or other communication networks. The other communication network may be an ethernet, a radio access network (radio access network, RAN), a wireless local area network (wireless local area networks, WLAN), etc. The communication interface 403 may be a module, a circuit, a communication interface, or any device capable of enabling communication.

Memory 404 for storing instructions. Wherein the instructions may be computer programs.

The memory 404 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device capable of storing static information and/or instructions, a random access memory (random access memory, RAM) or other type of dynamic storage device capable of storing information and/or instructions, an EEPROM, a CD-ROM (compact disc read-only memory) or other optical disk storage, an optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, etc.

It is noted that the memory 404 may exist separately from the processor 401 or may be integrated with the processor 401. Memory 404 may be used to store instructions or program code or some data, etc. The memory 404 may be located in the planning scheme determination device 400 or may be located outside the planning scheme determination device 400, without limitation. The processor 401 is configured to execute instructions stored in the memory 404 to implement the planning scheme determining method provided in the following embodiments of the present application.

In one example, processor 401 may include one or more CPUs, e.g., CPU0 and CPU1.

As an alternative implementation, the planning scheme determination apparatus 400 includes a plurality of processors.

As an alternative implementation, the planning scheme determination apparatus 400 further includes an output device and an input device. The output device is illustratively a display screen, speaker (spaker) or the like, and the input device is a keyboard, mouse, microphone or joystick or the like.

It should be noted that the planning scheme determining apparatus 400 may be a desktop computer, a portable computer, a network server, a mobile phone, a tablet computer, a wireless terminal, an embedded device, a chip system, or a device having a similar structure as in fig. 4. Furthermore, the constituent structures shown in fig. 4 do not constitute limitations on the respective apparatuses in fig. 1 and 4, and the respective apparatuses in fig. 1 and 4 may include more or less components than illustrated, or may combine some components, or may be arranged differently, in addition to the components shown in fig. 4.

In the embodiment of the application, the chip system can be composed of chips, and can also comprise chips and other discrete devices.

Further, actions, terms, and the like, which are referred to between embodiments of the present application, are not limited thereto. The message names of interactions between the devices or parameter names in the messages in the embodiments of the present application are just an example, and other names may be used in specific implementations without limitation.

The following describes a planning scheme determination method provided by the embodiment of the present application with reference to the planning scheme determination system shown in fig. 1. In which the terms and the like related to the actions of the embodiments of the present application are mutually referred to, without limitation. The message names of interactions between the devices or parameter names in the messages in the embodiments of the present application are just an example, and other names may be used in specific implementations without limitation. The actions involved in the embodiments of the present application are just an example, and other names may be adopted in the specific implementation, for example: the "included" of the embodiments of the present application may also be replaced by "carried on" or the like.

In order to solve the problems in the prior art, the embodiment of the application provides a planning scheme determining method for improving the accuracy of the planning scheme. As shown in fig. 5, the method includes:

S501, the planning scheme determining device obtains parameters of a target item and first decision models of a plurality of application scenes.

The first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects. The plurality of items includes a target item. The parameters include application scenarios, business requirements, and business characteristics.

By way of example, the business requirements may include at least one of: budget requirements, functional requirements, or personnel requirements. The above is merely an exemplary description of a business requirement, and the business requirement may also include other requirements (e.g., technical requirements), which the present application does not limit in any way.

In an alternative embodiment, the application scenario comprises at least one of: the intelligent planning scene of scientific research activities, the intelligent selection of cultivation scenes by talents, the internal and external knowledge analysis application scene or the flexible infrastructure application scene. The foregoing is merely an exemplary description of an application scenario, and other application scenarios may also be included in the application scenario, which is not limited in any way by the present application.

Optionally, the application scenario of the target item may be any application scenario of the multiple application scenarios.

Illustratively, the business characteristics of the target item include at least one of: item level, item duration, item yield, or item budget. The above is merely an exemplary description of a business feature of a target item, and the business feature of the target item may also include other features (e.g., item size), which the present application does not limit in any way.

As an optional implementation manner, the implementation process of the planning scheme determining device to obtain the first decision model of the multiple application scenarios may be: the planning scheme determining device may obtain offline features of the multiple application scenarios and online features of the multiple application scenarios, and construct an initial decision model of the multiple application scenarios based on the offline features of the multiple application scenarios. The planning scheme determining device may optimize the third decision model based on online features of the multiple application scenarios, to obtain a first decision model of the multiple application scenarios.

As another optional implementation manner, the implementation process of the planning scheme determining device to obtain the first decision model of the multiple application scenarios may further be: the user may upload the initial decision model for the plurality of application scenarios in the planning scheme determination device. The planning scheme determining device determines the application scenes of the initial decision models uploaded by the users and the output planning schemes, and combines the initial decision models with the same application scenes and/or similar output planning schemes to obtain a first decision model.

Optionally, the planning scheme determining device may store the first decision models and the initial decision models of the multiple application scenarios, and generate a model library, so that the subsequent planning scheme determining device selects the model from the model library for use.

It should be noted that each application scenario in the plurality of application scenarios may include a plurality of first decision models.

In one possible implementation manner, the planning scheme determining device may perform a functional test on the first decision models in the model library, and determine an application scenario of each first decision model and a service requirement that can be met by each first decision model. And, the planning scheme determining device performs a functional test on the first decision model in the model library, so that the validity of the first decision model can be determined. Since the planning scheme determination device is a first decision model obtained based on the offline sample feature and the online sample feature, the validity of the first decision model can indicate the validity of the offline sample feature and the online sample feature.

S502, the planning scheme determining device determines a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the service requirements.

The second decision model is used for outputting a planning scheme of the target project.

In an alternative embodiment, the implementation process of the planning scheme determining device in determining the second decision model from the first decision models of the multiple application scenarios based on the application scenarios and the service requirements may be: the planning scheme determining device determines the first decision model as a second decision model under the condition that the application scene of the first decision model is the same as that of the target project and the output result of the first decision model meets the service requirement.

It can be understood that the planning scheme determining device determines the second decision model from the first decision models of the multiple application scenes based on the application scene of the target item and the service requirement of the target item, so that the application scene of the second decision model can be ensured to conform to the application scene of the target item, and the output result of the second decision model can meet the service requirement of the target item.

In another optional embodiment, the implementation process of the planning scheme determining device in determining the second decision model from the first decision models of the multiple application scenarios based on the application scenarios may be: the planning scheme determining device determines the first decision model as a second decision model under the condition that the application scene of the first decision model is the same as the application scene of the target project.

Optionally, before the planning scheme determining device determines the second decision model from the first decision models of the multiple application scenarios based on the application scenarios, the planning scheme determining device may further acquire parameters of the multiple items and a situation that the multiple items use the first decision model, and compare the parameters of the target item with the parameters of the multiple items to determine the first decision model with higher availability of the target item. The planning scheme determination device may determine the second decision model from the first decision model with a higher availability of the target item.

For example, the case that the plurality of items use the first decision model may include the number of times the plurality of items use the first decision model, the business characteristics of the plurality of items input the first decision model, and the planning scheme of the first decision model output obtained by the plurality of items. The above-described exemplary description of the case where the first decision model is used only for a plurality of items, the case where the plurality of items use the first decision model may also include other cases (for example, a problem solved by the plurality of items using the first decision model), to which the present application is not limited in any way.

It can be understood that the planning scheme determining device determines the second decision model from the first decision models of the multiple application scenes based on the application scenes, so that the application scenes of the second decision model can be ensured to conform to the application scenes of the target items, and the planning scheme of the target items obtained by the planning scheme determining device under the application scenes by using the second decision model is accurate.

S503, the planning scheme determining device inputs the business characteristics of the target project into the second decision model to obtain the planning scheme of the target project.

As an example, take the second decision model as an example of an item budget planning model in a scientific research activity intelligent planning scenario: the planning scheme determination device inputs the project level, project duration, project labor, project yield, and project budget of the target project into the project budget planning model. The project budget planning model can directly output the planning scheme for the target project.

It can be appreciated that the user can determine the development trend of the target project according to the planning scheme of the target project, so that the subsequent planning scheme determining device can provide the user with the planning scheme which is more in line with the current development trend.

In an alternative embodiment, the planning scheme determining device may obtain the initial decision models of the multiple application scenarios, and optimize the initial decision models of the multiple application scenarios to obtain the first decision models of the multiple application scenarios, so that the planning scheme determining device may obtain a better decision model, and on the basis of the method embodiment shown in fig. 5, this embodiment provides a possible implementation manner, and in connection with fig. 5, as shown in fig. 6, an implementation process of the planning scheme determining device obtaining the first decision models of the multiple application scenarios may be determined by the following S601 to S603.

S601, the planning scheme determining device obtains online characteristics of a plurality of application scenes, offline characteristics of the plurality of application scenes and an initial decision model of the plurality of application scenes.

Wherein the online characteristics include project completion and/or project outcome yield. The offline feature includes at least one of: item attributes, item categories, or item sizes.

Alternatively, the above is only one exemplary depiction of an online feature and an offline feature. The online and offline features described above may also include other features. Illustratively, the online features may also include other dynamic features of the project (e.g., project progress) and the offline features may also include other static features of the project (e.g., project expertise), to which the present application is not limited.

As an optional implementation manner, the implementation process of the planning scheme determining device to obtain the online features of the multiple application scenes and the offline features of the multiple application scenes may be: the planning scheme determination device may directly select the online feature and the offline feature from the feature library.

In an alternative embodiment, the implementation process of the planning scheme determining device to obtain the initial decision model of the multiple application scenarios may be: the planning scheme determining equipment performs transformation and combination on the offline characteristics of the plurality of application scenes, determines the offline sample characteristics of the plurality of application scenes, and trains a preset model based on the offline sample characteristics of the plurality of application scenes to obtain an initial decision model of the plurality of application scenes.

In one possible implementation manner, the planning scheme determining device performs transformation and combination on offline features of a plurality of application scenarios, and the implementation process of determining the offline sample features of the plurality of application scenarios may be: the planning scheme determining device can randomly select part of offline features from the offline features of the plurality of application scenes, and perform multiple combination processing on the selected part of offline features to obtain a plurality of offline feature combination results. The planning scheme determining device may determine the above-mentioned plurality of offline feature combination results as offline sample features of a plurality of application scenarios.

As an optional implementation manner, the implementation process of training the preset model based on offline sample characteristics of multiple application scenarios by the planning scheme determining device to obtain the initial decision model of multiple application scenarios may: the planning scheme determining device splits the offline sample features into a training feature set and a testing feature set, and selects an appropriate preset model (e.g., a logistic regression model) according to the application scenario and business requirements of the project. The planning scheme determining device may train the preset model with the training feature set based on the algorithm of the training model to obtain an intermediate model with a smaller loss function, and input the test feature set into the intermediate model to evaluate the performance of the intermediate model. The planning scheme determination device adjusts (e.g., adjusts model types, adjusts training feature sets) the intermediate model that is not up to performance, until the performance of the intermediate model is up to standard, and determines the intermediate model that is up to performance as an initial decision model.

In one example, the above-mentioned evaluation index for evaluating the performance of the intermediate model may include at least one of: accuracy, precision, or recall. The above is merely an exemplary description of an evaluation index for evaluating the performance of the intermediate model, and the evaluation index for evaluating the performance of the intermediate model may further include other indexes (for example, F1 values), to which the present application is not limited.

Optionally, the above determining device for planning schemes trains the preset model based on offline sample features of multiple application scenarios, so as to obtain an exemplary description of an initial decision model of multiple application scenarios, where the obtaining of the initial decision model of multiple application scenarios may also be implemented through other training processes, which is not limited in this application.

For example, the above-described planning scheme determination device may determine the initial decision model of the plurality of application scenarios using big data techniques and/or artificial intelligence (artificial intelligence, AI) techniques, and the algorithm of the training model used in the planning scheme determination device in determining the initial decision model of the plurality of application scenarios may include at least one of: a machine learning algorithm, a deep learning algorithm, a federal learning algorithm, or an operations research algorithm. The foregoing is merely an exemplary description of techniques and algorithms for constructing an initial decision model for a plurality of application scenarios, and the techniques and algorithms may include other techniques and algorithms, and the present application is not limited in this regard.

Optionally, the planning scheme determining device may determine a network usage of the planning scheme determining device by the user, and allocate a corresponding network bandwidth to the user according to the network usage, so as to avoid a problem of bandwidth load in determining an initial decision model of multiple application scenarios by the planning scheme determining device.

It can be understood that for different application scenarios, the planning scheme determining device selects offline sample features of a plurality of application scenarios to construct initial decision models of the plurality of application scenarios, so that initial decision models conforming to each application scenario can be obtained, and the subsequent optimization of the initial decision models of each application scenario is facilitated.

S602, the planning scheme determining device matches the online characteristics of the plurality of application scenes with the offline sample characteristics of the plurality of application scenes to obtain the online sample characteristics of the plurality of application scenes.

In an alternative implementation manner, the implementation process of S602 may be: the planning scheme determining device determines the corresponding relation between the online features and the offline features based on the application scene, and determines the online features which are the same as the application scene to which the offline features belong as online sample features of the application scene to which the offline features belong.

S603, the planning scheme determining device optimizes initial decision models of the plurality of application scenes based on online sample characteristics of the plurality of application scenes to obtain first decision models of the plurality of application scenes.

In an alternative implementation manner, the implementation process of S603 may be: the planning scheme determining device inputs the online sample characteristics of the target application scene into an initial decision model of the target application scene to obtain an updated planning scheme of the items in the target application scene. The planning scheme determining device inputs the offline sample characteristics of the target application scene into an initial decision model of the target application scene to obtain a preset planning scheme of the items in the target application scene. Under the condition that the updated planning scheme is different from the preset planning scheme, the planning scheme determining device adjusts or retrains parameters of an initial decision model of the target application scene based on the online sample characteristics of the target application scene to obtain a first decision model of the target application scene. The planning scheme determination device determines a first decision model of the plurality of application scenarios based on the method.

Alternatively, the planning scheme determination apparatus may store the decision model that is not optimized (i.e., the initial decision model) and the decision model that has completed optimization (i.e., the first decision model) in a model library, and perform model diagnosis, model evaluation, and model verification on the models in the model library to determine the validity of the models.

In an optional implementation manner, after the planning scheme determining device obtains the first decision models of the multiple application scenarios, the planning scheme determining device may display, in real time, the number and types of the first decision models of each application scenario in the multiple application scenarios, and the planning scheme of the problems and outputs solved by the first decision models of each application scenario in the multiple application scenarios.

The technical scheme at least brings the following beneficial effects: the offline features can embody static features of the project, and the online features can embody dynamic features of the project, so that the planning scheme determining device optimizes an initial decision model determined based on the offline features based on the online features, the obtained first decision model data can be more accurate, and the output planning scheme is more accurate. And the planning scheme determining equipment matches the online characteristics with the offline characteristics, so that the online characteristics and the offline characteristics are all characteristics of the same item in the same application scene, and the situation that the model optimization errors are caused due to mismatching of the characteristics is avoided.

In an alternative embodiment, the planning scheme determining device may store the online features and the offline features, and perform validity analysis on the online features and the offline features, so that the subsequent planning scheme determining device may use the online features and the offline features in the process of building and optimizing the model, and may select the features with higher availability in the process of determining the sample features, so that the obtained decision model may be more accurate, and further improve the accuracy of the planning scheme.

S701, the planning scheme determining device stores the online features into a feature library.

The feature library may be, for example, a distributed file system (hadoop distributed file system, hdfs), the above is just one exemplary description of a feature library, and the feature library may be other systems, as the application is not limited in any way.

Alternatively, the feature library may provide features in the feature library for other access parties. Other access parties may determine that the device's API accesses the feature library through a planning scheme.

In an alternative implementation, the planning scheme determination device may update the online features in the feature library, since new online features may occur in multiple application scenarios, which may result in changes in the features of the application scenarios. The implementation process of updating the online features in the feature library by the planning scheme determining device may be: the planning scheme determining device obtains the latest online characteristics of the plurality of application scenes, and updates the online characteristics in the characteristic library based on the latest online characteristics of the plurality of application scenes.

For example, the planning scheme determining device may obtain real-time messages generated by a plurality of application scenarios, and determine the latest online features of the plurality of application scenarios based on the real-time messages. For example, the planning scheme determination device may acquire project execution progress messages of a plurality of application scenarios, and determine project completion degrees of the plurality of application scenarios based on the project execution progress messages of the plurality of application scenarios. The above is merely an exemplary illustration of a real-time message, and the real-time message may be other messages (e.g., a user question-answer message to a model, a user subscription model behavior, and a user attribute update message), which the present application does not limit in any way.

Alternatively, the planning scheme determining device may obtain the real-time messages of the plurality of application scenarios through Apache Kafka.

It can be understood that updating the online features in the feature library by the planning scheme determining device can ensure timeliness and effectiveness of the online features in the feature library, so that the planning scheme determining device can optimize the model based on the effective online features later, and can obtain better decision models of a plurality of application scenes, so that a decision model which accords with a target item better can be selected in the process of determining the second decision model, and further, the accuracy of the planning scheme is improved.

S702, the planning scheme determining device analyzes the effectiveness of the features in the feature library and determines the availability of the features in the feature library.

In an alternative implementation manner, the implementation process of S702 may be: the planning scheme determines that the equipment monitors the characteristics in the characteristic library and establishes a quality monitoring mechanism. The planning scheme determines that the equipment records the deletion rate of the features, the freshness of the features and the liveness of the features through the quality monitoring mechanism, and performs weighted summation on the deletion rate, the freshness and the liveness to obtain the availability of the features. Wherein the deletion rate of the features is used for representing the proportion of the features in the feature library to the features required in the process of constructing the model, the freshness of the feature is used to characterize the time interval between the time the feature was last used and the current time, and the liveness of the feature is used to characterize the number of times the feature was used within a preset time period.

Optionally, the planning scheme determining device controls the deletion rate, the freshness and the liveness within a preset range so as to ensure availability of features in the feature library, so that the feature library has better service quality.

The technical scheme at least brings the following beneficial effects: the planning scheme determining device performs validity analysis on the features in the feature library, determines the availability of the features in the feature library, and can select the features with higher availability in the process of selecting the offline sample features and the online sample features by the planning scheme determining device. The planning scheme determining device can improve the usability of the model by using the characteristic determining model with higher usability, so that the usability of the planning scheme is improved.

It is understood that the above-described planning scheme determination method may be implemented by the planning scheme determination device. The planning scheme determination device comprises a hardware structure and/or a software module corresponding to each function for realizing the functions. Those of skill in the art will readily appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is implemented as hardware or computer software driven hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosed embodiments.

The disclosed embodiment of the application can divide the functional modules according to the planning scheme determining device generated by the method example, for example, each functional module can be divided corresponding to each function, and two or more functions can be integrated in one processing module. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present application, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.

Fig. 8 is a schematic structural diagram of a planning scheme determining apparatus according to an embodiment of the present application. As shown in fig. 8, the plan determining apparatus 80 may be used to perform the plan determining method shown in fig. 5-7. The planning scheme determination device 80 includes: a communication unit 801 and a processing unit 802.

A communication unit 801, configured to obtain parameters of a target item and first decision models of a plurality of application scenarios; the first decision model of the plurality of application scenes is used for providing planning schemes of the plurality of projects; the plurality of items includes a target item; the parameters comprise application scenes, service requirements and service characteristics; a processing unit 802, configured to determine, based on the application scenario and/or the service requirement, a second decision model from the first decision models of the plurality of application scenarios; the second decision model is used for outputting a planning scheme of the target project; the processing unit 802 is further configured to input the business feature of the target item into the second decision model, to obtain a planning scheme of the target item.

In a possible implementation manner, the communication unit 801 is further configured to obtain online features of a plurality of application scenarios, offline features of a plurality of application scenarios, and an initial decision model of a plurality of application scenarios; the online features include project completion and/or project outcome yield; the offline feature includes at least one of: item attributes, item categories, or item scales; the processing unit 802 is further configured to match online features of the multiple application scenes with offline features of the multiple application scenes to obtain online sample features of the multiple application scenes; the processing unit 802 is further configured to optimize the initial decision models of the multiple application scenarios based on the online sample characteristics of the multiple application scenarios, so as to obtain first decision models of the multiple application scenarios.

In a possible implementation manner, the processing unit 802 is further configured to transform and combine offline features of the multiple application scenarios to determine offline sample features of the multiple application scenarios; the processing unit 802 is further configured to train the preset model based on offline sample characteristics of the multiple application scenarios, so as to obtain initial decision models of the multiple application scenarios.

In a possible implementation manner, the processing unit 802 is further configured to determine the first decision model as the second decision model when the application scenario of the first decision model is the same as the application scenario of the target item, and the output result of the first decision model meets the service requirement.

In a possible implementation manner, the processing unit 802 is further configured to determine the first decision model as the second decision model in a case where an application scenario of the first decision model is the same as an application scenario of the target item.

In a possible implementation, the processing unit 802 is further configured to store the online feature to a feature library; the processing unit 802 is further configured to perform validity analysis on the features in the feature library, and determine availability of the features in the feature library.

From the foregoing description of the embodiments, it will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of functional modules is illustrated, and in practical application, the above-described functional allocation may be implemented by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to implement all or part of the functions described above. The specific working processes of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, which are not described herein.

The present disclosure also provides a computer-readable storage medium having instructions stored thereon that, when executed by a processor of an electronic device, enable the electronic device to perform the method for determining a planning scheme provided by the embodiments of the present disclosure described above.

The disclosed embodiments also provide a computer program product containing instructions that, when run on an electronic device, cause the electronic device to perform the planning scheme determination method provided by the disclosed embodiments described above.

The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access Memory (Random Access Memory, RAM), a Read-Only Memory (ROM), an erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), a register, a hard disk, an optical fiber, a portable compact disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an application specific integrated circuit (Application Specific Integrated Circuit, ASIC). In embodiments of the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The present application is not limited to the above embodiments, and any changes or substitutions within the technical scope of the present application should be covered by the scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of determining a planning scheme, the method comprising:

acquiring parameters of a target item and first decision models of a plurality of application scenes; the first decision model of the plurality of application scenes is used for providing planning schemes of a plurality of projects; the plurality of items includes the target item; the parameters comprise application scenes, service requirements and service characteristics;

determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the business requirements; the second decision model is used for outputting a planning scheme of the target project;

and inputting the business characteristics of the target project into the second decision model to obtain a planning scheme of the target project.

2. The method of claim 1, wherein the determining a second decision model from the first decision models of the plurality of application scenarios based on the application scenarios and the business requirements comprises:

And determining the first decision model as the second decision model under the condition that the application scene of the first decision model is the same as that of the target project and the output result of the first decision model meets the service requirement.

3. The method of claim 1, wherein the determining a second decision model from the first decision models of the plurality of application scenarios based on the application scenarios comprises:

and determining the first decision model as the second decision model under the condition that the application scene of the first decision model is the same as the application scene of the target project.

4. The method of claim 3, wherein the obtaining a first decision model of the plurality of application scenarios comprises:

acquiring online characteristics of the plurality of application scenes, offline characteristics of the plurality of application scenes and initial decision models of the plurality of application scenes; the online features include project completion and/or project outcome yield; the offline feature includes at least one of: item attributes, item categories, or item scales;

matching the online characteristics of the plurality of application scenes with the offline characteristics of the plurality of application scenes to obtain online sample characteristics of the plurality of application scenes;

And optimizing the initial decision models of the plurality of application scenes based on the online sample characteristics of the plurality of application scenes to obtain first decision models of the plurality of application scenes.

5. The method of claim 4, wherein the obtaining the initial decision model for the plurality of application scenarios comprises:

transforming and combining the offline characteristics of the plurality of application scenes to determine the offline sample characteristics of the plurality of application scenes;

training a preset model based on the offline sample characteristics of the plurality of application scenes to obtain an initial decision model of the plurality of application scenes.

6. The method of claim 5, wherein the method further comprises:

storing the online features to a feature library;

and carrying out validity analysis on the features in the feature library, and determining the availability of the features in the feature library.

7. The method according to any one of claims 1-6, wherein the application scenario may comprise at least one of: the intelligent planning scene of scientific research activities, the intelligent selection of cultivation scenes by talents, the internal and external knowledge analysis application scene or the flexible infrastructure application scene.

8. A planning scheme determination device, characterized in that the device comprises: a communication unit and a processing unit;

the communication unit is used for acquiring parameters of a target item and first decision models of a plurality of application scenes; the first decision model of the plurality of application scenes is used for providing planning schemes of a plurality of projects; the plurality of items includes the target item; the parameters comprise application scenes, service requirements and service characteristics;

the processing unit is used for determining a second decision model from the first decision models of the plurality of application scenes based on the application scenes and/or the service demands; the second decision model is used for outputting a planning scheme of the target project;

the processing unit is further configured to input the business feature of the target item into the second decision model, and obtain a planning scheme of the target item.

9. A planning scheme determination apparatus, characterized by comprising: a processor and a communication interface; the communication interface is coupled to the processor for running a computer program or instructions to implement the planning scheme determination method as claimed in any one of claims 1-7.

10. A computer readable storage medium having instructions stored therein, characterized in that when executed by a computer, the computer performs the planning scheme determination method of any one of the preceding claims 1-7.