Self-adaptive calculation method and device for hierarchical training level index

Abstract

The embodiment of the invention provides a self-adaptive calculation method and device for a hierarchical training level index. The method comprises the following steps: configuring acquisition data of a last-stage index node in the hierarchical training level index; configuring a calculation model from each layer of index item to a root node in the hierarchical training level index according to a preset operator library; and according to the collected data of the last-stage index node, calculating and summarizing the calculation data of the calculation model from each layer of index item to the root node stage by stage from bottom to top to obtain an index evaluation result.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of training evaluation, and in particular, to a method and device for adaptively calculating hierarchical training level indicators. Background Art

[0002] In training fields such as sports and military, the quantification of training levels usually relies on indicator systems at different levels. By carefully breaking down, evaluating and summarizing various aspects of training, a comprehensive training level indicator is obtained.

[0003] At present, the quantification and evaluation of training level are mostly based on weight method and thousand-point method. Among them, the weight method quantifies the importance of each indicator item as a weight coefficient, and calculates the weighted average or weighted sum according to the performance of each indicator. This method relies on the experience of domain experts or historical data to determine the relative importance of different indicators. These weights are often fixed and cannot be automatically adjusted and optimized for different task requirements, different model types or dynamic changes in the training process. The thousand-point method maps all indicator items to a unified standard and then sums all items. This method simplifies the comparison between different indicator items, but ignores the differences in importance between different indicator items, resulting in information loss. In the training field, different training goals and projects may have different complexities. A single weight or thousand-point method may not accurately reflect these differences. Therefore, a more flexible and customizable aggregation method is urgently needed. Summary of the invention

[0004] The embodiments described herein provide a method and apparatus for adaptively calculating a hierarchical training level indicator, and a computer-readable storage medium storing a computer program.

[0005] According to a first aspect of the present disclosure, there is provided an adaptive calculation method for a hierarchical training level indicator, wherein the hierarchical training level indicator comprises a root node, multi-layer indicator items and a final indicator node that are logically connected from top to bottom, and the method comprises: configuring the collected data of the final indicator node in the hierarchical training level indicator; configuring a calculation model from each layer of indicator items to the root node in the hierarchical training level indicator system according to an operator library; and calculating and summarizing the calculation data of the calculation model from each layer of indicator items to the root node from bottom to top according to the collected data of the final indicator node, to obtain an indicator evaluation result.

[0006] In some embodiments of the present disclosure, configuring the collected data of the final indicator node in the hierarchical training level indicator includes: setting the upper and lower limits and data collection type of the data collection item corresponding to the final indicator node; determining whether the data value of the data collection item exceeds the preset upper and lower limits, and if so, discarding the data value; and performing data conversion and data cleaning on the collected data of heterogeneous data sources.

[0007] In some embodiments of the present disclosure, configuring a calculation model from each layer of indicator items to a root node in a hierarchical training level indicator system according to an operator library includes: configuring a calculation model from each layer of indicator items to a root node according to operator permutations and combinations in the operator library; and storing the calculation model from each layer of indicator items to a root node in a database in JSON format according to a progressive or nested relationship between indicator items at each layer.

[0008] In some embodiments of the present disclosure, the operator library includes a variety of basic operators, the basic operators include process control operators, data extraction operators, data conversion operators, common algorithm operators and user-defined operators, the process control operators include start operators, end operators, and jump operators, the common algorithm operators include statistical calculation operators, trigonometric function operators, logical calculation operators, matrix operation operators, mathematical operation operators, variance analysis operators, regression analysis operators, sensitivity analysis operators, cluster analysis operators, and fuzzy judgment operators.

[0009] In some embodiments of the present disclosure, the operator library processes the operator calculation method in a Java built-in manner.

[0010] In some embodiments of the present disclosure, based on the collected data of the last-level indicator node, the calculation data of the calculation model of each layer of indicator items to the root node are calculated and summarized from the bottom to the top to obtain the indicator evaluation result, including: identifying a specific operator calculation method based on the calculation model of each layer of indicator items to the root node; identifying a specific operator calculation method based on the calculation model of each layer of indicator items to the root node; calculating the indicator score of the bottom-level indicator item according to the operator calculation method based on the collected data of the last-level indicator node, and for each intermediate indicator item, summarizing the indicator score according to the hierarchical relationship between the indicator items until the indicator evaluation result of the root node is summarized.

[0011] In some embodiments of the present disclosure, based on the collected data of the last-level indicator node, the calculation data of the calculation model of the indicator items of each layer to the root node are calculated and summarized from the bottom to the top, and the indicator evaluation result also includes: adjusting the calculation model of the indicator items or the relationship between the indicator items according to the historical scores of the indicators at each level; deploying the adjusted calculation model and hierarchical training level indicators to actual applications, and calculating the scores of the indicators at each level in real time based on the adjusted calculation model.

[0012] In some embodiments of the present disclosure, the evaluation results of each indicator item can be displayed in a hierarchical manner through an interactive graphical interface by means of graphical drill-down, and the interactive graphical interface includes charts, heat maps, and tree diagrams; and anomalies in the indicator calculation process can be identified. If the calculation result of an indicator exceeds a preset threshold, the anomaly is automatically marked and an anomaly warning is triggered.

[0013] According to the second aspect of the present disclosure, an adaptive computing device for hierarchical training level indicators is provided. The device includes at least one processor; and at least one memory storing a computer program. When the computer program is executed by at least one processor, the device: configures the collected data of the last-level indicator node in the hierarchical training level indicator; configures the calculation model of each layer of indicator items to the root node in the hierarchical training level indicator according to a preset operator library; and calculates and summarizes the calculation data of the calculation model of each layer of indicator items to the root node from the bottom to the top according to the collected data of the last-level indicator node, and obtains the indicator evaluation result.

[0014] In some embodiments of the present disclosure, when the computer program is executed by at least one processor, the device configures the collected data of the final indicator node in the hierarchical training level indicator through the following operations: setting the upper and lower limits and data collection type of the data collection item corresponding to the final indicator node; determining whether the data value of the data collection item exceeds the preset upper and lower limits, and if so, eliminating the data value; and performing data conversion and data cleaning on the collected data of heterogeneous data sources.

[0015] In some embodiments of the present disclosure, when the computer program is executed by at least one processor, the device configures the calculation model of each layer of indicator items to the root node in the hierarchical training level indicator system according to the operator library through the following operations: configuring the calculation model for each layer of indicator items to the root node according to the operator permutations and combinations in the operator library; and storing the calculation model of each layer of indicator items to the root node in a database in JSON format according to the progressive or nested relationship between the indicator items of each layer.

[0016] In some embodiments of the present disclosure, when the computer program is executed by at least one processor, the device calculates and summarizes the calculation data of the calculation model of each layer of indicator items to the root node from the bottom to the top through the following operations to obtain an indicator evaluation result: according to the calculation model of each layer of indicator items to the root node, identify the specific operator calculation method; according to the operator calculation method, calculate the indicator score of the lowest level indicator item according to the collected data of the last level indicator node, and for each intermediate indicator item, summarize the indicator score according to the hierarchical relationship between the indicator items until the indicator evaluation result of the root node is summarized.

[0017] In some embodiments of the present disclosure, when the computer program is executed by at least one processor, the device obtains the indicator evaluation result through the following operations: adjusting the calculation model of the indicator items or the relationship between the indicator items according to the historical scores of the indicators at each level; deploying the adjusted calculation model and hierarchical training level indicators to actual applications, and calculating the scores of the indicators at each level in real time based on the adjusted calculation model.

[0018] In some embodiments of the present disclosure, when the computer program is executed by at least one processor, the device: displays the evaluation results of each indicator item in a hierarchical manner using an interactive graphical interface through graphical drill-down, and the interactive graphical interface includes charts, heat maps, and tree diagrams; and identifies anomalies in the indicator calculation process. If the calculation result of an indicator exceeds a preset threshold, the anomaly is automatically marked and an anomaly warning is triggered.

[0019] According to a third aspect of the present disclosure, a computer-readable storage medium storing a computer program is provided, wherein the computer program, when executed by a processor, implements the steps of the method for adaptively calculating a hierarchical training level indicator according to the first aspect of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly described below. It should be noted that the drawings described below only relate to some embodiments of the present disclosure, but are not intended to limit the present disclosure, wherein:

[0021] Figure 1 It is a schematic diagram of the architecture of hierarchical training level indicators;

[0022] Figure 2 An exemplary flow chart showing a method 200 for adaptively calculating a hierarchical training level indicator according to an embodiment of the present disclosure;

[0023] Figure 3 is a schematic block diagram of an adaptive calculation device for hierarchical training level indicators according to an embodiment of the present disclosure.

[0024] It should be noted that the elements in the drawings are schematic and not drawn to scale. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work also fall within the scope of protection of the present disclosure.

[0026] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which the subject matter of the present disclosure belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meanings in the context of the specification and the relevant art, and will not be interpreted in an idealized or overly formal form unless otherwise explicitly defined herein. In addition, terms such as "first" and "second" are only used to distinguish one component (or a portion of a component) from another component (or another portion of a component).

[0027] This solution proposes to solve the problems of single convergence algorithm and insufficient adaptability in training level quantization by configuring the computing model in a hierarchical manner and combining it with a flexible and configurable adaptive computing method. Figure 1 It is a schematic diagram of the architecture of hierarchical training level indicators. Figure 1 In the example, the hierarchical training level indicator includes the root node, first-level indicator items, second-level indicator items...N-level indicator items and the last-level indicator node, which are logically connected from top to bottom. Among them, the root node represents the top level of the entire indicator system, reflecting the final evaluation result of the training level. The first-level indicator items, second-level indicator items...N-level indicator items represent detailed indicators, such as "skill mastery level", "training effect", "physical fitness", etc. The last-level indicator item node includes specific data items such as "completion of technical movements", "reaction speed", and "strength improvement". The design of the hierarchical indicator system is usually closely related to business needs. The progressive and nested relationship of the business makes the indicators at all levels not only have their own meanings, but also form connections with other indicators.

[0028] In order to solve the problem of quantitative calculation of training level indicators, this paper proposes an adaptive calculation method for hierarchical training level indicators. Through flexible calculation model algorithm configuration, it breaks through the limitation of traditional methods that can only adopt a single indicator item aggregation algorithm, and enables indicators at different levels to use the most appropriate algorithm according to the characteristics of the task. Figure 2 An exemplary flow chart of a method 200 for adaptively calculating a hierarchical training level indicator according to an embodiment of the present disclosure is shown.

[0029] exist Figure 2 In block S202, the collected data of the last level indicator node in the hierarchical training level indicator is configured.

[0030] In one embodiment of the present disclosure, the structure of the constructed hierarchical training level indicators, the hierarchical relationship of the indicators, version information, etc. are stored in a database. For example, for the indicator items at each level, the indicator name, indicator description, data source, unit, weight setting, etc. are defined. By saving the hierarchical structure of indicators in the database, users can easily create, edit, and delete the indicator system through the interface. Whenever the indicator changes, a new version is generated and the detailed information of the change is recorded. The indicator system can be displayed in a tree structure on the front end, and version management and change logs can be displayed through different views. For example, the tree structure displays the hierarchical structure of the indicator, and addition, deletion, and modification operations can be performed; in the version management view, users can view the historical versions of a certain indicator and the related change logs; the version comparison view displays the changes between different versions through difference comparison.

[0031] In order to configure the data collection of the indicator node, you can set the upper and lower limits of the data collection items and the data collection type (real-time, scheduled, batch) corresponding to the final indicator node. By setting a reasonable range of data, ensure that the collected data does not deviate from the actual situation. For example, the collection of training time can be set with an upper limit of 24 hours and a lower limit of 0 hours. Different types of training data may have different collection methods. For example, collection through automated equipment (such as sensors) or collection through API interfaces. The data collection items will perform necessary verification and standardization during the collection process. For example, it will determine whether the data value of the data collection item exceeds the preset upper and lower limits. If it exceeds the upper and lower limits, the data value will be eliminated to avoid data errors or inconsistencies. For scheduled collection, the system will obtain data regularly based on the configured time interval; for real-time collection, the system needs to continuously obtain and verify data during the data collection process.

[0032] In order to support the compatibility of multiple data sources, data sources can be configured through JDBC to obtain data in the form of views. Furthermore, data conversion and data cleaning are performed on the collected data from heterogeneous data sources. Data conversion includes unifying data from different formats or protocols into a standard format that the system can handle. Different conversions can be made according to actual conditions, such as date format conversion, unit conversion, etc. Data cleaning includes removing invalid data, processing missing data, processing abnormal data, etc. For example, if a data field is empty or the value is negative, it may be necessary to clean or fill in these data.

[0033] Reference Figure 2 As shown, at box S204, the calculation model of each layer indicator item to the root node in the hierarchical training level indicator is configured according to the preset operator library.

[0034] Each indicator item can select a different algorithm model, and the calculation model is configured for each indicator item according to the operator combination in the operator library. According to the progressive or nested relationship between the indicator items at each layer, the calculation model from the indicator items at each layer to the root node is stored in the database in JSON format.

[0035] The operator library provides a variety of basic operators and user-defined operators, each of which represents an independent operation. For example, basic operators include process control operators, data extraction operators, data conversion operators, and common algorithm operators. Among them, process control operators are process control-related operators that represent the start, conditional jump, and completion of evaluation index calculations, including start operators, end operators, and jump operators. Data extraction operators are operators that represent the extraction of data from a variety of heterogeneous data sources according to different needs. Data conversion operators are operators that represent the conversion, calculation, combination, and splitting of data tables. Common algorithm operators represent operations that perform various mathematical calculations on data. For example, common algorithm operators include statistical calculation operators, trigonometric function operators, logical calculation operators, matrix operation operators, mathematical operation operators, variance analysis operators, regression analysis operators, sensitivity analysis operators, cluster analysis operators, fuzzy judgment operators, etc. User-defined operators are user-defined formulas or operation expressions. A Java built-in expression parser (such as Java ScriptEngine or ANTLR) may be used to parse and execute the user-defined formula.

[0036] Operators are matched and connected through directed lines. One end of the line is the export operator and the other end is the import operator. The data output interface parameters of the export operator and the data input interface parameters of the import operator are obtained to determine whether the formats of the two sets of interface parameters match. If they match, the two operators are connected through the line. If they do not match, a prompt that the connection cannot be made is given. The operator library implements the calculation methods of these operators through Java built-in methods. Each operator can be encapsulated into an interface or abstract class, and the specific operator implementation inherits the interface or abstract class.

[0037] Each level of indicator items needs to be configured with a calculation model, and these models are passed layer by layer through the parent-child relationship between nodes.

[0038] Finally Figure 2 In block S206, based on the collected data of the last-level indicator node, the calculation data of the calculation model of the indicator items of each layer to the root node are calculated and summarized from bottom to top to obtain the indicator evaluation result.

[0039] The bottom-up step-by-step calculation model decomposes the complex indicator system into multiple subtasks through a hierarchical structure. The calculation results of each subtask provide input for the upper-level task, and finally calculate the evaluation result of the root node. Specifically, according to the calculation model of each indicator item, the specific calculation method is identified. According to the operator calculation method, the indicator score of the lowest-level indicator item is calculated based on the collected data of the last-level indicator node. For each intermediate indicator item, the indicator score is summarized according to the hierarchical relationship between the indicator items until the indicator evaluation result of the root node is summarized.

[0040] For indicators with complex relationships, their interdependence needs to be considered during the calculation process. For example, an indicator at one level may depend on the calculation results of multiple sub-indicators. At this time, it is necessary to calculate according to the hierarchical relationship of the indicator nodes to ensure that the scores of all related indicator items can correctly reflect the overall score. If the score of an indicator is a weighted average of multiple sub-indicators, the system needs to weight and summarize the sub-indicator scores according to the configured weights. Some calculation models may be more complex and may include regression analysis, weighted scoring, or even prediction models based on machine learning. In this case, the system will apply the corresponding calculation method to obtain the final indicator score. The core of this process is to ensure that the data at each layer is accurately transmitted and reasonably weighted and synthesized at each layer, and finally achieve the purpose of comprehensive evaluation.

[0041] In some embodiments of the present disclosure, the evaluation results of each indicator item can be displayed in a hierarchical manner by means of graphical drill-down. For example, interactive graphical interfaces such as charts, heat maps, tree diagrams, etc. are used to display the scores of these levels and indicators. Clicking on an item in the chart will automatically expand the sub-indicators or details of the item, and display its calculation method and data source. Users can drill down from the overall indicator to the specific data of each sub-indicator, and view it layer by layer, even to the most detailed original data. Based on the filtering conditions and selection items, users can select the indicator items of interest for viewing.

[0042] Furthermore, reasonable thresholds can be set for each calculation link to identify anomalies in the indicator calculation process. If the calculation result of an indicator exceeds the preset threshold, the anomaly is automatically marked and an abnormal warning is triggered. As data changes or new data is loaded, the chart can be dynamically updated to ensure that the latest evaluation results are reflected in real time.

[0043] In one embodiment of the present disclosure, the calculation model of the indicator items or the relationship between the indicator items are adjusted according to the historical scores of the indicators at each level; the adjusted calculation model and the hierarchical training level indicators are deployed to practical applications, and the scores of the indicators at each level are calculated in real time based on the adjusted calculation model.

[0044] Specifically, according to the historical scores and change trends of indicators at each level, the relationship and influencing factors between different indicators are identified. For example, through statistical methods such as Pearson correlation coefficient and regression analysis, the relationship between different indicators is identified to determine which indicators have a significant impact on other indicators. The trend of real-time data is compared with that of historical data, and the calculation model of indicator items or the relationship between levels is adjusted according to the comparison results. This process ensures that the indicator system can adjust itself according to changes in the external environment and internal data, thereby ensuring the effectiveness and accuracy of the evaluation system in a dynamic environment.

[0045] Figure 3 is a schematic block diagram of an adaptive computing device for hierarchical training level indicators according to an embodiment of the present disclosure. Figure 3 As shown, the device 300 may include a processor 310 and a memory 320 storing a computer program. When the computer program is executed by the processor 310, the device 300 may perform the following operations: Figure 2 The steps of the method 200 for adaptively calculating the hierarchical training level indicator are shown. In one example, the apparatus 300 may be a computer device or a cloud computing node.

[0046] In one embodiment of the present disclosure, the processor 310 may include an indicator management module, an operator library module, an algorithm configuration module, and an indicator calculation module. Among them, the indicator management module is used to create and manage hierarchical training level indicators and manage the indicator hierarchy; the operator library module is used to provide a basic operator library, support various operation operations, and build an algorithm model; the algorithm configuration module is used to provide a configuration interface and build a calculation model through the operators in the operator library. The indicator calculation module is used to calculate the indicator evaluation results through data collection and configured algorithm models.

[0047] Specifically, the indicator management module allows users to design and create hierarchical training level indicators according to their needs. These hierarchical indicator structures are managed by the database and stored as tree structures or relational tables. According to one embodiment of the present disclosure, the indicator system supports version management. Each time an indicator is modified, a new version is automatically created to record the changes in the indicator in different versions.

[0048] The operator library provides a variety of basic operators, each of which represents an independent calculation operation. For example, basic processing operators include process control operators, data extraction operators, data conversion operators, common algorithm operators, and user-defined operators. Among them, process control operators are process control-related operators that represent the start, conditional jump, and completion of evaluation index calculations. Data extraction operators are operators that extract data from a variety of heterogeneous data sources according to different needs. Data conversion operators are operators that represent the conversion, calculation, combination, and splitting of data tables. Common algorithm operators represent operations that perform various mathematical calculations on data. For example, common algorithm operators include statistical calculation operators, matrix operation operators, mathematical operation operators, variance analysis operators, regression analysis operators, sensitivity analysis operators, cluster analysis operators, fuzzy judgment operators, etc. User-defined operators are operators that provide a general code framework and interface, and users can directly write special operator processing codes according to their needs.

[0049] Operators are matched and connected through directed connecting lines. One end of the connecting line is the export operator and the other end is the import operator. The data output interface parameters of the export operator and the data input interface parameters of the import operator are obtained to determine whether the formats of the two sets of interface parameters match. If they match, the two operators are connected through the connecting line. If they do not match, a prompt that the connection cannot be made is given. The indicator configuration module can complete a complex calculation model by combining multiple operators. Users can make detailed settings for the input parameters, output parameters, calculation order, etc. of each operator in the configuration.

[0050] According to an embodiment of the present disclosure, the device 300 can configure the collected data of the last-level indicator node in the hierarchical training level indicator; configure the calculation model of the indicator items of each layer to the root node in the hierarchical training level indicator according to a preset operator library; and calculate and summarize the calculation data of the calculation model of the indicator items of each layer to the root node from the bottom to the top according to the collected data of the last-level indicator node to obtain the indicator evaluation result.

[0051] In some embodiments of the present disclosure, the device 300 can set the upper and lower limits and data collection type of the data collection item corresponding to the final-level indicator node; determine whether the data value of the data collection item exceeds the preset upper and lower limits, and if so, eliminate the data value; and perform data conversion and data cleaning on the collected data from heterogeneous data sources.

[0052] In some embodiments of the present disclosure, the device 300 can configure a calculation model for each layer of indicator items to the root node according to the operator arrangement and combination in the operator library; according to the progressive or nested relationship between the indicator items of each layer, the calculation model of the indicator items of each layer to the root node is stored in the database in JSON format. Among them, the operator library includes a variety of basic operators, the basic operators include process control operators, data extraction operators, data conversion operators, common algorithm operators and user-defined operators, the process control operators include start operators, end operators, jump operators, and common algorithm operators include statistical calculation operators, trigonometric function operators, logical calculation operators, matrix operation operators, mathematical operation operators, variance analysis operators, regression analysis operators, sensitivity analysis operators, cluster analysis operators, and fuzzy judgment operators. The operator library uses Java built-in methods to process operator calculation methods.

[0053] In some embodiments of the present disclosure, the device 300 can identify a specific operator calculation method based on a calculation model from each layer of indicator items to a root node; according to the operator calculation method, the indicator score of the lowest level indicator item is calculated based on the collected data of the last level indicator node, and for each intermediate indicator item, the indicator score is summarized according to the hierarchical relationship between the indicator items until the indicator evaluation result of the root node is obtained.

[0054] Furthermore, the device 300 can adjust the calculation model of the indicator items or the relationship between the indicator items according to the historical scores of the indicators at each level; and deploy the adjusted calculation model and hierarchical training level indicators into practical applications, and calculate the scores of the indicators at each level in real time based on the adjusted calculation model.

[0055] In some embodiments of the present disclosure, the device 300 can display the evaluation results of each indicator item in a hierarchical manner using an interactive graphical interface through graphical drill-down, and the interactive graphical interface includes charts, heat maps, and tree diagrams; and identify anomalies in the indicator calculation process. If the calculation result of an indicator exceeds a preset threshold, the anomaly is automatically marked and an anomaly warning is triggered.

[0056] In an embodiment of the present disclosure, the processor 310 may be, for example, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a processor based on a multi-core processor architecture, etc. The memory 320 may be any type of memory implemented using data storage technology, including but not limited to random access memory, read-only memory, semiconductor-based memory, flash memory, disk storage, etc.

[0057] In addition, in the embodiment of the present disclosure, the apparatus 300 may also include an input device 330, such as a keyboard, a mouse, etc., for inputting the collected data of the indicator node. In addition, the apparatus 300 may also include an output device 340, such as a display, etc., for outputting the evaluation results of each indicator item and the root node.

[0058] In other embodiments of the present disclosure, a computer-readable storage medium storing a computer program is further provided, wherein the computer program can achieve the following when executed by a processor: Figure 2 The steps of the adaptive calculation method of the hierarchical training level indicator are shown.

[0059] In summary, according to the adaptive calculation method and device of the hierarchical training level indicator of the embodiment of the present disclosure, by flexibly configuring the algorithm model by level, combined with the user-defined operator library and the real-time exception capture mechanism, it is possible to more accurately quantify the training level and provide more powerful support for military training management and decision-making. In addition, this calculation method is easy to expand new levels or indicators. It only needs to add new calculation methods to the operator library or update the data collection method, and the entire calculation framework can adapt to the new content, avoiding the high cost of frequent reconstruction of the calculation model in the traditional method. The high adaptability and scalability of the system also provide sufficient flexibility for possible changes in the future, and can cope with changing training needs.

[0060] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the device and method according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, a program segment or an instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0061] Unless the context clearly indicates otherwise, the singular form of the words used herein and in the appended claims includes the plural and vice versa. Thus, when referring to the singular, the plural form of the corresponding term is generally included. Similarly, the words "comprise" and "include" are to be interpreted as inclusive rather than exclusive. Likewise, the terms "include" and "or" should be interpreted as inclusive unless such interpretation is expressly prohibited herein. Where the term "example" is used herein, particularly when it is located after a group of terms, the "example" is merely exemplary and illustrative and should not be considered exclusive or comprehensive.

[0062] Further aspects and scopes of adaptability become apparent from the description provided herein. It should be understood that various aspects of the present application can be implemented individually or in combination with one or more other aspects. It should also be understood that the description and specific embodiments herein are intended for purposes of illustration only and are not intended to limit the scope of the present application.

[0063] Several embodiments of the present disclosure are described in detail above, but it is obvious that those skilled in the art can make various modifications and variations to the embodiments of the present disclosure without departing from the spirit and scope of the present disclosure. The protection scope of the present disclosure is defined by the attached claims.

Claims

1. An adaptive calculation method for a hierarchical training level indicator, wherein the hierarchical training level indicator comprises a root node, multi-layer indicator items and a final indicator node logically connected from top to bottom, characterized in that: The method comprises: Configure the collected data of the last indicator node in the hierarchical training level indicator; According to the preset operator library, a calculation model of each layer indicator item in the hierarchical training level indicator to the root node is configured; and According to the collected data of the last-level indicator node, the calculation data of the calculation model of the indicator items of each layer to the root node are calculated and summarized from bottom to top to obtain the indicator evaluation result.

2. The adaptive calculation method of hierarchical training level index according to claim 1, characterized in that: The collected data of the last level indicator node in the configuration of the hierarchical training level indicator includes: Set the upper and lower limits and data collection type of the data collection items corresponding to the final indicator node; Determine whether the data value of the data collection item exceeds the preset upper and lower limits, and if so, remove the data value; and Perform data conversion and data cleaning on the collected data from heterogeneous data sources.

3. The adaptive calculation method of hierarchical training level index according to claim 1, characterized in that: The configuration of the calculation model of each layer index item in the hierarchical training level index to the root node according to the preset operator library includes: Configure the calculation model for each layer of indicator items to the root node according to the operator arrangement and combination in the operator library; and According to the progressive or nested relationship between the indicator items at each layer, the calculation model from the indicator items at each layer to the root node is stored in the database in JSON format.

4. The adaptive calculation method of hierarchical training level index according to claim 3 is characterized in that: The operator library includes a variety of basic operators, including process control operators, data extraction operators, data conversion operators, common algorithm operators and user-defined operators. The process control operators include start operators, end operators and jump operators. The common algorithm operators include statistical calculation operators, trigonometric function operators, logical calculation operators, matrix operation operators, mathematical operation operators, variance analysis operators, regression analysis operators, sensitivity analysis operators, cluster analysis operators and fuzzy judgment operators.

5. The adaptive calculation method of hierarchical training level index according to claim 3, characterized in that: The operator library processes operator calculation methods in a Java built-in manner.

6. The adaptive calculation method of hierarchical training level index according to claim 3, characterized in that: The step of calculating and aggregating the calculation data of the calculation model of each layer of indicator items to the root node from bottom to top based on the collected data of the last-level indicator node to obtain the indicator evaluation result includes: According to the calculation model from each layer of indicators to the root node, identify the specific operator calculation method; According to the operator calculation method, the index score of the bottom-level index item is calculated based on the collected data of the last-level index node, and for each intermediate index item, the index score is summarized according to the hierarchical relationship between the index items until the index evaluation result of the root node is obtained.

7. The adaptive calculation method of hierarchical training level index according to claim 6, characterized in that: The step of calculating and aggregating the calculation data of the calculation model of each layer of indicator items to the root node from bottom to top based on the collected data of the last-level indicator node to obtain the indicator evaluation result also includes: Adjust the calculation model of the indicator items or the relationship between the indicator items according to the historical scores of the indicators at each level; and The adjusted computing model and hierarchical training level indicators are deployed to practical applications, and the scores of indicators at each level are calculated in real time based on the adjusted computing model.

8. The adaptive calculation method of hierarchical training level index according to claim 1, characterized in that: The method further comprises: By means of graphical drill-down, the evaluation results of each indicator item are displayed hierarchically using an interactive graphical interface, wherein the interactive graphical interface includes a chart, a heat map, and a tree map; and Identify anomalies in the indicator calculation process. If the calculation result of an indicator exceeds the preset threshold, the anomaly will be automatically marked and an anomaly warning will be triggered.

9. An adaptive calculation device for hierarchical training level indicators, characterized in that: The device comprises: at least one processor; and at least one memory storing a computer program; Wherein, when the computer program is executed by the at least one processor, the device executes the steps of the adaptive calculation method of the hierarchical training level indicator according to any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the computer program implements the steps of the method for adaptively calculating a hierarchical training level indicator according to any one of claims 1 to 8.

Images