Military evaluation index construction and evaluation method and device based on deep learning

Abstract

The invention discloses a military assessment index construction and assessment method and device based on deep learning. The method comprises the steps of obtaining sample military data, performing feature extraction processing on the sample military data based on an expert marking method and machine learning, and constructing a military evaluation index database; constructing a first military assessment index system based on a rule template for the military assessment index database to obtain a first military assessment index system; model training processing based on a deep learning model is carried out on the military assessment index database to obtain a second military assessment index system, and the second military assessment index system is used for representing a military assessment model obtained through model training processing; and performing joint decision-making military assessment processing based on the first military assessment index system and the second military assessment index system on the to-be-assessed military data to obtain assessment result data. The technical effect of improving the accuracy of military assessment prediction and the interpretability of assessment indexes is achieved.

Description

Technical Field

[0001] The present application relates to the field of military combat assessment, and specifically to a method and device for constructing military assessment indicators and conducting assessments based on deep learning. Background Art

[0002] Modern warfare environments are increasingly complex and volatile. Traditional combat assessment techniques based on statistical, analytical, simulation, and comprehensive methods struggle to fully account for the systemic interconnections between combat elements and the dynamic and random nature of confrontations. The inventors discovered that by introducing artificial intelligence technology, particularly deep learning, the scientific nature and credibility of assessment metrics can be improved. However, relying solely on data-driven methods may not fully capture all changes in battlefield conditions, while rule-based methods lack flexibility.

[0003] Therefore, this application is filed to address the above problems. Summary of the Invention

[0004] The main purpose of this application is to provide a military evaluation index construction and evaluation method and device based on deep learning, so as to achieve the technical effect of improving the accuracy of military evaluation prediction and the interpretability of evaluation indicators.

[0005] To achieve the above objectives, in the first aspect of this application, a military evaluation index construction and evaluation method based on deep learning is proposed, including:

[0006] Acquire sample military data, perform feature extraction processing on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database;

[0007] Constructing a first military evaluation index system based on a rule template on the military evaluation index database to obtain a first military evaluation index system;

[0008] Performing a model training process based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained by the model training process;

[0009] The military data to be evaluated is subjected to joint decision-making military evaluation processing based on the first military evaluation index system and the second military evaluation index system to obtain evaluation result data.

[0010] Furthermore, the sample military data is subjected to feature extraction processing based on expert labeling and machine learning to construct a military evaluation index database, including:

[0011] Performing feature extraction processing based on data statistics and machine learning on the sample military data to obtain associated feature data, wherein the associated feature data is feature data used to represent the correlation relationship between military evaluation indicators;

[0012] Performing feature extraction processing on the sample military data based on preset customized military features to obtain military customized feature data;

[0013] Performing feature mapping processing on the sample military data based on a pre-trained language model to obtain semantic feature data, wherein the semantic feature data is feature data obtained by extracting features from language text in the sample military data;

[0014] The military evaluation index database is constructed based on the associated feature data, the military customized feature data and the semantic feature data.

[0015] Furthermore, the military data to be evaluated is subjected to joint decision-making military evaluation processing based on the first military evaluation index system and the second military evaluation index system, and the evaluation result data obtained includes:

[0016] Performing military evaluation processing on the military data to be evaluated based on the first military evaluation indicator system to obtain first process evaluation result data;

[0017] Performing military evaluation processing on the military data to be evaluated based on the second military evaluation indicator system to obtain second process evaluation result data;

[0018] The first process evaluation result data and the second process evaluation result data are jointly processed for decision-making to obtain the evaluation result data.

[0019] Furthermore, the first process evaluation result data and the second process evaluation result data are jointly processed for decision making, and the evaluation result data obtained includes:

[0020] Performing model prediction confidence feature extraction processing on the second process evaluation result data to obtain model confidence feature data, wherein the model confidence feature data is feature data used to represent the model prediction confidence;

[0021] If the model confidence feature data satisfies the preset confidence feature rules, obtaining the evaluation result data to be fused, the evaluation result data to be fused includes the first process evaluation result data and the second process evaluation result data, and performing fusion processing on the evaluation data to be fused based on the fusion rules to obtain the evaluation result data;

[0022] If the model confidence feature data satisfies the preset confidence feature rules, the evaluation result data is obtained, and the evaluation result data includes the first process evaluation result data.

[0023] Furthermore, it also includes:

[0024] Performing rule anomaly trigger detection on the first military evaluation indicator system to obtain rule anomaly trigger data;

[0025] Performing extraction processing on the rule abnormality triggering data based on time features and rule mapping features to obtain abnormal time feature data and abnormal rule mapping feature data, wherein the abnormal rule mapping feature data is feature data used to indicate abnormality in the trigger rule mapping relationship;

[0026] If the abnormal time feature data satisfies a preset dynamic update rule, the rule template of the first military evaluation index system is updated according to the abnormal rule mapping feature data to obtain an updated first military evaluation index system.

[0027] Furthermore, it also includes:

[0028] Acquire updated sample military data, perform incremental feature extraction on the sample military data, and obtain incremental feature data;

[0029] performing feature extraction processing based on the association relationship on the incremental feature data and the first military evaluation index system to obtain updated association feature data;

[0030] A model update process is performed on the military evaluation model of the second military evaluation index system according to the updated associated feature data to obtain an updated second military evaluation index system.

[0031] Performing model training processing based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system includes:

[0032] Furthermore, the Transformer is used as the framework of the deep learning model, including multiple stacked layers of the same structure, each layer including two main sublayers: a multi-head self-attention mechanism and a feedforward neural network; a two-layer linear transformation using the ReLU activation function is used as the feedforward neural network part, cross entropy is used as the loss function, and Xavier initialization and orthogonal initialization are used to initialize the parameters of the deep learning model to obtain a military evaluation model through training.

[0033] According to the second aspect of this application, a military evaluation index construction and evaluation device based on deep learning is proposed, including:

[0034] A feature extraction module is used to obtain sample military data, perform feature extraction processing on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database;

[0035] A first military evaluation index module is used to construct a first military evaluation index system based on a rule template on the military evaluation index database to obtain a first military evaluation index system;

[0036] a second military evaluation index module, configured to perform model training processing based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained through the model training processing;

[0037] The military evaluation module is used to perform joint decision-making military evaluation processing on the military data to be evaluated based on the first military evaluation index system and the second military evaluation index system to obtain the evaluation result data.

[0038] According to the third aspect of the present application, a computer-readable storage medium is proposed, which stores computer instructions, and the computer instructions are used to enable the computer to execute the above-mentioned deep learning-based military evaluation index construction and evaluation method.

[0039] According to the fourth aspect of the present application, an electronic device is proposed, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor executes the above-mentioned deep learning-based military evaluation indicator construction and evaluation method.

[0040] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0041] In this application, sample military data is obtained, and feature extraction processing based on expert labeling and machine learning is performed on the sample military data to construct a military evaluation index database; a first military evaluation index system based on a rule template is constructed on the military evaluation index database to obtain a first military evaluation index system; a model training processing based on a deep learning model is performed on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained by the model training processing; and a joint decision-making military evaluation processing based on the first military evaluation index system and the second military evaluation index system is performed on the military data to be evaluated to obtain evaluation result data. By integrating deep learning with rule templates to construct and evaluate military evaluation indicators, the technical effect of improving the accuracy of military evaluation predictions and the interpretability of evaluation indicators is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings that constitute part of this application are used to provide a further understanding of this application and make other features, objects and advantages of this application more apparent. The illustrative embodiment drawings of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0043] Figure 1 A flowchart of a deep learning-based military evaluation index construction and evaluation method provided in this application;

[0044] Figure 2 A flowchart of a deep learning-based military evaluation index construction and evaluation method provided in this application;

[0045] Figure 3 A schematic diagram of a deep learning-based military evaluation index construction and evaluation device provided in this application. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] In this application, terms such as "upper," "lower," "left," "right," "front," "back," "top," "bottom," "inner," "outer," "center," "vertical," "horizontal," "transverse," and "longitudinal" indicate positions or locations based on the positions or locations shown in the accompanying drawings. These terms are primarily intended to better describe this application and its embodiments and are not intended to limit the devices, elements, or components indicated to having a specific orientation, or to being constructed or operated in a specific orientation.

[0049] Furthermore, some of the above terms may be used to express other meanings besides indicating a position or location. For example, the term "on" may also be used to indicate a dependency or connection in certain circumstances. Those skilled in the art will understand the specific meanings of these terms in this application based on the specific circumstances.

[0050] Furthermore, the terms "installed," "disposed," "provided with," "connected," "connected," and "socketed" should be interpreted broadly. For example, "connected" can mean a fixed connection, a removable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection, an indirect connection through an intermediary, or an internal communication between two devices, elements, or components. Those skilled in the art will understand the specific meanings of these terms in this application based on the specific circumstances.

[0051] In the field of military assessment, traditional statistical methods rely on manual feature engineering and struggle to capture complex nonlinear relationships. Pure deep learning models, while capable of automatically extracting features, lack transparency and interpretability. This application proposes a deep learning-based military assessment indicator construction and evaluation method. By integrating deep learning with rule-based templates, this method achieves the technical effect of improving the accuracy of military assessment predictions and the interpretability of evaluation indicators.

[0052] In an optional embodiment of the present application, a military evaluation index construction and evaluation method based on deep learning is proposed. Figure 1A flowchart of a deep learning-based military evaluation index construction and evaluation method provided in this application, such as Figure 1 As shown, the method includes the following steps:

[0053] S101: Obtain sample military data, perform feature extraction on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database;

[0054] Sample military data is collected from various data sources, such as databases and various APIs. This includes various types of military data, including military reconnaissance data, military intelligence data, military operations data, military training data, military research data, military technical standards data, military regulations and guidelines data, and other military-related data. Military data obtained from multiple data sources includes unstructured and / or structured data, with varying data formats and quality levels. Data preprocessing is performed on these data sources to filter out abnormal data and obtain data that is convenient for subsequent rule design and model training. Data preprocessing includes: data cleaning, which involves removing duplicate records, filling in missing values, and correcting incorrect entries; and data labeling, which involves labeling the cleaned data using expert labeling methods to obtain important features annotated by military experts and define target variables. For example, when analyzing enemy troop movement patterns, information such as the time, location, and participants of specific tactical actions is labeled.

[0055] In an optional embodiment of the present application, a military evaluation index construction and evaluation method based on deep learning is proposed, including:

[0056] The sample military data is subjected to feature extraction processing based on data statistics and machine learning to obtain associated feature data, which is feature data used to represent the association relationship between military evaluation indicators; the sample military data is subjected to feature extraction processing based on preset customized military features to obtain military customized feature data; the sample military data is subjected to feature mapping processing based on a pre-trained language model to obtain semantic feature data, wherein the semantic feature data is feature data used to represent feature data obtained by feature extraction of language text in the sample military data; and a military evaluation indicator database is constructed based on the associated feature data, the military customized feature data and the semantic feature data.

[0057] In an optional embodiment of the present application, the above-mentioned cleaned data is subjected to feature extraction and conversion based on feature engineering to obtain feature data for model training, including: analyzing the original data by statistical methods to obtain correlation features between military features; analyzing the original data by machine learning algorithms to obtain feature data for representing the correlation relationship between data, the original data can be analyzed by principal component analysis (PCA), or the original data can be analyzed and processed by t-SNE dimensionality reduction to obtain correlation feature data for representing the correlation relationship of military evaluation indicators. Due to the uniqueness of military tasks, features are customized according to military tasks, and data corresponding to customized military features in the original data are extracted according to the customized features, such as spatial distribution features based on geographic information system GIS or time series features reflecting the status of troops; for unstructured natural language text, the BERT pre-trained language model is used to map it to a high-dimensional vector space, retaining semantic information, and facilitating subsequent operations with other types of data to obtain semantic feature data; a military evaluation index database is constructed based on the correlation feature data, military customized feature data, and semantic feature data.

[0058] S102: constructing a first military evaluation indicator system based on a rule template in the military evaluation indicator database to obtain a first military evaluation indicator system;

[0059] Constructing a rule template involves integrating military domain knowledge to obtain rule data; performing rule definition processing on the rule data to define the rule template format, and constructing the rule template. Integrating military domain knowledge includes obtaining expert opinions; extracting military research data, technical standards data, military regulations data, and military guidance data; integrating the collected expert opinions with the aforementioned military data to obtain rule data; and performing rule template definition processing on the integrated rule data. For example, a rule template is defined based on the IF-THEN format, i.e., "If a condition is met, then perform a certain action."

[0060] In some optional embodiments of the present application, a method for constructing and evaluating military evaluation indicators based on deep learning is proposed, which also includes:

[0061] A rule anomaly trigger detection is performed on the first military evaluation index system to obtain rule anomaly trigger data; the rule anomaly trigger data is extracted and processed based on time features and rule mapping features to obtain abnormal time feature data and abnormal rule mapping feature data, wherein the abnormal rule mapping feature data is feature data used to indicate abnormal trigger rule mapping relationships; if the abnormal time feature data meets the preset dynamic update rules, the rule template of the first military evaluation index system is updated according to the abnormal rule mapping feature data to obtain an updated first military evaluation index system.

[0062] S103: Performing model training processing based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system;

[0063] The second military evaluation index system is used to represent the military evaluation model obtained through model training;

[0064] A deep learning model based on the Transformer architecture was trained using data from the military evaluation indicator database. The architecture consists of multiple stacked layers of identical structure, each of which is divided into two main sublayers: a multi-head self-attention mechanism and a feed-forward neural network. The former allows the model to focus on information from the entire sequence at the same time, while the latter is used to further transform the feature space. To maintain consistent input and output dimensions, residual connections and layer normalization operations can be added after each sublayer. The calculation formula is as follows:

[0065] MultiHead(Q,K,V)=Concat(head1,...,head h )W O ,in,

[0066]

[0067] Here Q, K, and V are query, key, and value matrices respectively; are the corresponding projection matrices respectively; W O is the output transformation matrix; h represents the number of heads. The attention score is calculated by dot product similarity and a scaling factor is applied. Propagate with stable gradients:

[0068]

[0069] Subsequently, a two-layer linear transformation using the ReLU activation function is used as the feedforward neural network part:

[0070] FFN(r)=max(0,xW1+b1)W2+b2,

[0071] In order to keep the input and output dimensions consistent, residual connection and layer normalization operations can be added after each sublayer.

[0072] The Xavier initialization method is used in combination with orthogonal initialization to enhance the independence between matrices and thus improve the efficiency of gradient propagation. For the weight matrix W, the Xavier initialization is set as:

[0073] Using cross entropy loss, specifically, assuming that the probability distribution of the model output is y^ and the true label is y, the total loss can be defined as:

[0074]

[0075] Where NN is the number of samples and CC is the number of categories or indicators. In addition, a regularization term is introduced to prevent overfitting and randomly discard some neurons to simulate an uncertain environment.

[0076] Adam optimizer is used as the model optimization algorithm, and the update rules are:

[0077] η represents the initial learning rate, β1 and β2 are the decay factors of the first-order moment estimate and the second-order moment estimate, respectively, with default values ​​of 0.9 and 0.999, respectively, and ε is a small constant to avoid division by zero errors.

[0078] In some optional embodiments of the present application, a method for constructing and evaluating military evaluation indicators based on deep learning is proposed, including: obtaining updated sample military data, performing incremental feature extraction processing on the sample military data to obtain incremental feature data; performing feature extraction processing based on correlation relationships on the incremental feature data and the first military evaluation indicator system to obtain updated correlation feature data; performing model update processing on the military evaluation model of the second military evaluation indicator system according to the updated correlation feature data to obtain an updated second military evaluation indicator system.

[0079] S104: Performing joint decision-making military evaluation processing on the military data to be evaluated based on the first military evaluation index system and the second military evaluation index system to obtain evaluation result data.

[0080] In some optional embodiments of the present application, a military evaluation index construction and evaluation method based on deep learning is proposed. Figure 2 A flowchart of a deep learning-based military evaluation index construction and evaluation method provided in this application, such as Figure 2 As shown, the method includes the following steps:

[0081] S201: performing military evaluation processing on the military data to be evaluated based on a first military evaluation indicator system to obtain first process evaluation result data;

[0082] The military data to be evaluated is identified and processed based on task characteristics to obtain task characteristics corresponding to the military data to be evaluated. Task characteristics include task characteristics of multiple dimensions, including combat phase characteristics, arms characteristics, task type characteristics, etc. For example, combat phase characteristics include strategy, campaign, tactics, etc., arms characteristics include army, navy, air force, etc., and task type characteristics include offense, defense, support, etc. The above different dimensions include different subcategories. According to the above task characteristic data, the rule template corresponding to the task characteristic data is matched in the first military evaluation system, and the military data to be evaluated is filtered or the evaluation results are output according to the rule template. The rule templates in the first military evaluation system are traversed. If there is no rule template data corresponding to the above task characteristic data, the military evaluation processing based on the second military evaluation indicator system is triggered.

[0083] S202: performing military evaluation processing on the military data to be evaluated based on the second military evaluation indicator system to obtain second process evaluation result data;

[0084] In an optional embodiment of the present application, let x represent the preprocessed input feature vector, f(·; θ) represent the deep learning model (the aforementioned Transformer architecture), then the predicted output is:

[0085] in, is a probability distribution vector of length C, where C represents the number of categories or the number of evaluation indicators.

[0086] By calculating the cross-entropy loss function, which is used to evaluate the difference between the model output and the true label, and combining the idea of ​​Bayesian decision theory, that is, minimizing the expected risk is converted to maximizing the posterior probability, this means that we should choose the option that maximizes the posterior probability as the final answer.

[0087] S203: Perform joint decision-making processing on the first process evaluation result data and the second process evaluation result data to obtain evaluation result data.

[0088] In some optional embodiments of the present application, a military evaluation index construction and evaluation method based on deep learning is proposed, including: extracting and processing the model prediction confidence features of the second process evaluation result data to obtain model confidence feature data, wherein the model confidence feature data is feature data used to represent the model prediction confidence; if the model confidence feature data meets the preset confidence feature rules, obtaining the evaluation result data to be fused, the evaluation result data to be fused includes the first process evaluation result data and the second process evaluation result data, performing fusion processing based on the fusion rules on the evaluation data to be fused to obtain evaluation result data, and calculating the comprehensive score by assigning different weights to the rule template and the deep learning model; if the model confidence feature data meets the preset confidence feature rules, obtaining the evaluation result data, the evaluation result data includes the first process evaluation result data.

[0089] In some optional embodiments of the present application, for example, in a combat evaluation scenario, the effectiveness of a ground force offensive operation is evaluated. Past evaluation indicator case data related to such operations is retrieved from the database, and a preliminary analysis is performed in combination with the basic information of the current operation; a combat evaluation is performed on the current operation information based on the second military evaluation indicator system, and these evaluation indicator data are processed using a pre-trained deep learning model. In combination with factors such as the force deployment, combat objectives, enemy information, and battlefield situation information of this combat operation, relevant evaluation indicators are constructed. A combat evaluation is performed on the current operation information based on the first military evaluation indicator system, and the evaluation indicators that should be included under the current conditions are determined according to the corresponding clauses in the rule template library. The evaluation indicators determined by the above-mentioned first military evaluation indicator system and the evaluation indicators constructed by the above-mentioned second military evaluation indicator system are jointly decided and processed to obtain evaluation result data including the evaluation indicators.

[0090] In some optional embodiments of the present application, a military evaluation index construction and evaluation device based on deep learning is proposed. Figure 3 This application provides a schematic diagram of a military evaluation index construction and evaluation device based on deep learning, such as Figure 3 ,include:

[0091] A feature extraction module 31 is used to obtain sample military data, perform feature extraction processing on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database;

[0092] A first military evaluation index module 32 is configured to construct a first military evaluation index system based on a rule template on the military evaluation index database to obtain a first military evaluation index system;

[0093] a second military evaluation index module 33 for performing a model training process based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained through the model training process;

[0094] The military evaluation module 34 is configured to perform a joint decision-making military evaluation process on the military data to be evaluated based on the first military evaluation index system and the second military evaluation index system to obtain the evaluation result data.

[0095] The specific manner of executing the operations of each unit in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated on here.

[0096] In summary, in this application, sample military data is obtained, and feature extraction processing based on expert labeling and machine learning is performed on the sample military data to construct a military evaluation index database; a first military evaluation index system based on a rule template is constructed on the military evaluation index database to obtain a first military evaluation index system; a model training processing based on a deep learning model is performed on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained by the model training processing; and a joint decision-making military evaluation processing based on the first military evaluation index system and the second military evaluation index system is performed on the military data to be evaluated to obtain evaluation result data. By integrating deep learning with rule templates to construct and evaluate military evaluation indicators, the technical effect of improving the accuracy of military evaluation predictions and the interpretability of evaluation indicators is achieved.

[0097] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0098] Obviously, those skilled in the art should understand that the above-mentioned units or steps of the present application can be implemented using a general-purpose computing device. They can be concentrated on a single computing device or distributed across a network composed of multiple computing devices. Alternatively, they can be implemented using program codes executable by the computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0099] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A military evaluation index construction and evaluation method based on deep learning, characterized in that: include: Acquire sample military data, perform feature extraction processing on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database; Constructing a first military evaluation index system based on a rule template on the military evaluation index database to obtain a first military evaluation index system; Performing a model training process based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained by the model training process; The military data to be evaluated is subjected to joint decision-making military evaluation processing based on the first military evaluation index system and the second military evaluation index system to obtain evaluation result data.

2. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 1 is characterized in that: Performing feature extraction processing on the sample military data based on expert labeling and machine learning to construct a military evaluation index database includes: Performing feature extraction processing based on data statistics and machine learning on the sample military data to obtain associated feature data, wherein the associated feature data is feature data used to represent the correlation relationship between military evaluation indicators; Performing feature extraction processing on the sample military data based on preset customized military features to obtain military customized feature data; Performing feature mapping processing on the sample military data based on a pre-trained language model to obtain semantic feature data, wherein the semantic feature data is feature data obtained by extracting features from language text in the sample military data; The military evaluation index database is constructed based on the associated feature data, the military customized feature data and the semantic feature data.

3. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 1 is characterized in that: Performing joint decision-making military evaluation processing on the military data to be evaluated based on the first military evaluation index system and the second military evaluation index system to obtain the evaluation result data includes: Performing military evaluation processing on the military data to be evaluated based on the first military evaluation indicator system to obtain first process evaluation result data; Performing military evaluation processing on the military data to be evaluated based on the second military evaluation indicator system to obtain second process evaluation result data; The first process evaluation result data and the second process evaluation result data are jointly processed for decision-making to obtain the evaluation result data.

4. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 3 is characterized in that: Performing joint decision-making processing on the first process evaluation result data and the second process evaluation result data to obtain the evaluation result data includes: Performing model prediction confidence feature extraction processing on the second process evaluation result data to obtain model confidence feature data, wherein the model confidence feature data is feature data used to represent the model prediction confidence; If the model confidence feature data satisfies the preset confidence feature rules, obtaining the evaluation result data to be fused, the evaluation result data to be fused includes the first process evaluation result data and the second process evaluation result data, and performing fusion processing on the evaluation data to be fused based on the fusion rules to obtain the evaluation result data; If the model confidence feature data satisfies the preset confidence feature rules, the evaluation result data is obtained, and the evaluation result data includes the first process evaluation result data.

5. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 1 is characterized in that: Also includes: Performing rule anomaly trigger detection on the first military evaluation indicator system to obtain rule anomaly trigger data; Performing extraction processing on the rule abnormality triggering data based on time features and rule mapping features to obtain abnormal time feature data and abnormal rule mapping feature data, wherein the abnormal rule mapping feature data is feature data used to indicate abnormality in the trigger rule mapping relationship; If the abnormal time feature data satisfies a preset dynamic update rule, the rule template of the first military evaluation index system is updated according to the abnormal rule mapping feature data to obtain an updated first military evaluation index system.

6. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 1 is characterized in that: Also includes: Acquire updated sample military data, perform incremental feature extraction on the sample military data, and obtain incremental feature data; performing feature extraction processing based on the association relationship on the incremental feature data and the first military evaluation index system to obtain updated association feature data; A model update process is performed on the military evaluation model of the second military evaluation index system according to the updated associated feature data to obtain an updated second military evaluation index system.

7. The method for constructing and evaluating military evaluation indicators based on deep learning according to claim 1 is characterized in that: Performing model training processing based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system includes: The Transformer-based framework of the deep learning model includes multiple stacked layers of the same structure, each of which includes two main sublayers: a multi-head self-attention mechanism and a feedforward neural network. A two-layer linear transformation using the ReLU activation function is used as the feedforward neural network part, cross entropy is used as the loss function, and Xavier initialization and orthogonal initialization are used to initialize the parameters of the deep learning model to train a military assessment model.

8. A military evaluation index construction and evaluation device based on deep learning, characterized in that: include: A feature extraction module is used to obtain sample military data, perform feature extraction processing on the sample military data based on expert labeling and machine learning, and construct a military evaluation index database; A first military evaluation index module is used to construct a first military evaluation index system based on a rule template on the military evaluation index database to obtain a first military evaluation index system; a second military evaluation index module, configured to perform model training processing based on a deep learning model on the military evaluation index database to obtain a second military evaluation index system, wherein the second military evaluation index system is used to represent the military evaluation model obtained through the model training processing; The military evaluation module is used to perform joint decision-making military evaluation processing on the military data to be evaluated based on the first military evaluation index system and the second military evaluation index system to obtain the evaluation result data.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, which are used to enable the computer to execute the deep learning-based military evaluation index construction and evaluation method described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor executes the deep learning-based military evaluation indicator construction and evaluation method described in any one of claims 1-7.

Images