Military situation assessment and prediction method and system based on big data
Through big data technology, a situation evaluation index system and prediction model are constructed, which solves the complexity of situation data and insufficient prediction, and achieves accurate prediction and combat guidance for future situation changes.
Patent Information
- Application Number
- CN202410936873.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-26
AI Technical Summary
Military situation data has large amounts of data, complexity, multi-source and heterogeneous data, which is not conducive to analysis and utilization. The existing situation analysis cannot accurately predict the trend changes in the next stage.
Using a big data-based method, the elements of military situation historical data are extracted through the MapReduce model, the situation evaluation index system is constructed, and the total situation value is calculated using hierarchical analysis method, and the historical change sequence of situation value is constructed based on time variables, and the feedforward neural network is used and improvements are improved. The particle swarm algorithm constructs a situation prediction model and outputs a military situation prediction sequence.
The quality and availability of collation and analysis of complex situation data has been improved, accurate prediction of future military situation changes has been achieved, and the effectiveness of combat guidance has been enhanced.
Smart Images

Figure CN119026973A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of military situation analysis, and specifically relates to a military situation assessment and prediction method and system based on big data. Background Art
[0002] Information Operations (IO) refers to military actions taken in the information environment to influence, manipulate and control information in support of military strategies and objectives. Information operations play a vital role in modern warfare as it can significantly improve the effectiveness of military operations.
[0003] In information operations, situation analysis is a method of evaluating and understanding a specific situation or environment. It is used to identify, evaluate and interpret various factors in order to better understand the current state and predict future development trends. Situation analysis is a key component of the decision-making process of information operations. It helps commanders and decision makers understand the battlefield environment and develop and implement effective strategies and tactics.
[0004] At present, there are at least the following problems in the practical application of situation analysis: 1. Military situation data is large in volume, complex, multi-source and heterogeneous, which is not conducive to the analysis and utilization of military situation data; 2. Most of the existing situation analysis can only provide commanders with information on the past and current military situation, and cannot accurately predict changes in the situation in the next stage. Summary of the invention
[0005] The purpose of the present invention is to provide a military situation assessment and prediction method and system based on big data, so as to solve the problems in the prior art that the military situation data is large in volume, complex, multi-source and heterogeneous, which is not conducive to the analysis and utilization of the military situation data, and the situation analysis can only provide commanders with the past and current military situation, and cannot accurately predict the situation changes in the next stage.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a military situation assessment and prediction method and system based on big data, the method comprising: Obtain a historical military situation data set, extract elements from the historical military situation data set based on a pre-built big data extraction model, and obtain situation estimation elements; The military situation assessment index system is constructed with situation estimation elements, and the total situation value of the military situation assessment index system is calculated based on the hierarchical analysis method; Based on the time variable and the total situation value of the military situation assessment index system, a historical change sequence of situation values is constructed; The historical change sequence of situation values is used as input to a pre-built situation prediction model, and the situation prediction model outputs a military situation prediction sequence.
[0007] Preferably, the big data extraction model is a MapReduce model, and the MapReduce model includes: a Map function and a Reduce function; Based on the pre-built big data extraction model, elements of military situation historical data are extracted to obtain situation estimation elements, including: Preprocess the military situation historical data set to obtain the target data set; Based on the Map function, the target data set is divided into several sub-data sets, and the several sub-data sets are key-value processed to obtain a Map intermediate result, wherein the Map intermediate result includes several key-value pairs; Based on the Reduce function, the intermediate results of the Map are merged to obtain a Reduce data table, which contains situation estimation elements.
[0008] Preferably, the Map intermediate results are merged based on the Reduce function to obtain a Reduce data table, including: The Reduce function receives a number of key-value pairs output from the Map function, processes the key-value pairs based on a density clustering algorithm, and obtains cluster core points; Based on the fuzzy mean clustering algorithm, cluster core points are clustered to obtain several key-value pair clustering groups; Construct a Reduce data table based on several key-value pair clustering groups.
[0009] Preferably, the preprocessing includes at least: data cleaning, data simplification and data integration.
[0010] Preferably, the military situation assessment indicator system includes multiple levels, each level includes at least one assessment indicator, each assessment indicator has at least one evaluation item, the assessment indicators of the lower level are used as the evaluation items of the assessment indicators of the upper level, and the evaluation items of the assessment indicators of the lower level are the corresponding situation estimation elements; The score of each evaluation indicator is obtained by weighting at least one evaluation item corresponding to each evaluation indicator, and the score of the top-level evaluation indicator is used as the total situation value of the military situation assessment indicator system.
[0011] Preferably, the method further comprises: constructing a situation prediction model based on a feedforward neural network, comprising: Obtaining situation estimation elements at a number of sampling moments within a preset historical period and total situation values corresponding to the situation estimation elements at the number of sampling moments, using the situation estimation elements at the number of sampling moments as multiple groups of training sample data, and using the total situation values corresponding to the situation estimation elements at the number of sampling moments as training label values; Initialize the parameters of the feedforward neural network, use the training sample data to train the feedforward neural network until the feedforward neural network converges, and obtain the trained feedforward neural network, and use the trained feedforward neural network as the situation prediction model.
[0012] Preferably, the feedforward neural network comprises: an input layer, a hidden layer and an output layer, and the parameters of the feedforward neural network comprise: the dimension of the input layer, the dimension of the output layer and the number of neurons in the hidden layer; In the process of training the feedforward neural network using the training sample data, the method further includes: optimizing the parameters of the feedforward neural network based on the improved particle swarm algorithm, including: Constructing an objective function based on the predicted output value and the expected output value of the feedforward neural network, and using the parameters of the feedforward neural network as decision variables of the objective function; wherein the expected output value is a training label value; The objective function is iteratively solved based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, which include: the optimal input layer dimension, the optimal output layer dimension and the optimal number of neurons in the hidden layer.
[0013] Preferably, the objective function is iteratively solved based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, including: Using the decision variables as the input of the improved particle swarm algorithm, initializing the parameters of the improved particle swarm algorithm; Based on the objective function, the fitness of each particle in the current particle population is calculated, and the historical best position and the best position of the particle population are updated; Replace the best position of the population based on the replacement strategy to obtain the best population position after replacement; The inertia weight of the particle swarm algorithm is updated based on the fitness of each particle in the current particle population to obtain an updated inertia weight; Update the particle's velocity and position based on the replaced best population position, the historical best position, and the updated inertia weight; The fitness of each particle is recalculated based on the updated particle speed and position, and the historical best position and the population best position of the particle population are updated repeatedly; the population best position is replaced based on the replacement strategy to obtain the replaced best population position; the inertia weight of the particle swarm algorithm is updated based on the fitness of each particle of the current particle population to obtain the updated inertia weight; the particle speed and position are updated based on the replaced best population position, the historical best position and the updated inertia weight; until the maximum number of iterations is reached, the decision variable corresponding to the global optimal particle position is obtained as the optimal parameter of the feedforward neural network.
[0014] Preferably, the replacement strategy is: randomly selecting a particle from the current particle population according to a roulette random algorithm, and replacing the optimal position of the population with the position of the selected particle.
[0015] In a second aspect, the present invention provides a military situation assessment and prediction system based on big data, which is used to implement the above-mentioned military situation assessment and prediction method based on big data, and the system includes: A data acquisition module is used to acquire a historical military situation data set, extract elements from the historical military situation data set based on a pre-built big data extraction model, and obtain situation estimation elements; The situation assessment module is used to construct a military situation assessment index system based on situation estimation elements and calculate the total situation value of the military situation assessment index system based on the hierarchical analysis method; A sequence construction module is used to construct a historical change sequence of situation values based on the time variable and the total situation value of the military situation assessment indicator system; The situation prediction module is used to use the historical change sequence of situation values as the input of a pre-built situation prediction model, and the situation prediction model outputs a military situation prediction sequence.
[0016] Beneficial effects: 1. The present invention extracts elements from the military situation historical data set through a pre-selected big data extraction model, realizes the sorting and analysis of massive multi-source heterogeneous and complex situation data, can improve the quality, availability, consistency and interoperability of the data, reduce the complexity of the data, and is conducive to improving the processing efficiency and accuracy of the evaluation and prediction of the situation data; 2. By constructing a military situation assessment index system, the present invention can objectively calculate the overall situation value, determine the change of the overall situation value according to the time dimension, and use the situation prediction model to predict the changes in the future military situation. Taking the future changes in the military situation as a combat guide is conducive to improving the military's all-round combat capability. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 This is a process of a military situation assessment and prediction method based on big data provided by an embodiment of the present invention; Figure 2 It is a block diagram of a military situation assessment and prediction system based on big data provided by one embodiment of the present invention. DETAILED DESCRIPTION
[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.
[0019] Embodiment 1 Figure 1 This is a process of a military situation assessment and prediction method based on big data provided by an embodiment of the present invention, such as Figure 1 As shown, this embodiment provides a military situation assessment and prediction method and system based on big data, and the method includes: Step S10: Obtain a military situation historical data set, extract elements from the military situation historical data set based on a pre-built big data extraction model, and obtain situation estimation elements.
[0020] In this embodiment, the data in the military situation history data set can be document text, image data, video data, audio data, etc. The military situation history data set includes: combat capability data of the two combatants (mobility data, firepower data, logistics data, etc.), combat environment data (traffic data, weather data, geographic data, social data, etc.), combat target data (troop deployment data, combat facility data, combat mission data, etc.), etc.
[0021] However, the above-mentioned multi-source heterogeneous complex data is not easy to analyze and store; therefore, the big data extraction model is a MapReduce model, and the MapReduce model includes: Map (mapping) function and Reduce (reduction) function.
[0022] Among them, the Map function is used to divide the input data (military situation historical data set) into multiple small blocks, each small block is processed independently, and then each small block is mapped to a key-value pair; the key-value pairs output by the Map function are sent to the Reduce function, all values with the same key are grouped together, and the Reduce function processes each key-value pair to generate an output list; the MapReduce model can process data in parallel on multiple computers, decompose complex problems into simple parts, and then summarize the results; it is convenient for the later evaluation and prediction of the military situation.
[0023] Specifically, based on the pre-built big data extraction model, elements of military situation historical data are extracted to obtain situation estimation elements, including: Step S101: pre-processing the military situation history data set to obtain a target data set; Step S102: splitting the target data set based on the Map function to obtain a number of sub-data sets, performing key value processing on the sub-data sets to obtain a Map intermediate result, wherein the Map intermediate result includes a number of key-value pairs; Step S103: Based on the Reduce function, the intermediate results of the Map are merged to obtain a Reduce data table, which contains situation estimation elements.
[0024] In this embodiment, for step S101, the preprocessing at least includes: data cleaning, data reduction and data integration.
[0025] Among them, data cleaning: Since the confusing data in the military situation history data set may account for more than 30% of the total data, some programs need to be used for cleaning, such as filling missing values, smoothing noisy data, or resolving inconsistencies in the data.
[0026] Among them, data reduction: data reduction is divided into numerical reduction and dimensionality reduction. Regression or sampling methods can be used for numerical reduction. Data dimensionality reduction can use a variety of algorithms, such as PCA (principal component analysis), SOM (self-organizing map) and SVD (singular value decomposition) algorithms.
[0027] Among them, data integration: merging data sets from different sources, and merging data sets with associated relationships.
[0028] In this embodiment, by preprocessing the military situation history data set, the performance and prediction accuracy of the model can be significantly improved.
[0029] In this embodiment, for step S103, the Map intermediate results are merged based on the Reduce function to obtain a Reduce data table, including: Step a10: the Reduce function receives a number of key-value pairs outputted from the Map function, processes the key-value pairs based on a density clustering algorithm, and obtains cluster core points; Step a20: clustering the cluster core points based on the fuzzy mean clustering algorithm to obtain a number of key-value pair clustering groups; Step a30: construct a Reduce data table according to a number of key-value pair clustering groups.
[0030] In this embodiment, the density clustering algorithm preferably adopts the DBSCAN algorithm, which is a density-based spatial clustering algorithm; the fuzzy C-Means Clustering (FCM) algorithm divides the data set into n clusters, and each data point in the data set is highly correlated with a cluster, and the data points far away from the cluster center have a lower degree of correlation with the cluster.
[0031] Among them, the main steps of the FCM algorithm include: 1. Initialize the cluster center and the number of clusters; 2. For each data point, calculate its distance to each cluster center; 3. Calculate the membership of each data point to each cluster center. The membership is determined based on the distance from the data point to the cluster center and the cluster radius. 4. Update the cluster center according to the membership degree; Repeat steps 2-4 until the change in cluster center is less than a certain threshold or the upper limit of the number of iterations is reached.
[0032] Among them, the main steps of the DBSCAN algorithm include: Defining density: DBSCAN algorithm defines clusters by density. The density around a data point is determined by the number of data points in its neighborhood. Core Points: A core point means that there must be at least MinPts points in its neighborhood. Therefore, it and its neighborhood points form a high-density area. Border Points: A border point is a point that is in the neighborhood of a core point but is not a core point itself. They may belong to a cluster, but may not be so clear because of their low density. Noise Points: Points that are neither core points nor boundary points are called noise points. They may be outliers or abnormal values. Discovering clusters: The algorithm starts from the core point and gradually expands the neighborhood until no new core points appear. Each core point may become the core of a cluster; Merge clusters: When the algorithm expands the neighborhood, if two core points belong to the same neighborhood, they are merged into the same cluster.
[0033] In this embodiment, the FCM algorithm can obtain flexible clustering results, that is, data points can fuzzily belong to multiple clusters, which makes FCM have advantages when processing data with complex structures and fuzzy boundaries. However, the FCM algorithm also has some disadvantages, such as being sensitive to noise and outliers, and the number of clusters needs to be specified in advance. The present invention processes several key-value pairs using the DBSCAN algorithm, can identify noise points and potential clustering core points in the key-value pairs, and then uses the fuzzy mean clustering algorithm to cluster the clustering core points, which can obtain better clustering results and provide accurate basic data for the military situation assessment index system of step S20.
[0034] Step S20: construct a military situation assessment index system based on situation estimation elements, and calculate the total situation value of the military situation assessment index system based on the hierarchical analysis method.
[0035] In this embodiment, the military situation assessment indicator system includes multiple levels, each level includes at least one assessment indicator, each assessment indicator has at least one evaluation item, the assessment indicators of the lower level are used as the evaluation items of the assessment indicators of the upper level, and the evaluation items of the assessment indicators of the lower level are the corresponding situation estimation elements; The score of each evaluation indicator is obtained by weighting at least one evaluation item corresponding to each evaluation indicator, and the score of the top-level evaluation indicator is used as the total situation value of the military situation assessment indicator system.
[0036] In this embodiment, the military situation assessment index system has four levels. The first level includes a first-level indicator: the overall military situation assessment index; the second level includes three second-level indicators: combat capability index, combat environment index and combat target index; each second-level indicator includes multiple third-level indicators, and the third-level indicators serve as the third level. Each third-level indicator includes multiple fourth-level indicators, and the fourth-level indicators serve as the fourth level.
[0037] Among them, the third-level indicators under the combat capability index include: enemy combat capability index, our combat capability index and ally combat capability index; and the fourth-level indicators under the enemy combat capability index include: enemy mobility capability index, enemy firepower capability index and enemy logistics index, etc.; the fourth-level indicators are determined by situation assessment elements, for example: the enemy mobility capability index can be determined by scoring based on the enemy mobility data (number of armored vehicles, number of tanks, number of drones, etc.), and the scoring adopts the entropy weight method.
[0038] Similarly, the four-level indicators under our combat capability indicators include: our mobility capability indicators, our firepower capability indicators and our logistics indicators, etc.; the four-level indicators under the ally combat capability indicators include: ally mobility capability indicators, ally firepower capability indicators and ally logistics indicators, etc.
[0039] Among them, the third-level indicators under the combat environment indicators include: traffic indicators, weather indicators, geographical indicators and social indicators, etc.; the fourth-level indicators under the traffic indicators include: land indicators, air indicators and sea indicators, etc.; the fourth-level indicators under the weather indicators include: land meteorological indicators, ocean meteorological indicators, etc.; the fourth-level indicators under the geographical indicators include: landform indicators, river basin indicators, etc.; the fourth-level indicators under the social indicators include: economic indicators, cultural indicators, etc.
[0040] Among them, the third-level indicators under the combat target indicators include: air indicators, ground indicators and sea indicators, etc.; the fourth-level indicators under the air indicators include: air facility indicators, air force indicators, etc.; the fourth-level indicators under the ground indicators include: ground facility indicators, ground force indicators, etc.; the fourth-level indicators under the sea indicators include: sea facility indicators, sea force indicators, etc.
[0041] In this embodiment, the entropy weight method is used to calculate the score of each four-level indicator, and then the score of each three-level indicator is obtained by weighted summing the scores of multiple four-level indicators under the three-level indicator. The score of each indicator is calculated layer by layer, and the score of the first-level indicator is used as the total situation value of the military situation assessment index system, and the other scores of each level are also used as the situation value of the indicator at that level.
[0042] Step S30: construct a historical change sequence of situation values based on the time variable and the total situation value of the military situation assessment index system.
[0043] In this embodiment, the historical change sequence of situation values can reflect the changes in the total situation value in the time dimension; at the same time, the scores of indicators at other levels can also be included in the historical change sequence of situation values. At this time, the historical change sequence of situation values can reflect the changes in the situation values of each level of the military situation assessment indicator system.
[0044] Step S40: Using the historical change sequence of situation values as input to a pre-built situation prediction model, the situation prediction model outputs a military situation prediction sequence.
[0045] By constructing a military situation assessment index system, the present invention can objectively calculate the overall situation value, determine the change of the overall situation value according to the time dimension, and use the situation prediction model to predict the change of the future military situation. Taking the future military situation change as a combat guide is beneficial to improving the military's all-round combat capability.
[0046] As a further optimization of this embodiment, the method further includes: constructing a situation prediction model based on a feedforward neural network, including: Step b10: obtaining situation estimation elements at a number of sampling moments in a preset historical period and total situation values corresponding to the situation estimation elements at the number of sampling moments, using the situation estimation elements at the number of sampling moments as multiple sets of training sample data, and using the total situation values corresponding to the situation estimation elements at the number of sampling moments as training label values; Step b20: Initialize the parameters of the feedforward neural network, and train the feedforward neural network using the training sample data until the feedforward neural network converges to obtain a trained feedforward neural network, and use the trained feedforward neural network as a situation prediction model.
[0047] In this embodiment, the feedforward neural network includes: an input layer, a hidden layer and an output layer, and the parameters of the feedforward neural network include: the dimension of the input layer, the dimension of the output layer and the number of neurons in the hidden layer; therefore, the topological structure of the feedforward neural network is relatively simple, easy to understand and implement. It consists of multiple layers of nodes, and the output of each layer of nodes is in turn used as the input of the next layer, and does not form a closed loop; it has the characteristics of few control parameters, fast convergence speed, and strong generalization ability, but the initial value of the network's own parameters directly affects the prediction accuracy and learning speed.
[0048] In order to improve the prediction accuracy of the feedforward neural network, in the process of training the feedforward neural network using training sample data, the method further includes: optimizing the parameters of the feedforward neural network based on the improved particle swarm algorithm, including: Step b10: constructing an objective function based on the predicted output value and the expected output value of the feedforward neural network, and using the parameters of the feedforward neural network as decision variables of the objective function; wherein the expected output value is a training label value; Step b20: Iteratively solve the objective function based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, wherein the optimal parameters of the feedforward neural network include: the optimal input layer dimension, the optimal output layer dimension and the optimal number of neurons in the hidden layer.
[0049] Among them, the expression of the objective function is: ; In the formula, F is the objective function, N is the total number of training sample data sets, i For the i Group training sample data, For the i The predicted output value of the training sample data set, For the i The expected output value of the training sample data set,F The smaller the value, the closer the particle is to the optimal value.
[0050] In step b20, specifically, the objective function is iteratively solved based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, including: Step b201: using the decision variables as the input of the improved particle swarm algorithm to initialize the parameters of the improved particle swarm algorithm; Step b202: Based on the objective function, calculate the fitness of each particle in the current particle population, and update the historical best position and the population best position of the particle population; Step b203: replacing the best position of the population based on the replacement strategy to obtain the best population position after replacement; Step b204: updating the inertia weight of the particle swarm algorithm based on the fitness of each particle in the current particle population to obtain an updated inertia weight; Step b205: updating the speed and position of the particle based on the replaced best population position, the historical best position and the updated inertia weight; Step b206: Recalculate the fitness of each particle based on the updated particle speed and position, and repeat steps b202 to b206 until the iteration end condition (such as the maximum number of iterations) is reached. The decision variable corresponding to the global optimal particle position can be obtained as the optimal parameter of the feedforward neural network.
[0051] In this embodiment, the update expressions of the particle speed and position are: ; ; In the formula, is the number of l The speed of a particle, In the t+1th iteration l The position of a particle, In the tth iteration l The position of a particle, is the number of l The speed of a particle, is the inertia weight, For the l The historical optimal position of a particle, and are the first coefficient and the second coefficient respectively, and are all random numbers in the interval [0,1]. The best population position after replacement in the tth iteration, j is the first particle randomly selected from all particlesj Particles.
[0052] In this embodiment, the replacement strategy is: randomly selecting a particle from the current particle population according to the roulette random algorithm, and replacing the optimal position of the population with the position of the selected particle; by randomly selecting a particle and replacing the optimal position of the population with the position of the particle, it is possible to effectively avoid falling into a local minimum solution during the iterative solution process.
[0053] The steps to randomly select a particle using the roulette random algorithm are: 1. Assign a random weight to each particle. The random weight can be any positive number. 2. Calculate the sum of all random weights; 3. Generate a random number between 1 and the sum of random weights; 4. Traverse each particle and its corresponding random weight, subtract the sum of the traversed random weights from the random weight each time, and select the particle when the random number is less than or equal to the remaining random weight; otherwise, continue traversing.
[0054] In this embodiment, the inertia weight affects the moving speed of the particles, which in turn affects the search accuracy and search speed of the particle positions; therefore, the inertia weight is updated to improve the search capability of the algorithm.
[0055] The updating formula of the inertia weight in this embodiment is: ; In the formula, For the l The current inertia weight of each particle, and are the maximum and minimum values of the inertia weight, respectively. For the l The current fitness value of a particle, is the minimum fitness value of all particles in the current population, is the average fitness value of all particles in the current population.
[0056] In this embodiment, during the iterative solution process, when the fitness of a particle is large, it means that the particle is far away from the optimal position. At this time, the inertia weight is the maximum value, which improves the search speed of the algorithm; when the fitness of the particle is equal to or less than the average fitness value of all particles in the current population, the minimum value of the inertia weight is dynamically adjusted according to the current fitness of each particle, the minimum fitness value of all particles in the current population, and the average fitness value of all particles in the current population. The position and speed of each particle are updated with the adjusted inertia weight. At this time, the inertia weight is small, which allows the algorithm to search for the optimal solution more finely.
[0057] The present invention extracts elements from the military situation historical data set through a pre-selected big data extraction model, realizes the sorting and analysis of massive multi-source heterogeneous and complex situation data, can improve the quality, availability, consistency and interoperability of the data, reduce the complexity of the data, and is conducive to improving the processing efficiency and accuracy of the evaluation and prediction of the situation data; and by constructing a military situation evaluation index system, the total situation value can be objectively calculated, the change of the total situation value can be determined according to the time dimension, and the situation prediction model can be used to predict the change of the future military situation, and the future military situation changes can be used as combat guidance, which is conducive to improving the all-round combat capability of the military.
[0058] Embodiment 2 Figure 2 is a block diagram of a military situation assessment and prediction system based on big data provided by an embodiment of the present invention, such as Figure 2 As shown, this embodiment provides a military situation assessment and prediction system based on big data, which is used to implement the military situation assessment and prediction method based on big data of embodiment 1. The system includes: A data acquisition module is used to acquire a historical military situation data set, extract elements from the historical military situation data set based on a pre-built big data extraction model, and obtain situation estimation elements; The situation assessment module is used to construct a military situation assessment index system based on situation estimation elements and calculate the total situation value of the military situation assessment index system based on the hierarchical analysis method; A sequence construction module is used to construct a historical change sequence of situation values based on the time variable and the total situation value of the military situation assessment indicator system; The situation prediction module is used to use the historical change sequence of situation values as the input of a pre-built situation prediction model, and the situation prediction model outputs a military situation prediction sequence.
[0059] The present invention extracts elements from the military situation historical data set through a pre-selected big data extraction model, realizes the sorting and analysis of massive multi-source heterogeneous and complex situation data, can improve the quality, availability, consistency and interoperability of the data, reduce the complexity of the data, and is conducive to improving the processing efficiency and accuracy of the evaluation and prediction of the situation data; and by constructing a military situation evaluation index system, the total situation value can be objectively calculated, the change of the total situation value can be determined according to the time dimension, and the situation prediction model can be used to predict the change of the future military situation, and the future military situation changes can be used as combat guidance, which is conducive to improving the all-round combat capability of the military.
[0060] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0062] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A military situation assessment and prediction method based on big data, It is characterized in that The method comprises: Obtain a historical military situation data set, extract elements from the historical military situation data set based on a pre-built big data extraction model, and obtain situation estimation elements; The military situation assessment index system is constructed with situation estimation elements, and the total situation value of the military situation assessment index system is calculated based on the hierarchical analysis method; Based on the time variable and the total situation value of the military situation assessment index system, a historical change sequence of situation values is constructed; The historical change sequence of situation values is used as input to a pre-built situation prediction model, and the situation prediction model outputs a military situation prediction sequence.
2. According to the military situation assessment and prediction method based on big data according to claim 1, It is characterized in that The big data extraction model is a MapReduce model, and the MapReduce model includes: a Map function and a Reduce function; Based on the pre-built big data extraction model, elements of military situation historical data are extracted to obtain situation estimation elements, including: Preprocess the military situation historical data set to obtain the target data set; Based on the Map function, the target data set is divided into several sub-data sets, and the several sub-data sets are key-value processed to obtain a Map intermediate result, wherein the Map intermediate result includes several key-value pairs; Based on the Reduce function, the intermediate results of the Map are merged to obtain a Reduce data table, which contains situation estimation elements.
3. According to the military situation assessment and prediction method based on big data as claimed in claim 2, It is characterized in that Based on the Reduce function, the intermediate results of Map are merged to obtain the Reduce data table, including: The Reduce function receives a number of key-value pairs output from the Map function, processes the key-value pairs based on a density clustering algorithm, and obtains cluster core points; Based on the fuzzy mean clustering algorithm, cluster core points are clustered to obtain several key-value pair clustering groups; Construct a Reduce data table based on several key-value pair clustering groups.
4. According to the military situation assessment and prediction method based on big data as claimed in claim 2, It is characterized in that The preprocessing includes at least: data cleaning, data simplification and data integration.
5. According to the military situation assessment and prediction method based on big data as claimed in claim 1, It is characterized in that The military situation assessment indicator system includes multiple levels, each level includes at least one assessment indicator, each assessment indicator has at least one evaluation item, the assessment indicators of the lower level are used as the evaluation items of the assessment indicators of the upper level, and the evaluation items of the assessment indicators of the lower level are the corresponding situation estimation elements; The score of each evaluation indicator is obtained by weighting at least one evaluation item corresponding to each evaluation indicator, and the score of the top-level evaluation indicator is used as the total situation value of the military situation assessment indicator system.
6. According to the military situation assessment and prediction method based on big data as claimed in claim 1, It is characterized in that The method further includes: constructing a situation prediction model based on a feedforward neural network, including: Obtaining situation estimation elements at a number of sampling moments within a preset historical period and total situation values corresponding to the situation estimation elements at the number of sampling moments, using the situation estimation elements at the number of sampling moments as multiple groups of training sample data, and using the total situation values corresponding to the situation estimation elements at the number of sampling moments as training label values; Initialize the parameters of the feedforward neural network, use the training sample data to train the feedforward neural network until the feedforward neural network converges, and obtain the trained feedforward neural network, and use the trained feedforward neural network as the situation prediction model.
7. According to the military situation assessment and prediction method based on big data as claimed in claim 6, It is characterized in that The feedforward neural network includes: an input layer, a hidden layer and an output layer, and the parameters of the feedforward neural network include: the dimension of the input layer, the dimension of the output layer and the number of neurons in the hidden layer; In the process of training the feedforward neural network using the training sample data, the method further includes: optimizing the parameters of the feedforward neural network based on the improved particle swarm algorithm, including: Constructing an objective function based on the predicted output value and the expected output value of the feedforward neural network, and using the parameters of the feedforward neural network as decision variables of the objective function; wherein the expected output value is a training label value; The objective function is iteratively solved based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, which include: the optimal input layer dimension, the optimal output layer dimension and the optimal number of neurons in the hidden layer.
8. According to the military situation assessment and prediction method based on big data as claimed in claim 7, It is characterized in that The objective function is iteratively solved based on the improved particle swarm algorithm to obtain the optimal parameters of the feedforward neural network, including: Using the decision variables as the input of the improved particle swarm algorithm, initializing the parameters of the improved particle swarm algorithm; Based on the objective function, the fitness of each particle in the current particle population is calculated, and the historical best position and the best position of the particle population are updated; Replace the best position of the population based on the replacement strategy to obtain the best population position after replacement; The inertia weight of the particle swarm algorithm is updated based on the fitness of each particle in the current particle population to obtain an updated inertia weight; Update the particle's velocity and position based on the replaced best population position, the historical best position, and the updated inertia weight; The fitness of each particle is recalculated based on the updated particle speed and position, and the historical best position and the population best position of the particle population are updated repeatedly; the population best position is replaced based on the replacement strategy to obtain the replaced best population position; the inertia weight of the particle swarm algorithm is updated based on the fitness of each particle of the current particle population to obtain the updated inertia weight; the particle speed and position are updated based on the replaced best population position, the historical best position and the updated inertia weight; until the maximum number of iterations is reached, the decision variable corresponding to the global optimal particle position is obtained as the optimal parameter of the feedforward neural network.
9. According to the military situation assessment and prediction method based on big data as claimed in claim 8, It is characterized in that The replacement strategy is: randomly select a particle from the current particle population according to the roulette random algorithm, and replace the optimal position of the population with the position of the selected particle.
10. A military situation assessment and prediction system based on big data, used to implement the military situation assessment and prediction method based on big data as described in any one of claims 1 to 9, It is characterized in that The system comprises: A data acquisition module is used to acquire a historical military situation data set, extract elements from the historical military situation data set based on a pre-built big data extraction model, and obtain situation estimation elements; The situation assessment module is used to construct a military situation assessment index system based on situation estimation elements and calculate the total situation value of the military situation assessment index system based on the hierarchical analysis method; A sequence construction module is used to construct a historical change sequence of situation values based on the time variable and the total situation value of the military situation assessment indicator system; The situation prediction module is used to use the historical change sequence of situation values as the input of a pre-built situation prediction model, and the situation prediction model outputs a military situation prediction sequence.
Citation Information
Cited By
Simulation situation data-based military assessment method, apparatus and device, and storage medium
CN120430197A
Military evaluation index construction and evaluation method and device based on deep learning
CN120670745A
