A classification processing method and system for information interaction data

Abstract

The present invention relates to a classification processing method and system for information interaction data, belonging to the technical field of information network data processing, the method comprising: obtaining a combination of data to be sorted; classifying the combination of data to be sorted by a pre-established data classifier to obtain the classification type of each data in the combination of data to be sorted; matching the basic sorting rules corresponding to each data from a preset basic sorting rule library according to the classification type of each data; sorting the combination of data to be sorted based on the basic sorting rules corresponding to each data to obtain the data combination sending order; distributing and processing each data in the combination of data to be sorted in turn according to the data combination sending order. The present application can realize the clear classification and sorting of congested data combinations, thereby solving the problem of full coverage of data combinations and improving the congestion regulation coverage performance.

Description

Technical Field

[0001] The present invention belongs to the technical field of information network data processing, and in particular relates to a classification processing method and system for information interaction data. Background Art

[0002] The test training environment is a type of cross-domain distributed heterogeneous LVC simulation system (LVC simulation can connect the scattered live simulation, virtual simulation and constructive simulation into an interactive comprehensive simulation environment through a computer network.), which has a large geographical span and usually includes several physical ranges or bases; at the same time, the test training environment contains various types of physical and semi-physical equipment and digital systems. It is a virtual and real combination of simulation environment for testing, training and equipment assessment. From the perspective of system composition, the essence of the test training environment can be regarded as a logical range, which has the characteristics of geographical distribution and logical unity. It consists of multiple physical ranges, each of which includes several local area networks; further, the local area network can include several resource nodes, which are the basic units of the test training environment. From the perspective of system functions, the test training environment includes test task-related functions and environment operation support functions. Functions related to test tasks include: test planning, test management, test monitoring, test playback, test evaluation, etc.; environmental operation support functions include: object modeling, resource encapsulation, middleware, shared resource library, shared database and heterogeneous gateway, etc., and system functions are the basis for building the software system.

[0003] When conducting test and training tasks, the test and training environment is integrated into an information system network, and the software system interconnects the resources of each node and can interoperate. In order to meet the efficient and reliable operation of the test and training tasks, a high-performance data interaction system is required. High Level Architecture (HLA) and Test and Training Enabling Architecture (TENA) are two widely used data interaction and control standards, and their goals are to achieve data transmission between distributed network resources and further achieve interoperability. On the user side, the construction levels of different physical ranges are different, the informationization and intelligence levels of various types of equipment resources are different, and the technical system is obviously diversified; in terms of communication, the backbone network construction between physical ranges is incomplete, and so on. In the construction of the logical range, there are various objective requirements from the user side for data transmission, and it is necessary to further improve the service capabilities of data interaction in the test and training environment.

[0004] At present, the basic way of data interaction in the test and training environment is: the object model constitutes the data source or data sink, the middleware unifies the interactive information sent by the object model into a data format consistent with the entire network, and the middleware performs data format conversion and transmission between object models. The middleware can provide a variety of communication methods for different business application requirements. This interaction method focuses on node requirements, processes and sends interaction requests to the transport layer in a timely manner, and can better support various test and training tasks under the premise of sufficient bandwidth of communication facilities. However, when there are too many access nodes in the test and training environment and various types of data are intertwined, random congestion of various data combinations will occur.

[0005] For the problem of communication congestion in the test network, we usually target specific congested areas, analyze them and take measures to alleviate them. However, it is difficult to achieve both accuracy of the solution and coverage of the congested areas or phenomena. In particular, for the internal data interaction needs of multi-node networks, due to the large randomness of data source combinations and the large number of combination types, it is difficult for existing congestion regulation measures to achieve full coverage. It is urgent to adopt a new method that can not only effectively regulate congestion but also comprehensively cover the sources of congestion of various random data combinations, and improve the congestion relief service capabilities; in addition, it has the ability to perceive and regulate the network environment globally, avoid the limitations of a single node environment, adapt to task requirements to the maximum extent, and significantly improve the regulation effect. Summary of the invention

[0006] The present invention is intended to provide a classification processing method and system for information interaction data to solve the problem that existing congestion regulation measures are difficult to achieve full coverage processing. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0007] In one aspect, the present invention provides a method for classifying and processing information interaction data, comprising:

[0008] Get the data combination to be sorted;

[0009] Classifying the data combination to be sorted by using a pre-established data classifier to obtain the classification type of each data in the data combination to be sorted;

[0010] According to the classification type of each data, the basic sorting rule corresponding to each data is matched from the preset basic sorting rule library;

[0011] Sorting the data combinations to be sorted based on the basic sorting rules corresponding to each data to obtain a sending order of the data combinations;

[0012] Each data in the data combination to be sorted is distributed and processed in turn according to the sending order of the data combination.

[0013] Preferably, the method further comprises:

[0014] According to the logical range communication architecture, the trial training network environment is divided into node level, local area network level and range level. Among them, the trial training network environment includes multiple nodes, multiple local area networks and multiple physical ranges;

[0015] The basic sorting rules at the node level, the basic sorting rules at the local area network level and the basic sorting rules at the range level are designed in advance respectively. Among them, the basic sorting rules at the node level, the basic sorting rules at the local area network level and the basic sorting rules at the range level constitute the preset basic sorting rule library.

[0016] Preferably, the design of the basic sorting rules at the node level, the basic sorting rules at the local area network level and the basic sorting rules at the range level in advance respectively specifically includes:

[0017] Based on the node data type, the data sending order between nodes is designed at the node level to obtain the basic sorting rules at the node level;

[0018] On the basis of the sorting at the node level, regarding the local area network as a node, the data sending order between local area networks is designed at the local area network level based on the node importance to obtain the basic sorting rules at the local area network level;

[0019] On the basis of the sorting at the local area network level, regarding the range as a node, the data sending order between ranges is designed at the range level based on the business urgency to obtain the basic sorting rules at the range level.

[0020] Preferably, the data classifier is a decision tree classifier constructed using the random forest algorithm.

[0021] Preferably, the process of constructing a decision tree classifier using the random forest algorithm is as follows:

[0022] Assume that the size of the training set T is N, the number of features is Q, and the size of the random forest is K. Traverse the size of the random forest K times:

[0023] Step a: Sample N times with replacement from the training set T to form a new sub-training set D, where the training set T includes all data combinations in the trial network environment;

[0024] Step b: Randomly select q features, where q < Q. Among them, the features in the trial network environment include node data type, node importance, business urgency, sending frequency, and network level;

[0025] Step c: Use the new sub-training set D and the q features to learn a complete decision tree;

[0026] Repeat the above steps a - c to obtain a random forest.

[0027] Preferably, the decision tree classifier is a decision tree classifier constructed based on sparse matrix features, wherein the sparse matrix M of the decision tree classifier is:

[0028]

[0029] in,

[0030]

[0031] in,

[0032]

[0033] Where, M represents the physical range matrix, i represents the physical range number; M i represents the LAN matrix in the ith physical range, j represents the LAN number in the ith physical range; M i,j represents the node matrix in the jth local area network in the i-th physical range, and k represents the node number in the jth local area network in the i-th physical range.

[0034] Preferably, the classification design of the decision tree classifier is:

[0035] For the element m of the interaction between the lth node and the hth node in the sparse matrix M l,h , 1 indicates that the current data combination contains the element, and 0 indicates that the element is not included;

[0036] When data ,and When the data m l,h The data interaction between the corresponding ranges belonging to level 1 in the decision tree adopts the basic sorting rules of the range level;

[0037] When data ,and When the data m l,h The data interaction between the corresponding local area networks belonging to level 2 in the decision tree adopts the local area level basic sorting rule;

[0038] When data ,and When the data m l,h The data interaction between corresponding nodes belonging to level 3 in the decision tree adopts the node-level basic sorting rule.

[0039] On the other hand, the present invention also provides a classification processing system for information interaction data, comprising:

[0040] A data acquisition module is configured to acquire a combination of data to be sorted;

[0041] A classification processing module is configured to classify the data combination to be sorted by using a pre-established data classifier to obtain a classification type of each data in the data combination to be sorted;

[0042] A rule matching module is configured to match the basic sorting rule corresponding to each data from a preset basic sorting rule library according to the classification type of each data;

[0043] A sorting processing module is configured to sort the data combinations to be sorted based on the basic sorting rules corresponding to each data, so as to obtain a sending order of the data combinations;

[0044] And, the data sending module is configured to distribute and process each data in the data combination to be sorted in sequence according to the data combination sending order.

[0045] On the other hand, the present invention further provides an electronic device, comprising: a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the classification processing method of information interaction data as described above is implemented.

[0046] On the other hand, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for classifying and processing information interaction data as described above is implemented.

[0047] The present invention has the following advantages:

[0048] The classification and processing method of information interaction data of the present invention first pre-designs a preset basic sorting rule library to provide a basic data sorting function, and pre-constructs a data classifier to provide the object and basis of classification; then, the data classifier is used to classify the data combination to be sorted to clarify its classification type, and then the corresponding basic sorting rules are clarified according to the classification type, so as to sort the data combination to obtain the data combination sending order, and finally, the data combination to be sorted is distributed and processed according to the data combination sending order, which can realize the clear classification and sorting of congested data combinations, thereby solving the full coverage problem of data combinations and improving the congestion regulation coverage performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the flow chart of some embodiments of a method for classifying and processing information interaction data of the present invention;

[0050] Figure 2 A schematic diagram of a three-layer structure of a data source in a trial training network environment of the present invention;

[0051] Figure 3 A schematic diagram of a decision tree and a random forest structure of the present invention;

[0052] Figure 4 It is a schematic diagram of the structure of some embodiments of a classification processing system for information interaction data of the present invention;

[0053] Figure 5 The present invention is a structural block diagram of some embodiments of an electronic device of the present invention. DETAILED DESCRIPTION

[0054] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0055] In some embodiments, see Figure 1 As shown, a classification processing method for information interaction data in an embodiment of the present invention includes:

[0056] Step 100: Obtain a data combination to be sorted;

[0057] In this step, when a new data source combination appears under congestion, the data source combination is obtained. The data source combination needs to be sent after the data sending order is determined. Here, it is defined as the data combination to be sorted.

[0058] Step 200: Classify the data combination to be sorted by using a pre-established data classifier to obtain the classification type of each data in the data combination to be sorted;

[0059] In this step, the data combinations to be sorted are classified by a pre-established data classifier. Once the classification is clear, the data can be sent and sorted using known logical judgments, thereby achieving transmission order management.

[0060] Step 300: According to the classification type of each data, a basic sorting rule corresponding to each data is matched from a preset basic sorting rule library;

[0061] In this step, a basic sorting rule library is pre-designed through offline identification or expert experience. The preset basic sorting rule library includes multiple basic sorting rules. The corresponding basic sorting rules can be determined according to the classification of the data, so as to determine the data sending order according to the corresponding basic sorting rules.

[0062] Step 400: sorting the data combinations to be sorted based on the basic sorting rules corresponding to each data to obtain a sending order of the data combinations;

[0063] In this step, after determining the basic sorting rule corresponding to each data, the data combination sending order of the entire data combination to be sorted can be obtained.

[0064] Step 500: Distribute and process each data in the data combination to be sorted in turn according to the order in which the data combinations are sent.

[0065] The classification and processing method of information interaction data in the embodiment of the present invention first pre-designs a preset basic sorting rule library to provide a basic data sorting function, and pre-constructs a data classifier to provide the object and basis of classification; then, the data classifier is used to classify the data combination to be sorted to clarify its classification type, and then the corresponding basic sorting rule is clarified according to the classification type, so as to sort the data combination to obtain the data combination sending order, and finally, the sorted data combination is distributed according to the data combination sending order, so as to achieve a clear classification and sorting solution for congested data combinations, thereby solving the full coverage problem of data combinations and improving the congestion regulation coverage performance. In addition, it has the ability to perceive and regulate the network environment globally, avoids the limitations of a single node environment, adapts to task requirements to the maximum extent, and significantly improves the regulation effect.

[0066] In some embodiments, the classification processing method of information interaction data of the present invention further includes:

[0067] According to the logical range communication architecture, the test network environment is divided into node level, LAN level and range level.

[0068] Among them, the test network environment includes multiple nodes, multiple local area networks and multiple physical shooting ranges;

[0069] See also Figure 2 As shown, in this embodiment, the test training network environment is regarded as a logical shooting range, which includes multiple physical shooting ranges. There is data interaction between each physical shooting range. Each physical shooting range includes a local area network composed of multiple nodes. There is data interaction between local area networks and between nodes. In this embodiment, based on the combination of data sources, the data source of the test training network environment is divided into a three-level structure, namely, node level, local area network level and shooting range level, and corresponding sorting rules are constructed at different levels.

[0070] The node-level basic sorting rules, the local area network-level basic sorting rules and the range-level basic sorting rules are pre-designed respectively, wherein the node-level basic sorting rules, the local area network-level basic sorting rules and the range-level basic sorting rules constitute a preset basic sorting rule library.

[0071] In this embodiment, node-level basic sorting rules, local area network-level basic sorting rules, and range-level basic sorting rules are pre-designed according to corresponding sorting optimization indicators.

[0072] In some embodiments, the classification and processing method of information interaction data of the present invention pre-designs basic node-level sorting rules, basic local area network-level sorting rules, and basic range-level sorting rules, specifically including:

[0073] Design the data sending order between nodes at the node level based on the node data type to obtain the basic sorting rules at the node level;

[0074] On the basis of node level sorting, the LAN is regarded as a node, and the data transmission order between LANs is designed at the LAN level based on the importance of the node, so as to obtain the basic sorting rules at the LAN level;

[0075] On the basis of LAN hierarchical sorting, the target range is regarded as a node, and the data sending order between target ranges is designed at the target range level based on business urgency to obtain the basic sorting rules at the target range level.

[0076] In this embodiment, the sorting rules are first designed at the node level, and then expanded to the LAN level based on the node-level sorting, that is, the LAN is regarded as a node, and the data transmission order at the LAN level is designed; further expanded to the physical range level, that is, the physical range is regarded as a node, and the data transmission order between physical ranges is designed. In this way, the data transmission of the entire test environment is clearly layered, the communication link is clear, and basic sorting problems can be solved.

[0077] The specific design method of basic sorting rules is:

[0078] Basic rankings include node ranking, local area network ranking and range ranking, which are obtained through offline identification or expert experience.

[0079] (1) Node sorting design

[0080] For the test training network, data sorting at the node level is the basis for achieving sorting of the entire network, and node sorting is also the basis for sorting at the range level. For commonly occurring data source combinations, a set of sorting rules is constructed to solve the order management under different combinations.

[0081] First, based on the running process of the trial task, several typical data source combinations are selected to form basic node-level sorting samples. Indicates the data source in the node. i represents the node number, and the maximum value is the maximum number of nodes in the network environment; j represents the data type, and the values ​​are 1, 2, and 3. 1 represents the remote procedure call class, 2 represents the data distribution class, and 3 represents the message queue class. For example Represents the message queue data of the 9th node. Within a certain node, data sorting is mainly carried out based on three types of data, namely the value of j. The following relational judgment formula gives the sorting design.

[0082] Take a sampling of 3 data as an example, which are , , , we can know from the permutation and combination method that since x1, x2, and x3 can all take the values ​​1, 2, and 3, there are Since the value of j is proportional to its importance, the design ranking is as follows:

[0083] Arrays , the order of elements in the array is also the initial order;

[0084] when hour, , Adjust the order, i.e. ;

[0085] when hour, , Adjust the order, ;

[0086] when hour, , Adjust the order, .

[0087] Arrays The final order of the elements in is the order of the data source.

[0088] (2) LAN sorting design

[0089] At the LAN level, data sources and data sinks are both nodes within the LAN. Business scenarios and features include: data source refers to the total amount of data of a node as a whole, that is, the node throughput demand; sorting object refers to the combination of data sources at that point in time when the throughput demand of multiple data sources exceeds the LAN bandwidth; sorting optimization indicators include priority for important nodes, priority for urgent business needs, priority for nodes with low sending frequency, etc.

[0090] Assume that a local area network includes two switches and six resource nodes. Each switch is connected to three resource nodes, and then the two switches are interconnected. When the three nodes in switch 1 need to transmit data with three different nodes in switch 2, three groups of transmission links need to be established and sorted. The priority of important nodes in the data sink is used as the optimization indicator.

[0091] Arrays , i refers to the LAN number, , , Refers to the link number. The numbering order is consistent with the importance.

[0092] Arrays , the order of elements in the array is also the initial order.

[0093] when hour, , Adjust the order, i.e. ;

[0094] when hour, , Adjust the order, i.e. ;

[0095] when hour, , Adjust the order, i.e. .

[0096] The final order of the elements in array L is the ordering of the data links.

[0097] (3) Shooting range ranking design

[0098] The range level includes physical range and logical range. The physical range is for data interaction between LANs. The data source and data sink belong to the same physical range but not the same LAN. The logical range is for data interaction between physical ranges. The data source and data sink belong to different physical ranges within the logical range. The ranking indicator of the range level is mainly business urgency.

[0099] Assume that a physical range includes three LANs, which are interconnected through core switches. Each core switch is connected to several switches within the LAN. When nodes in different LANs interact with each other, they need to interact through the core switch.

[0100] Arrays , i refers to the physical range number, , , Refers to the link number. The numbering sequence is consistent with the urgency of the service.

[0101] Arrays , the order of elements in the array is also the initial order.

[0102] when hour, , Adjust the order, i.e. ;

[0103] when hour, , Adjust the order, i.e. ;

[0104] when hour, , Adjust the order, i.e. .

[0105] Arrays The final order of the elements in is the ordering of the data links.

[0106] Similar to the physical range, the logical range can be designed and ordered as follows:

[0107] Arrays , i refers to the logical range number, usually 1; , , Refers to the link number. The numbering sequence is consistent with the urgency of the service.

[0108] Arrays , the order of elements in the array is also the initial order.

[0109] when hour, , Adjust the order, i.e. ;

[0110] when hour, , Adjust the order, i.e. ;

[0111] when hour, , Adjust the order, i.e. .

[0112] The final order of the elements in array N is the ordering of the data links.

[0113] In some embodiments, the data classifier in the classification processing method of information interaction data of the present invention is a decision tree classifier constructed using a random forest algorithm.

[0114] When there are new data groups that need to be sorted, the random forest algorithm is used to classify the new data combinations.

[0115] Based on the node sorting, a decision tree is constructed for the LAN sorting and the range sorting. Figure 3 The decision tree and random forest structures in the experimental training network environment are given.

[0116] In the test field environment, simulate the operation of the task. During the operation, continuously collect data combinations, conduct training at the range, local area network and node levels, and obtain the corresponding logical judgment models. When encountering data combinations that need to be classified in the field environment, use these already acquired logical judgment models for classification.

[0117] Specifically, in the method for classifying and processing information interaction data of the present invention according to the embodiments of the present invention, the process of constructing a decision tree classifier using the random forest algorithm is as follows:

[0118] Assume that the size of the training set T is N, the number of features is Q, and the size of the random forest is K. Traverse the random forest K times:

[0119] Step a: Sample N times with replacement from the training set T to form a new sub-training set D, where the training set T includes all data combinations in the test network environment;

[0120] Step b: Randomly select q features, where q < Q, and the features in the test network environment include node data type, node importance, service urgency, sending frequency, and network level;

[0121] Step c: Use the new sub-training set D and q features to learn a complete decision tree;

[0122] Repeat the above steps a-c to obtain a random forest.

[0123] Preferably, the decision tree classifier of the present invention according to the embodiments of the present invention is a decision tree classifier constructed based on sparse matrix features. The designs of node sorting, local area network sorting, and range sorting are sortings under typical data source combinations, and known sorting rules are designed. During the execution of the test training task, when a new data source combination appears under congestion conditions, classification is performed through a random forest decision tree. Once the classification attribution is clear, known logical judgments will be used for data sending sorting, thereby realizing the control of the transmission order. For this reason, the embodiments of the present invention propose a classification method and classifier based on sparse matrix feature extraction. Specifically as follows:

[0124] All nodes are arranged diagonally in the sparse matrix, and according to the division of the physical range and the local area network within the range of the test and training network, the physical range, local area network, and each node are arranged in sequence.

[0125] Among them, the sparse matrix M for constructing the decision tree classifier is:

[0126]

[0127] Among them,

[0128]

[0129] Among them,

[0130]

[0131] Among them, M represents the physical range matrix, and i represents the physical range number; M irepresents the LAN matrix in the ith physical range, j represents the LAN number in the ith physical range; M i,j represents the node matrix in the jth local area network in the i-th physical range, and k represents the node number in the jth local area network in the i-th physical range.

[0132] Furthermore, the classification design of the decision tree classifier in this embodiment is:

[0133] For the element m of the interaction between the lth node and the hth node in the sparse matrix M l,h , 1 means the current data combination contains the element, 0 means it does not contain the element;

[0134] When data ,and When, data The data interaction between corresponding ranges belonging to level 1 in the decision tree adopts the basic sorting rules at the range level;

[0135] When data ,and When, data The data interaction between the corresponding local area networks belonging to level 2 in the decision tree adopts the basic sorting rules of the local area level;

[0136] When data ,and When, data The data interaction between corresponding nodes belonging to level 3 in the decision tree adopts the basic sorting rules at the node level.

[0137] In summary, the classification and processing method of the information interaction data of the embodiment of the present invention is based on the logical range communication architecture, and divides the test network environment into three levels, namely the node level, the local area network level and the range level. The congestion is classified and sorted according to the three levels, and the basic sorting of the node level, the local area network level and the range level is designed to provide a basis for the classification and processing of the test network environment; then, on the basis of the basic sorting design, a decision tree is constructed for the test network environment through the random forest algorithm to provide classification support for various types of random data source combinations, and sorting rules are matched through classification; thirdly, a sparse matrix is ​​constructed to provide an analysis basis for random data sources, and accurate classification of any combination is achieved. The classification and processing method of the information interaction data of this embodiment constructs a full coverage classification processing method for any random data source combination from the overall perspective of the network environment through the application of decision trees and sparse matrices, avoiding the insufficiency of the difficulty in balancing sorting effect and coverage, and effectively improving the congestion regulation service capability of a multi-node large network environment.

[0138] On the other hand, see Figure 4 As shown, the embodiment of the present invention also provides a classification processing system for information interaction data, including:

[0139] The data acquisition module 10 is configured to acquire a combination of data to be sorted;

[0140] The classification processing module 20 is configured to classify the data combination to be sorted by using a pre-established data classifier to obtain a classification type of each data in the data combination to be sorted;

[0141] The rule matching module 30 is configured to match the basic sorting rule corresponding to each data from a preset basic sorting rule library according to the classification type of each data;

[0142] The sorting processing module 40 is configured to sort the data combination to be sorted based on the basic sorting rule corresponding to each data to obtain the sending order of the data combination;

[0143] Furthermore, the data sending module 50 is configured to distribute and process each data in the data combination to be sorted in sequence according to the data combination sending order.

[0144] The specific details of each module in the above-mentioned classification processing system for information interaction data have been described in detail in the corresponding classification processing method for information interaction data, so they will not be repeated here.

[0145] On the other hand, an embodiment of the present invention further provides an electronic device, including: a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the classification processing method of information interaction data described in the above embodiment is implemented.

[0146] Specifically, the above-mentioned memory and processor can be general-purpose memory and processor, which are not specifically limited here. When the processor runs the computer-readable instructions stored in the memory, it can perform the classification processing of the information interaction data described in the above-mentioned embodiment.

[0147] Figure 5 is a structural diagram of an electronic device according to an embodiment of the present invention, such as Figure 5As shown, the electronic device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a device for process modeling. The internal memory provides an environment for the operation of the operating system and the device for process modeling in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the housing of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0148] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a structural diagram of the part related to the technical solution of the present disclosure, and does not constitute a limitation on the electronic device to which the technical solution of the present application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the classification and processing method of information interaction data described in the above embodiment is implemented.

[0150] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0151] It should be noted that the above detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present application belongs.

[0152] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments described in the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0153] 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 sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0154] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0155] For ease of description, spatially relative terms, such as "above", "above", "on the upper surface of", "above", etc., may be used herein to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" may include both "above" and "below". The device may also be positioned in other different ways, such as rotated 90 degrees or in other orientations, and the spatially relative descriptions used herein are interpreted accordingly.

[0156] In the above detailed description, reference is made to the accompanying drawings, which form a part of this document. In the accompanying drawings, similar symbols typically identify similar components unless the context indicates otherwise. The illustrated embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be used, and other changes may be made, without departing from the spirit or scope of the subject matter presented herein.

[0157] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A classification processing method for information interaction data, characterized in that: Including: Obtain the data combination to be sorted; Classify the data combination to be sorted through a pre-established data classifier to obtain the classification type of each data in the data combination to be sorted; Match the basic sorting rule corresponding to each data from a preset basic sorting rule library according to the classification type of each data; Perform sorting processing on the data combination to be sorted based on the basic sorting rule corresponding to each data to obtain the data combination sending order; Perform distribution processing on each data in the data combination to be sorted in sequence according to the data combination sending order; Among them, according to the logical range communication architecture, the trial training network environment is divided into node level, local area network level and range level. The trial training network environment includes multiple nodes, multiple local area networks and multiple physical ranges; Design the node-level basic sorting rule, local area network-level basic sorting rule and range-level basic sorting rule respectively in advance. Among them, the node-level basic sorting rule, the local area network-level basic sorting rule and the range-level basic sorting rule constitute the preset basic sorting rule library; The steps of respectively designing the node-level basic sorting rule, local area network-level basic sorting rule and range-level basic sorting rule in advance specifically include: Design the data sending order between nodes at the node level based on the node data type to obtain the node-level basic sorting rule; On the basis of sorting at the node level, regard the local area network as a node, and design the data sending order between local area networks at the local area network level based on the node importance to obtain the local area network-level basic sorting rule; On the basis of sorting at the local area network level, regard the range as a node, and design the data sending order between ranges at the range level based on the business urgency to obtain the range-level basic sorting rule.

2. The classification processing method of information interaction data according to claim 1 is characterized in that: The data classifier is a decision tree classifier constructed by using the random forest algorithm.

3. The classification processing method of information interaction data according to claim 2 is characterized in that: The process of constructing a decision tree classifier by using the random forest algorithm is: Assume that the size of the training set T is N, the number of features is M, and the size of the random forest is K. Traverse the random forest K times: Step a: Sample N times with replacement from the training set T to form a new sub-training set D, where the training set T includes all data combinations in the test network environment; Step b: Randomly select m features, where m < M, and the features in the test network environment include node data type, node importance, business urgency, sending frequency and network level; Step c: Use the new sub-training set D and the m features to learn a complete decision tree; Repeat the above steps a-c to obtain a random forest.

4. The classification processing method of information interaction data according to claim 3 is characterized in that: The data classifier is a decision tree classifier constructed based on sparse matrix features. The sparse matrix M for constructing the random forest is: Where, Where, Where M represents the physical range matrix, i represents the physical range number; M i represents the LAN matrix in the ith physical range, j represents the LAN number in the ith physical range; M i,j represents the node matrix in the jth local area network in the i-th physical range, and k represents the node number in the jth local area network in the i-th physical range.

5. The classification processing method of information interaction data according to claim 4 is characterized in that: The classification design of the data classifier is: For the element m of the interaction between the lth node and the hth node in the sparse matrix M l,h , 1 indicates that the current data combination contains the element, and 0 indicates that the element is not included; When the data m l,h =1, and m l,h ∈M, the data m l,h The data interaction between the corresponding ranges belonging to level 1 in the decision tree adopts the basic sorting rules of the range level; When the data m l,h =1, and m l,h ∈M i When the data m l,h The data interaction between the corresponding local area networks belonging to level 2 in the decision tree adopts the local area level basic sorting rule; When the data m l,h =1, and m l,h ∈M ij When the data m l,h The data interaction between corresponding nodes belonging to level 3 in the decision tree adopts the node-level basic sorting rule.

6. A classification processing system for information interaction data, characterized in that: Including: A data acquisition module configured to acquire the data combination to be sorted; A classification processing module configured to classify the data combination to be sorted through a pre-established data classifier to obtain the classification type of each data in the data combination to be sorted; A rule matching module is configured to match the basic sorting rule corresponding to each data from a preset basic sorting rule library according to the classification type of each data; A sorting processing module is configured to sort the data combinations to be sorted based on the basic sorting rules corresponding to each data, so as to obtain a sending order of the data combinations; And, a data sending module is configured to distribute and process each data in the data combination to be sorted in sequence according to the data combination sending order; Wherein, the test training network environment is divided into a node level, a local area network level and a range level according to the logical range communication architecture, and the test training network environment includes a plurality of nodes, a plurality of local area networks and a plurality of physical ranges; Pre-designing node-level basic sorting rules, local area network-level basic sorting rules and shooting range-level basic sorting rules respectively, wherein the node-level basic sorting rules, the local area network-level basic sorting rules and the shooting range-level basic sorting rules constitute the preset basic sorting rule library; The node-level basic sorting rules, the local area network-level basic sorting rules and the range-level basic sorting rules are respectively pre-designed, specifically including: Designing a data transmission order between nodes at the node level based on the node data type to obtain the node-level basic sorting rule; On the basis of node level sorting, the LAN is regarded as a node, and the data transmission order between LANs is designed at the LAN level based on the importance of the node, so as to obtain the basic sorting rule at the LAN level; On the basis of the LAN level sorting, the shooting range is regarded as a node, and the data sending order between shooting ranges is designed at the shooting range level based on the business urgency to obtain the shooting range level basic sorting rule.

7. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the classification processing method for information interaction data according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for classifying and processing information interaction data as described in any one of claims 1 to 5 is implemented.

Images