CN118468532B

CN118468532B - Fusion processing method and system based on polymorphic information data

Info

Publication number: CN118468532B
Application number: CN202410527158.8A
Authority: CN
Inventors: 周成军; 曹占广; 李�浩; 曲玉言; 刘若青; 金锦; 司光亚; 吴�琳; 尹宗润; 孟祥林; 韩红亮; 宋士兵; 张体奎; 刘洋; 曹毅; 王飞; 刘经德; 唐宇波; 杨镜宇; 周文
Original assignee: National Defence University Of People's Liberation Army Joint Operation Institute
Current assignee: National Defence University Of People's Liberation Army Joint Operation Institute
Priority date: 2024-04-29
Filing date: 2024-04-29
Publication date: 2024-10-25
Anticipated expiration: 2044-04-29
Also published as: CN118468532A

Abstract

The invention belongs to the technical field of geographic information and computer graphics, and provides a fusion processing method and system based on polymorphic information data. The method comprises the following steps: according to entity attributes corresponding to each chess piece in the chess situation data, the deduction data of each chess piece is mapped from a high-dimensional space to a two-dimensional matrix in a two-dimensional space so as to form a filtering condition; performing aggregation calculation on the attributes of the chessmen entities to generate a display control matrix of the situation display elements; transforming the control matrix formed by the implicit and explicit control according to the behavior cognition calculation result; evaluating the behavior tendency of a user, and transforming the implicit control matrix; when receiving user input, the transformed implicit and explicit control matrix is obtained to filter the soldier chess situation data and event situation data. The method and the device can reduce the time complexity of calculation, can rapidly switch the display and hidden states of the situation information, and remarkably reduce the number of rendering elements, thereby greatly improving the visual rendering efficiency of the situation and reducing the interference of the situation information.

Description

Fusion processing method and system based on polymorphic information data

Technical Field

The invention relates to the technical field of geographic information and computer graphics, in particular to a fusion processing method and system based on polymorphic information data.

Background

The soldier chess is a tool for simulating and judging the game activities of two or more parties by using a map and chessmen which visually represent the battlefield environment and the countermeasure force and taking rules (data) formed by practical experience statistical analysis as the main and models as the auxiliary.

In the field of soldier chess simulation, based on different battlefield domains such as land battlefield, sea battlefield, air battlefield, electronic battlefield, network battlefield, space battlefield and the like, a large number of various countermeasure entities are gathered, and the chess situation data and event situation data in the simulation deduction process have the characteristics of massive, heterogeneous, multi-source and other big data. The large-scale polymorphic data greatly increases the information capacity of the patterns of the soldier chess situation. However, for the existing method for realizing the large-scale polymorphic information data visualization of the simulation situation of the chess, the operation is complex and the time is long.

The soldier chess situation relies on the chessboard, the chessmen and related event information to comprehensively display real-time situation information. The multi-mode expression situation display comprises a basic situation of the chess and a key expansion event situation. In addition, there is still a great room for improvement in various aspects such as how to effectively perform real-time comprehensive display control on complex situation information composed of chessboard, chessmen, events, tasks and the like, how to perform section display or appointed display on the complex information, how to reduce information overload, reduce information content confusion, and how to assist in improving understanding ability of users on battlefield situation information.

Therefore, it is necessary to provide a fusion processing method and system for polymorphic information data to solve the above-mentioned problems.

Disclosure of Invention

The invention aims to provide a fusion processing method and system based on polymorphic information data, which are used for solving the technical problems that in the prior art, real-time comprehensive display of large-scale complex situation information consisting of chessboard, chessmen, events, tasks and the like cannot be effectively realized, how to perform section display or appointed display on the complex information, how to reduce information overload, reduce information content confusion, how to assist in improving the understanding capability of users on battlefield situation information and the like.

The first aspect of the invention provides a fusion processing method based on polymorphic information data, comprising the following steps: according to entity attributes corresponding to each chess piece in the chess situation data, the deduction data of each chess piece is mapped from a high-dimensional space to a two-dimensional matrix in a two-dimensional space so as to form a filtering condition; performing aggregation calculation on the entity attributes of each piece to generate a display control matrix of the situation display element; evaluating the behavior tendency of a user, and transforming the implicit control matrix; when receiving user input, the transformed implicit and explicit control matrix is obtained to filter the soldier chess situation data and event situation data.

According to an alternative embodiment, the aggregating and calculating the entity attribute of each chess piece to generate the implicit control matrix of the situation display element includes: dividing each group of two-dimensional matrixes in a plurality of groups of two-dimensional matrixes according to a specified dividing granularity, and dividing the two-dimensional matrixes into a plurality of two-dimensional matrixes to form a new two-dimensional matrix; performing aggregation calculation on the formed new two-dimensional matrix to obtain an intermediate implicit control matrix; and merging the intermediate implicit control matrix generated by merging to obtain a final implicit control matrix.

According to an alternative embodiment, further comprising: singular value decomposition is carried out on a first matrix and a second matrix to be polymerized, and then the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix:

Σ_c＝αΣ_a+βΣ_b

Wherein, sigma_c represents that the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix; Σ_a represents a first matrix obtained by performing singular value decomposition on a first matrix a to be aggregated; Σ_b represents a second matrix obtained by performing singular value decomposition on a second matrix b to be aggregated; the α and β are weight coefficients corresponding to the first matrix and the second matrix after singular value decomposition, respectively.

According to an alternative embodiment, further comprising: the entropy method is used for assigning the weight coefficient, and the calculation formula of the alpha weight coefficient is as follows:

α＝-lnnp(a)lnp(a)

wherein alpha is a weight coefficient corresponding to the first matrix a after singular value decomposition; the first matrix a is a matrix of n rows and m columns, n is the total number of rows of the first matrix a, and ln represents a logarithmic operation based on e.

P (a) =Σ ⁿ _i＝1,a_ij,a_ij is an element of the ith row and j column in the first matrix a after singular value decomposition, i and j are positive integers, and n is the total number of rows of the first matrix a.

According to an alternative embodiment, the evaluating the user behavioral tendencies comprises: a neural network is adopted to establish a user behavior tendency prediction model, a display state formed by a subordinate party, a battle zone, a user type and an actual state is used to form a user display state label, a positive sample and a negative sample are defined, user input parameters marked with the user display state label are used as a training data set, and the user behavior tendency prediction model is trained to obtain a trained user behavior tendency prediction model; the user input parameters are input into a trained user behavior tendency prediction model, and user behavior tendency predicted values corresponding to the user input parameters are output, wherein the user behavior tendency predicted values are multi-dimensional vectors or multi-dimensional matrixes.

According to an alternative embodiment, the feature matrix is screened, and the implicit control matrix is screened according to the feature matrix, so that situation display matched with the behavior tendency of the user is obtained.

According to an alternative embodiment, a chain address method is adopted, and a multi-level index matrix is established according to attribute information or data corresponding to each piece after classification processing and layering processing, wherein the multi-level index matrix comprises association relations of different levels and association relations of different types of the same level.

A second aspect of the present invention provides a fusion processing system based on polymorphic information data, which uses the fusion processing method based on polymorphic information data according to the first aspect of the present invention, the fusion processing system comprising: the forming processing module is used for mapping deduction data of each chess piece from a high-dimensional space to a two-dimensional matrix in a two-dimensional space according to entity attributes corresponding to each chess piece in the chess situation data so as to form filtering conditions; the generation processing module is used for carrying out aggregation calculation on the attributes of each entity to generate a display and hidden control matrix of the situation display element; the evaluation processing module evaluates the behavior tendency of the user and transforms the implicit control matrix; and the filtering processing module is used for acquiring the transformed implicit and explicit control matrix when receiving user input so as to filter the soldier chess situation data and the event situation data.

A third aspect of the present invention provides an electronic apparatus, comprising: one or more processors; a storage means for storing one or more programs; the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the fusion processing method based on polymorphic information data according to the first aspect of the present invention.

A fourth aspect of the present invention provides a computer-readable medium having stored thereon a computer program which, when executed by a processor, implements the fusion processing method based on polymorphic information data according to the first aspect of the present invention.

The invention has the beneficial effects that:

Compared with the prior art, the invention maps the deduction data of each chess piece from a high-dimensional space to a two-dimensional matrix in a two-dimensional space according to the entity attribute corresponding to each chess piece in the chess situation data to form a filtering condition; performing aggregation calculation on the entity attributes of each piece to generate a display-hidden control matrix of the situation display element, and transforming the display-hidden control matrix by evaluating the behavior tendency of the user; when user input is received, the transformed display and hidden control matrix is obtained to filter the soldier chess situation data and event situation data, so that the calculation time complexity can be reduced, the display and hidden states of the situation information can be rapidly switched, the number of rendering elements is obviously reduced, the situation visual rendering efficiency is greatly improved, and meanwhile, the situation information interference is reduced.

Drawings

FIG. 1 is a flow chart of steps of an example of a fusion processing method based on polymorphic information data of the present invention;

FIG. 2 is a block diagram of an example of entity attribute classification in the fusion processing method based on polymorphic information data of the present invention;

FIG. 3 is a schematic diagram showing an example of a tag index relation of a tag index table structure in a fusion processing method based on polymorphic information data according to the present invention;

FIG. 4 is a schematic diagram of user trend attribute partitioning in an example of a fusion processing method based on polymorphic information data to which the present invention is applied;

FIG. 5 is a schematic diagram of an example of a fusion processing system based on polymorphic information data of the present invention;

FIG. 6 is a schematic structural diagram of an embodiment of an electronic device according to the present invention;

fig. 7 is a schematic diagram of an embodiment of a computer readable medium according to the present invention.

Detailed Description

It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.

In view of the above, the present invention provides a fusion processing method based on polymorphic information data. According to entity attributes corresponding to each chess piece in the soldier chess situation data, the deduction data of each chess piece are mapped from a high-dimensional space to a two-dimensional matrix in a two-dimensional space to form filtering conditions; performing aggregation calculation on the entity attributes of each piece to generate a display-hidden control matrix of the situation display element, and transforming the display-hidden control matrix by evaluating the behavior tendency of the user; when user input is received, the transformed display and hidden control matrix is obtained to filter the soldier chess situation data and event situation data, so that the calculation time complexity can be reduced, the display and hidden states of the situation information can be rapidly switched, the number of rendering elements is obviously reduced, the situation visual rendering efficiency is greatly improved, and meanwhile, the situation information interference is reduced.

Example 1

The simulation training scenario generation method of the present invention will be described in detail with reference to fig. 1,2, 3 and 4.

Fig. 1 is a flowchart showing steps of an example of a fusion processing method based on polymorphic information data according to the present invention.

First, in step S101, according to the entity attribute corresponding to each piece in the soldier chess situation data, the deduction data of each piece is mapped from the high-dimensional space to the two-dimensional matrix in the two-dimensional space, so as to form the filtering condition.

Specifically, according to entity attributes corresponding to each chess piece in the soldier chess situation data, deduction data of each chess piece is mapped from a high-dimensional space to a two-dimensional matrix in a two-dimensional space.

In this example, the chess situation data specifically refers to related data to be displayed according to a task scene or a user requirement, for example, in a business task scene displayed with a countermeasure capability, where the chess situation data specifically includes large-scale data related to countermeasure entities such as personnel, infrastructure, countermeasure objects, and countermeasure targets of each party participating in a countermeasure process, the number of the involved entities is 100 tens of thousands, and the data amount can reach 10 billions.

In one embodiment, in a pawn system, the pawn is a representation of a countermeasure entity or a countermeasure event. The attribute of each entity (hereinafter, simply referred to as "entity attribute") may reach as many as hundred classes, and each class of entity attribute is an independent attribute (for example, "independent attribute 1" in fig. 2), or is a combined attribute having a subordinate relationship (for example, major class 1-minor class 11 and minor class 12-sub-class 111 and sub-class 112 in fig. 2), and particularly, refer to fig. 2.

Specifically, the attribute value space of the entity attribute is typically a high-dimensional space, and the entity attribute information (i.e., attribute information of the chess pieces) in the high-dimensional space specifically includes a affiliated party, group category information (e.g., army category information), group type, group prototype, group passion number, group level, superior group, position, speed, movement direction, task, target, and the like. The attribute information of each chess piece is stored in a JSON format, for example, and each type of entity attribute is stored in a numerical form, for example, data access is performed through an index.

Further, the entity attribute is configured in the form of a chess piece card, and specifically includes: the chess pieces of the army entity, the chess pieces of the army entity and the chess pieces of the battlefield facility. The chess card of the army entity comprises a basic attribute label, an affiliated army label, a hitting strength label, a defending strength label, a guaranteeing material demand label and the like. The chess piece card of the arming entity comprises a basic attribute label, an affiliated army label, an army label, a guarantee label and the like. Chess pieces cards of battlefield facilities include basic attribute tags, battle strength tags, defend strength tags, etc. Each type of pawn card contains basic attribute tags, such as a army entity classification tag, which can be subjected to data screening, such as tag screening, according to time, spatial location, membership, army type, main battle equipment, ammunition consumption number, ammunition residue number, equipment strength attributes.

The chessman card adopts the following scheme according to the label screening:

a tag inverted Index is established, and the data structure of the tag inverted Index comprises three parts, namely an inverted table (Posting List), a Term Dictionary (Term Dictionary) and a Term Index (Term Index). Wherein the term dictionary is a value object dictionary of various labels of the chess pieces and cards. The inverted list stores a chessman card ID list corresponding to the tag value, and the data structure is a numerical array. The term index is an index structure built on the term dictionary, and aims to improve the query speed of the term dictionary.

First, an inverted list of tags is constructed. The inverted list contains all the words that appear in the tag. The inverted list is constructed by traversing all tags, extracting and removing the heavy related vocabulary.

Next, the metadata structure model of the tag represents the data element, i.e., the tag and the Address of each record having the tag, in the form of a tag-Address (Value) Value pair. The processing of the metadata index reduces the number of direct disk accesses. The address is a unique identification number or ID of the attribute in the entire inverted table.

The tag index table structure comprises key values, tag metadata and addresses of the Hash table, wherein the addresses point to node storage positions of next-stage tags or specific numerical value storage positions of the tags. Traversing the tag index table by adopting a term index algorithm. The tag index table structure includes tag index relationships, see in particular fig. 3.

Because the data volume of the chess is extremely large, the space required for storing the inverted list is extremely large, so that data compression is required in implementation. The use of storage space is reduced by compressing data, so that the cost of disk I/O and network transmission can be reduced, and the reading speed is further increased. Common compression algorithms for inverted tables are FOR (Frame Of Reference) and RBM (RoaringBitMap). Since the data ID is typically continuously incremented or stepwise incremented, the value gap is not excessive, and therefore the FOR algorithm is employed. The execution process is divided into the following three steps:

Step S201: the ID list of the data store before compression is an ordered set of values. Such as: 55,300,311,332,443, 472.

Step S202: the difference before the sequence ID is calculated. Such as: 55,245,11,21,111,29, wherein 245=300-55, 11=311-300, and so on.

Step S203: the difference calculated in step S202 is subjected to list blocking, and when each block contains K values, for example, k=3, the above example is divided into two blocks, K1 (55,245,11), K2 (21,111,29), the maximum value in each block is taken, and the maximum number of bits required for storing the block data is calculated. If the maximum value in K1 is 245, 8 bits may be used for storage, then each number in K1 may be stored in 8 bits. Similarly, K2 may be stored in 7bit bits. The number of memory bits required for each block is fixed using one byte (8 bit) memory alone.

In the above example, assuming that the ID type is the int type, the storage is 4*6 =24 bytes before compression, and is 3×8+3×7+2×8=61 bits after compression, 8 bytes are required, so that the use of storage space is significantly reduced.

In another embodiment, a term indexing algorithm is used to build an inverted index term dictionary for binary searching of corresponding data.

The built inverted index term dictionary is an ordered array, each term of the array stores term and corresponding inverted table data ID. And loading the inverted index term dictionary into a memory when inquiring data, and searching by two halves.

In practical implementation, the chessman card label data are classified in layers through a multi-layer Hash function. The data structure is stored by using a multi-stage Hash table and a skip table, and is as follows: the first layer Hash table is K- { a, B, C … }, the second layer Hash table is table access address of a- { A1, a2..an }, and { A1, A2, … An } is stored using a skip table.

Specifically, node type: containing values and pointers to the next node (e.g., child node), and to the next node at each level. The query algorithm queries the storage position of the attribute in a skip list mode, and extracts the value of the corresponding attribute and the node pointer of the related attribute.

When inquiring, firstly, acquiring keys of the Hash table through the Hash function, acquiring a set of a lower-level Hash table, and when the child node is reached, acquiring node data through jump table inquiry. The time complexity of the Hash table query is O (1), the skip table query is O (m×log), and m is the number of nodes to be traversed by each level of index of the skip table, for example, set to 3.

The space complexity of the multi-stage Hash table and the jump table is O (N), the occupied memory of the whole term is excessively large, and only the Hash value and the ID of the object are stored.

In a specific embodiment, taking two labels of the affiliated party and the army type as examples, all chessman cards are traversed first to generate a display control matrix.

Specifically, according to the army compiling level setting of the army chess system, generating an army type tree, wherein the affiliated parties are divided into a red party, a blue party, a green party, a white party and the like, establishing a chess piece index in a two-dimensional array mode, namely, through the army type tree, ordering all chess pieces according to the army compiling level, each affiliated party respectively generates a one-dimensional array of the affiliated party, and the one-dimensional arrays of the affiliated parties are used as column elements and combined into a two-dimensional matrix of n rows and 4 columns according to the sequence of the red party, the blue party, the green party and the white party, and the spare position is supplemented with 0.

Whether each chessman is displayed or not is defined as a display hidden value, and is expressed by a Boolean value, wherein the display is 1 (true), and the display is 0 (false). Thus, a two-dimensional boolean matrix can be obtained, wherein each boolean value represents whether the corresponding pawn is displayed on a situation map.

For example, a tree list of a certain team type is a table of m rows and 1 columns, and the corresponding converted matrix form is as follows:

[a₁₁ a₁₂..a_1m]^T

In the above matrix, each element represents an entity node related to a certain team type hierarchy, each node defines the hierarchy of the index through the hierarchy attribute, and is divided into Top, mid, bott, the Top type node is the topmost node, the Mid type node has an upper level and a lower level, and the Bott type node is only the upper level. The Mid and the Bott type node attributes are kept synchronous with the Top type node attributes, for example, when the Top node filtering control attribute is true, the attributes of all the sub-nodes in the lower stage are synchronous to be true.

Specifically, the entity attribute of each chess piece, namely the chess piece card, is mapped from a high-dimensional space into a plurality of groups of two-dimensional matrixes in a two-dimensional space, specifically, the attribute information of each chess piece after classification is quantized and mapped into a plurality of groups of two-dimensional matrixes.

Further, each set of two-dimensional matrices forms a class of attribute slices. Each class of attribute slices corresponds to a class of thematic views. The sets of two-dimensional matrices are represented, for example, using a boolean matrix, wherein matrix elements are represented using 1 s and matrix elements are not represented using 0 s.

One or more filter condition lists displaying filter conditions are further formed according to the plurality of groups of two-dimensional matrixes. One or more lists of filter criteria are formed, for example, based on parameters of the application task display parameters, user ratings, device types, antagonism ranks, etc.

In practical application, one or more filtering condition lists are formed based on the filtering rule. The filtering rules may also be formed based on the type of entity attributes and the attribute encoding. For example, when the attribute value of the entity attribute is 1, data pushing is performed. When the attribute value of the entity attribute is 0, data pushing is not performed.

Optionally, a list of filtering conditions is generated based on the level affiliate or the level type. For example, the list of filtering conditions mainly comprises a matrix of the army's parties, each element in the matrix representing the presence or absence of the data of the chess, in particular in the form of boolean values; the filtering condition list is a table with m rows and 3 columns, and the corresponding matrix form is as follows:

Wherein b ₁₁、b₁₂、b₁₃、...b_m1、b_m2、b_m3 is used to represent each element in the matrix.

And sequentially carrying out classification processing and layering processing on entity attributes corresponding to each chess piece in the soldier chess situation data of different levels.

And establishing a multi-level index matrix according to the attribute information or data corresponding to each piece after classification and layering treatment by adopting a chain address method, wherein the multi-level index matrix comprises association relations of different levels, association relations of different types of the same level and the like.

Optionally, each multi-level index matrix contains elements for identifying levels and categories and is endowed with appointed values, so that the entity corresponding to the relevant chessman card can be effectively and quickly traversed.

The multi-level index matrix comprises a plurality of layers of matrixes, specifically, a first layer of matrixes is used for storing index information and classification information for quick retrieval, namely, the index information and the classification information of a second layer of matrixes are stored, and the first layer of matrixes are primary hash matrixes. The second layer matrix is used for storing index information and classification information of the third layer matrix, and the second layer matrix is a second-level hash matrix. And by analogy, the h-1 layer matrix is used for storing index information and classification information of the h layer matrix, and the h-1 layer matrix is an h-level hash matrix.

Preferably, a hash algorithm is used to quickly traverse a multi-level index matrix, wherein the index represented by each matrix element is calculated by the following formula:

index_(i,j)＝i*k+j

Wherein i represents the row index of the matrix element in the multi-level index matrix; j represents the column index of the matrix element in the multi-level index matrix; k represents the values corresponding to the different subordinate types of the filtering condition.

Quick search is carried out in the first layer matrix, and the Value corresponding to the Key in the first-level hash matrix (Key, value) is entered into a second-level hash index table; the second layer matrix stores index information and classification information of the third layer matrix, so that address information of required data is progressively searched layer by layer, and a corresponding value of the data is obtained.

It should be noted that the foregoing is merely illustrative of the present invention and is not to be construed as limiting thereof.

Next, in step S102, aggregation calculation is performed on the respective chess piece card attributes, and a display control matrix of the situation display element is generated.

Specifically, aggregate calculation is performed on the card attributes of each piece so as to sequentially perform classification processing and layering processing on the entity attributes corresponding to each piece in the soldier chess situation data of different levels.

For the aggregation calculation, the following steps are specifically included.

Step S301: and (3) dividing each group of two-dimensional matrixes in the plurality of groups of two-dimensional matrixes obtained in the step (S101) according to a specified division granularity, and dividing the two-dimensional matrixes into a plurality of two-dimensional matrixes to form a new two-dimensional matrix.

The matrix is divided according to a row priority mode, for example, the number of division rows is 10, namely, the matrix in a plurality of groups of two-dimensional matrices is decomposed into a plurality of new matrices with 10 rows and x columns, so as to form a new matrix group (namely, an initial display and hidden control matrix is obtained). Specifically, the specified division granularity includes the number of division lines, the number of division columns, the number of division times, the specific area to be divided, and the like.

It should be noted that, since the initial implicit control matrix is a sparse matrix, most of the elements in the matrix are zero, the matrix can be divided into smaller elements, so that zero elements are ignored during the aggregation operation, thereby reducing the calculation amount and improving the calculation efficiency. For example, the matrix is divided into a plurality of 10 rows x columns by selecting to divide the matrix once every 10 rows of the matrix and dividing the matrix into a plurality of 10 rows x columns by using a row-first dividing method.

Step S302: and performing aggregation calculation on the formed new two-dimensional matrix to obtain the intermediate implicit control matrix.

For example, the first matrix a and the second matrix b are combined to generate a third matrix c, i.e. an intermediate implicit control matrix.

And respectively carrying out singular value decomposition on the first matrix a and the second matrix b to be aggregated to obtain corresponding U, sigma and V matrixes.

a＝U_aΣ_aV_a*

b＝U_bΣ_bV_b*

Singular value decomposition is carried out on a first matrix and a second matrix to be polymerized, and then the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix:

Σ_c＝αΣ_a+βΣ_b

Further comprises: the entropy method is used for assigning the weight coefficient, and the calculation formula of the alpha weight coefficient is as follows:

α＝-lnnp(a)lnp(a)

Further comprises: the entropy method is used for assigning the weight coefficient, and the calculation formula of the beta weight coefficient is as follows:

β＝-lnnp(b)lnp(b)

Wherein α is a weight coefficient corresponding to the singular value decomposed second matrix b; the first matrix a is a matrix of n rows and m columns, n is the total number of rows of the second matrix a, and ln represents a logarithmic operation based on e.

P (b) =Σ ⁿ _i＝1,b_ij,b_ij is the element of the ith row and j column in the second matrix b after singular value decomposition, i and j are positive integers, and n is the total number of rows of the second matrix b.

Preferably, the value of α is not equal to the value of β.

The aggregated singular value matrix Σc and the U and V matrices of the original matrices a and b are used to reconstruct the aggregated intermediate implicit control matrix c.

c＝U_aΣ_cV_a*

C represents the U and V matrices of the original matrices a and b to reconstruct an aggregated intermediate implicit control matrix;

Step S303: and merging the intermediate implicit control matrix generated by merging to obtain a final implicit control matrix.

Specifically, a plurality of intermediate implicit control matrices are combined into one large matrix in a certain order. Initializing an empty result matrix, storing the merged result, copying the elements in the plurality of intermediate implicit control matrices into the result matrix (final implicit control matrix) according to a row-first mode, and repeating the step S303 until all the intermediate implicit control matrices execute the merging operation, thereby obtaining the final implicit control matrix.

When the matrix size of the implicit control matrix is n×n, the time complexity is O (n 2 log), where log represents a 2-based logarithmic operation on n. Thus, the time complexity is largely dependent on the matrix size and granularity of the segmentation. The final implicit control matrix obtained by the aggregation calculation can reduce the time complexity of calculation.

It should be noted that, the multiplication operation of the initial implicit control matrix (specifically, the sparse matrix) includes a large number of discrete indirect addressing operations, so that the system computing resources are occupied in a large amount, which increases the data processing time. The memory space of the implicit control matrix can be compressed to be extremely small to facilitate storage and calculation by compressing the implicit control matrix through a sparse matrix compression algorithm. Since the number of non-zeros in each row of the matrix varies greatly in all rows and such sparse matrices are generated in a row-first manner, the apparent-hidden control matrix is stored in compression by using the CSC format (Compressed Sparse Column) method, and the following three types of information of the sparse matrix are stored in detail: the values, the row indexes and the column offsets of the effective non-zero elements in the display and hidden control matrix are stored in a value matrix A_data, the row numbers are stored in a row number matrix A_indexes, and the column offsets are stored in an A_ indptr matrix. Because of the particularity of the implicit control matrix, all elements of the numerical matrix are 1, and the index of the implicit control matrix can be established only by determining A_indices and A_ indptr.

Sparse matrix A

[1 2 1 0 1 4 4 1 2 0 2 3 0 4 2 4]

Line number matrix A_indices

[0 2 3 6 7 9 12 14 16]

Column offset A_ indptr

The column offset a_ indptr is a one-dimensional array, whose length is equal to the number of columns of the matrix plus one. The value indptr [ j ] represents the index position of the first non-zero element in column j. The line number matrix a_indices is a one-dimensional array. Through the array of column offsets A_ indptr, non-zero elements of each column can be quickly found, so that matrix operations can be accelerated.

For the j-th column in sparse matrix A, its non-zero element position in the A_indices array may be obtained by a column offset A_ indptr [ j ].

By the mode, non-zero elements can be effectively prevented from being searched by traversing the whole sparse matrix, so that the efficiency of accessing the elements in the sparse matrix is improved, namely the efficiency of traversing the implicit control matrix can be effectively improved.

For classification processing and layering processing, the attribute with correlation is aggregated into a group, such as the type of a party and an army is the correlation attribute, the main battle equipment and the equipment strength attribute are the correlation attribute, the ammunition consumption number and the ammunition residue number are the correlation attribute, the attribute correlation is divided according to the model applied by the chess system, and certain differences exist for different chess systems. A tree list or a linear list is used for the attribute groups having the hierarchical relationship. And mapping a part of entity attributes (such as entity attributes corresponding to a part of screened basic attribute labels and the like) from a high-dimensional space to a Boolean matrix in a two-dimensional space to be used as a implicit control matrix for conditional filtering. Each set of "correlation properties" forms a boolean matrix with a set of independent properties. For example, bit operations of the boolean matrix are utilized to implement batch control of entity display states (i.e., situation display elements).

Optionally, the display state (i.e., situation display element) of each entity is controlled according to the filtering rule.

Specifically, the filtering rules include a implicit control matrix. For example in the form of an integrated display control matrix configuration file.

For example, when data display is performed, the comprehensive display control matrix configuration file is obtained, and the comprehensive display control matrix configuration file is loaded for performing display control. The comprehensive display control matrix configuration file comprises a comprehensive display control matrix.

Specifically, the implicit control matrix includes a tree list and a filtering condition list (i.e., the filtering condition list formed in step S101). For example, different tree lists are respectively constructed according to team types, event types, data types, and the like. For example, each matrix element corresponds to a different army type when the type is not established, and can be divided into a plurality of levels. Matrix elements and levels in the tree list can be freely configured according to actual requirements. For example, a first level of the tree list may be set as a army, a navy, an air force, and a second level set as a first group army, a second group army. Adding a fork in the tree list: the equipment army correspondingly increases the next-stage force constitution of the equipment army at the second stage, does not influence the elements in the original tree-shaped list, and adds new elements at the last of the list.

And (3) screening the soldier chess situations with different classifications by inputting elements of the filtering matrix by a user, for example, checking navy army forces in the state of an offshore training field, traversing all relevant chessmen, and comparing whether the army attribute of each chessman is the current battlefield threshold value of the offshore training field.

And particularly, the filtering matrix generated by the filtering condition list is stored, for example, a corresponding configuration file is generated, and the situation of the offshore training field can be obtained rapidly when the situation of the offshore training field is selected again later.

Performing AND operation on the attribute of the tree list and the attribute of the side of the filtering list by establishing a display and hidden control matrix, and displaying situation information of the soldiers and chesses under the combination condition if and only if the attribute of the tree list and the attribute of the filtering list are true; the method can rapidly switch the display and hidden states of the situation information, and remarkably reduces the number of rendering elements, thereby greatly improving the visual rendering efficiency of the situation and reducing the interference of the situation information.

For example, the attribute of the current element of the implicit control matrix in the tree list and the attribute of the filtering list are subjected to AND operation, and when the attribute of the current element of the implicit control matrix in the tree list is true and when the attribute of the current element of the implicit control matrix in the attribute of the filtering list is true, the attribute value of the current element of the implicit control matrix is true.

Next, in step S103, the user behavior tendency is evaluated, and the implicit control matrix is transformed.

According to the behavior characteristics and the requirements of the user, the tendency of the user to the data display is determined to be analyzed, and the requirements of the user to the data display are measured by dividing a plurality of attributes. And determining the weight according to the correlation, importance and other factors of the attribute. The weight is determined based on texts generated by past display control configuration of the user and the user tendency investigation files, the subject words in the texts are extracted for attribute matching, and the weight value is determined according to the occurrence frequency of the words.

In a specific embodiment, for example, a neural network is used to build a user behavior tendency prediction model, a display state (such as available, damaged or destroyed) formed by a party, a battle zone, a user type and an actual state is used to form a user display state label, a positive sample and a negative sample are defined, user input parameters marked with the user display state label are used as a training data set, and the user behavior tendency prediction model is trained to obtain a trained user behavior tendency prediction model.

User input parameters (e.g., using a party, belonging combat zone, user type) are input to the trained user behavior trend prediction model, and user behavior trend prediction values corresponding to the user input parameters are output, e.g., as a multidimensional vector or multidimensional matrix.

And transforming the implicit control matrix according to the predicted value (such as a multidimensional vector or a multidimensional matrix) of the user behavior tendency output by the model to obtain a transformed implicit control matrix.

In another embodiment, the aggregated implicit control matrix is modified using the model-output user behavior trend prediction values (e.g., as a multi-dimensional vector or multi-dimensional matrix).

In particular, the partitioning of user-tendencies attributes of, for example, a chess system, specifically includes command hierarchy, display hierarchy, data preferences, attention status, details. See in particular fig. 4.

The modification of the aggregated saliency control matrix refers to the aggregation of the original saliency control matrix and the characteristic saliency control matrix described below.

And matching the keywords in the entity attributes with the acquired subject words focused by the user, such as the affiliated party, the affiliated combat zone, the army, the current state (available/damaged/destroyed), and setting the attribute focused by the user with the entity as default display.

And establishing a characteristic implicit control matrix by screening and judging matrix elements, wherein the row and column numbers of the characteristic implicit control matrix are the same as those of the original implicit control matrix. And finding the position of the corresponding attribute by using a search algorithm, wherein the Boolean value of the position is 1, and the Boolean value of the rest elements is 0, so as to form a matrix.

The original implicit control matrix and the characteristic implicit control matrix are aggregated, and the operation formula is as follows:

e(i,j)＝C(i,j)+b(i,j)

e(i,j)＝1,(if e(i,j)>0)

e (i, j) represents a matrix after screening, i represents a matrix row number, and j represents a matrix column number;

c (i, j) is the original display and hidden control matrix, i represents the matrix row number, and j represents the matrix column number;

b (i, j) is a feature saliency control matrix, i represents a matrix row number, and j represents a matrix column number.

Through the polymerization process, the polymerized implicit control matrix can be obtained.

Next, in step S104, when receiving the user input, classification filtering processing is performed on the soldier chess situation data and the event situation data according to the transformed implicit control matrix.

In a specific embodiment, when the processing system receives user input, the transformed implicit and explicit control matrix is obtained to filter the soldier chess situation data and the event situation data.

Specifically, the user input specifically includes a military service display category, a displayed army name, a displayed army level, a displayed range of countermeasure areas.

And particularly acquiring a display and hidden control matrix configuration file in real time, and calling the display and hidden control matrix in the acquired display and hidden control matrix configuration file to filter the situation data of the chess and the situation data of the event.

For example, for the chess situation data, the classification and filtering are performed according to the grade attribute of the chess pieces.

1) And selecting corresponding level options on the level navigation tree by hooking to obtain a level filtering set.

2) Traversing all the chesses, judging whether the grade attribute of the chesses is in the grade filtering set, if so, displaying the chesses, and if not, hiding the chesses.

And then, classifying and filtering the event situation data according to the event situation type.

1) On an event situation navigation tree, checking event situation types to obtain an event filtering set

2) And traversing all the current events, judging whether the type attribute of each event is in the event filtering set, if so, displaying the chessman, otherwise, hiding the current event, and not receiving the type event on the bus.

And classifying event situation data through an event situation navigation tree, wherein the event situation data comprises interaction related events and entity related events. Wherein the entity-related events include command-down events, aim-up events, find scout target events, strike events, damage events, satellite transit events, etc.

Compared with the prior art, the invention maps the deduction data of each chess piece from a high-dimensional space to a two-dimensional matrix in a two-dimensional space according to the entity attribute corresponding to each chess piece in the chess situation data to form a filtering condition; performing aggregation calculation on the attributes of the chessmen entities to generate a display-hidden control matrix of the situation display element, and transforming the display-hidden control matrix by evaluating the behavior tendency of the user; when user input is received, the transformed display and hidden control matrix is obtained to filter the soldier chess situation data and event situation data, so that the calculation time complexity can be reduced, the display and hidden states of the situation information can be rapidly switched, the number of rendering elements is obviously reduced, the situation visual rendering efficiency is greatly improved, and meanwhile, the situation information interference is reduced.

Example 2

The following are system embodiments of the present invention that may be used to perform method embodiments of the present invention. For details not disclosed in the system embodiments of the present invention, please refer to the method embodiments of the present invention.

Fig. 5 is a schematic diagram of the structure of an example of a fusion processing system based on polymorphic information data according to the present invention.

Next, a fusion processing system based on polymorphic information data using the fusion processing method based on polymorphic information data according to the first aspect of the present invention will be described with reference to fig. 5.

As shown in fig. 5, the fusion processing system 500 includes a formation processing module 510, a generation processing module 520, an evaluation processing module 530, and a filtering processing module 540.

In a specific embodiment, the formation processing module 510 maps the deduction data of each chess piece from the high-dimensional space to a two-dimensional matrix in the two-dimensional space according to the entity attribute corresponding to each chess piece in the chess situation data, so as to form the filtering condition. The generation processing module 520 performs aggregation calculation on the attributes of the chessmen entities to generate a display control matrix of the situation display element; the evaluation processing module 530 evaluates the behavior tendency of the user and transforms the implicit control matrix; the filtering processing module 540 obtains the transformed implicit control matrix when receiving the user input to filter the soldier chess situation data and the event situation data.

According to an alternative embodiment, the aggregation calculation is performed on the entity attributes of each chess piece to generate a display control matrix of the situation display element, including: dividing each group of two-dimensional matrixes in a plurality of groups of two-dimensional matrixes according to a specified dividing granularity, and dividing the two-dimensional matrixes into a plurality of two-dimensional matrixes to form a new two-dimensional matrix; performing aggregation calculation on the formed new two-dimensional matrix to obtain an intermediate implicit control matrix; and merging the intermediate implicit control matrix generated by merging to obtain a final implicit control matrix.

Further comprises: singular value decomposition is carried out on a first matrix and a second matrix to be polymerized, and then the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix:

Σ_c＝αΣ_a+βΣ_b

Then, the entropy method is used for assigning the weight coefficient, and the calculation formula of the alpha weight coefficient is as follows:

α＝-lnnp(a)lnp(a)

For evaluating user behavioral trends, comprising: a neural network is adopted to establish a user behavior tendency prediction model, a display state formed by a subordinate party, a battle zone, a user type and an actual state is used to form a user display state label, a positive sample and a negative sample are defined, user input parameters marked with the user display state label are used as a training data set, and the user behavior tendency prediction model is trained to obtain a trained user behavior tendency prediction model; the user input parameters are input into a trained user behavior tendency prediction model, and user behavior tendency predicted values corresponding to the user input parameters are output, wherein the user behavior tendency predicted values are multi-dimensional vectors or multi-dimensional matrixes.

According to an alternative embodiment, a chain address method is adopted, and a multi-level index matrix is established according to attribute information or data corresponding to each piece after classification processing and layering processing, wherein the multi-level index matrix comprises association relations of different levels and association relations of different types of the same level

Note that, since the fusion processing method performed by the fusion processing system of fig. 5 is substantially the same as that in the example of fig. 1, the description of the same portions is omitted.

Example 3

Fig. 6 is a schematic structural view of an embodiment of an electronic device according to the present invention.

As shown in fig. 6, the electronic device is in the form of a general purpose computing device. The processor may be one or a plurality of processors and work cooperatively. The invention does not exclude that the distributed processing is performed, i.e. the processor may be distributed among different physical devices. The electronic device of the present invention is not limited to a single entity, but may be a sum of a plurality of entity devices.

The memory stores a computer executable program, typically machine readable code. The computer readable program may be executable by the processor to enable an electronic device to perform the method, or at least some of the steps of the method, of the present invention.

The memory includes volatile memory, such as Random Access Memory (RAM) and/or cache memory, and may be non-volatile memory, such as Read Only Memory (ROM).

Optionally, in this embodiment, the electronic device further includes an I/O interface, which is used for exchanging data between the electronic device and an external device. The I/O interface may be a bus representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

It should be understood that the electronic device shown in fig. 6 is only one example of the present invention, and the electronic device of the present invention may further include elements or components not shown in the above examples. For example, some electronic devices further include a display unit such as a display screen, and some electronic devices further include a man-machine interaction element such as a button, a keyboard, and the like. The electronic device may be considered as covered by the invention as long as the electronic device is capable of executing a computer readable program in a memory for carrying out the method or at least part of the steps of the method.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, as shown in fig. 7, the technical solution according to the embodiment of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, and includes several commands to cause a computing device (may be a personal computer, a server, or a network device, etc.) to perform the above-described method according to the embodiment of the present invention.

The software product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. The readable storage medium can also be any readable medium that can communicate, propagate, or transport the program for use by or in connection with the command execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The computer-readable medium carries one or more programs, which when executed by one of the devices, cause the computer-readable medium to implement the data interaction methods of the present disclosure.

Those skilled in the art will appreciate that the modules may be distributed throughout several devices as described in the embodiments, and that corresponding variations may be implemented in one or more devices that are unique to the embodiments. The modules of the above embodiments may be combined into one module, or may be further split into a plurality of sub-modules.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, and which includes several commands to cause a computing device (may be a personal computer, a server, a mobile terminal, or a network device, etc.) to perform the method according to the embodiments of the present invention.

It should be noted that the foregoing detailed description is exemplary and is intended to provide further explanation of the application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

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

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A fusion processing method based on polymorphic information data is characterized by comprising the following steps:

according to entity attributes corresponding to each chess piece in the chess situation data, the deduction data of each chess piece is mapped from a high-dimensional space to a two-dimensional matrix in a two-dimensional space so as to form a filtering condition;

Aggregation calculation is carried out on the attributes of the chessmen entities to generate a display control matrix of the situation display elements, which comprises the following steps: dividing each group of two-dimensional matrixes in a plurality of groups of two-dimensional matrixes according to a specified dividing granularity, and dividing the two-dimensional matrixes into a plurality of two-dimensional matrixes to form a new two-dimensional matrix; performing aggregation calculation on the formed new two-dimensional matrix to obtain an intermediate implicit control matrix; merging the intermediate implicit and explicit control matrix obtained by the aggregation calculation to obtain a final implicit and explicit control matrix; further comprises: singular value decomposition is carried out on a first matrix and a second matrix to be polymerized, and then the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix:

Σ_c＝αΣ_a+βΣ_b

Wherein, sigma_c represents that the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix; Σ_a represents a first matrix obtained by performing singular value decomposition on a first matrix a to be aggregated; Σ_b represents a second matrix obtained by performing singular value decomposition on a second matrix b to be aggregated; alpha and beta are weight coefficients corresponding to the first matrix and the second matrix after singular value decomposition, respectively; the entropy method is used for assigning the weight coefficient, and the calculation formula of the alpha weight coefficient is as follows:

α＝-lnnp(a)lnp(a)

wherein alpha is a weight coefficient corresponding to the first matrix a after singular value decomposition; the first matrix a is a matrix of n rows and m columns, n is the total number of rows of the first matrix a, and ln represents a logarithmic operation based on e;

p (a) =Σ ⁿ _i＝1a_ij,a_ij is the element of the ith row and j column in the first matrix a after singular value decomposition, i and j are positive integers, and n is the total number of rows of the first matrix a;

Evaluating the user behavior tendency, specifically inputting the user input parameters into a trained user behavior tendency prediction model, outputting a user behavior tendency predicted value corresponding to the user input parameters, and correcting the aggregated and calculated implicit and explicit control matrix by using the user behavior tendency predicted value output by the model; transforming the implicit control matrix according to the predicted value of the user behavior tendency output by the model;

When receiving user input, the transformed implicit and explicit control matrix is obtained to filter the soldier chess situation data and event situation data.

2. The fusion processing method based on polymorphic information data according to claim 1, wherein the evaluation of the user behavior tendency comprises:

A neural network is adopted to establish a user behavior tendency prediction model, a display state formed by a subordinate party, a battle zone, a user type and an actual state is used to form a user display state label, a positive sample and a negative sample are defined, user input parameters marked with the user display state label are used as a training data set, and the user behavior tendency prediction model is trained to obtain a trained user behavior tendency prediction model;

the user behavior tendency predicted value is a multidimensional vector or a multidimensional matrix.

3. The fusion processing method based on polymorphic information data according to claim 1, comprising:

screening the feature matrix, and screening the implicit control matrix according to the feature matrix to obtain situation display matched with the behavior tendency of the user.

4. The fusion processing method based on polymorphic information data according to claim 1, wherein,

And establishing a multi-level index matrix according to the attribute information or data corresponding to each piece after classification and layering treatment by adopting a chain address method, wherein the multi-level index matrix comprises association relations of different levels and association relations of different types of the same level.

5.A fusion processing system based on polymorphic information data, characterized in that it uses the fusion processing method based on polymorphic information data according to any one of claims 1 to 4, comprising:

The forming processing module is used for mapping deduction data of each chess piece from a high-dimensional space to a two-dimensional matrix in a two-dimensional space according to entity attributes corresponding to each chess piece in the chess situation data so as to form filtering conditions;

The generation processing module is used for carrying out aggregation calculation on the entity attributes of the chessmen to generate a display and hidden control matrix of the situation display element, and specifically comprises the following steps: dividing each group of two-dimensional matrixes in a plurality of groups of two-dimensional matrixes according to a specified dividing granularity, and dividing the two-dimensional matrixes into a plurality of two-dimensional matrixes to form a new two-dimensional matrix; performing aggregation calculation on the formed new two-dimensional matrix to obtain an intermediate implicit control matrix; merging the intermediate implicit and explicit control matrix obtained by the aggregation calculation to obtain a final implicit and explicit control matrix; further comprises: singular value decomposition is carried out on a first matrix and a second matrix to be polymerized, and then the first matrix and the second matrix after singular value decomposition are subjected to weighted aggregation to obtain an intermediate implicit control matrix:

Σ_c＝αΣ_a+βΣ_b

α＝-lnnp(a)lnp(a)

The evaluation processing module is used for evaluating the user behavior tendency, specifically inputting the user input parameters into a trained user behavior tendency prediction model, outputting a user behavior tendency predicted value corresponding to the user input parameters, and correcting the aggregated and calculated implicit control matrix by using the user behavior tendency predicted value output by the model; transforming the implicit control matrix according to the predicted value of the user behavior tendency output by the model;

And the filtering processing module is used for acquiring the transformed implicit and explicit control matrix when receiving user input so as to filter the soldier chess situation data and the event situation data.