CN114238648B - Game countermeasure behavior decision method and device based on knowledge graph - Google Patents
Game countermeasure behavior decision method and device based on knowledge graph Download PDFInfo
- Publication number
- CN114238648B CN114238648B CN202111364693.9A CN202111364693A CN114238648B CN 114238648 B CN114238648 B CN 114238648B CN 202111364693 A CN202111364693 A CN 202111364693A CN 114238648 B CN114238648 B CN 114238648B
- Authority
- CN
- China
- Prior art keywords
- node
- game
- countermeasure
- conclusion
- knowledge graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/80—Special adaptations for executing a specific game genre or game mode
- A63F13/803—Driving vehicles or craft, e.g. cars, airplanes, ships, robots or tanks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/042—Backward inferencing
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/80—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game specially adapted for executing a specific type of game
- A63F2300/8017—Driving on land or water; Flying
Abstract
The invention provides a game countermeasure behavior decision method and a game countermeasure behavior decision device based on a knowledge graph, wherein the method comprises the following steps: acquiring fact data in a game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process; matching the fact data with an antagonism behavior decision model to obtain a behavior decision result of the game antagonism; wherein the countermeasure behavior decision model is constructed based on a knowledge graph, and is used for representing the generative rules of the game countermeasure. The rule matching process is converted into the inquiry searching process of the knowledge graph, the time complexity is only related to the complexity of the inquiry statement, and is decoupled from the scale of the rule set, and compared with the traditional mode of scanning and matching one by one based on the text rule set, the rule matching efficiency under the large-scale rule base scene is greatly improved, and further the effective improvement of the behavior decision efficiency in the game countermeasure process is realized.
Description
Technical Field
The invention relates to the technical field of behavior decision, in particular to a game countermeasure behavior decision method and device based on a knowledge graph.
Background
Generating a regular form such as P 1 (x)∧P 2 (x)∧...∧P m (x)→Q 1 (x)∧Q 2 (x)∧...∧Q n (x) It is a common knowledge representation method, which expresses causal relationships in the form of "IF-THEN". Wherein, P 1 (x)∧P 2 (x)∧...∧P m (x) Is an IF part, also called a rule front-piece, P i (x) The predicate is used for representing that the variable x is asserted, and the assertion result can be true, representing that the assertion is established, or can be false, representing that the assertion is not established; Λ is a "conjunction" symbol, indicating that the assertions of two predicates before and after the symbol are simultaneously satisfied. Q 1 (x)∧Q 2 (x)∧...∧Q n (x) The THEN part is also called a rule back-piece, and the rule back-piece is formed by combining a plurality of conclusion primitives through logical relations, and the conclusion primitives include a function, such as printing a statement, or executing an action by an entity x. An "→" implication "symbol indicating that the back-piece is in effect if the front-piece holds. The expression form of the production rule reflects the behavior characteristics of human beings for solving a class of problems, and the solution method of the problems can be deduced by circularly applying the rules, which is one of the most widely applied expression forms of the rules.
The game confrontation is the adversary strategy interaction which is carried out by the opponent and the my party by using respective strength units in order to achieve respective targets. The behavior decision in the scene mainly refers to a process of judging the current situation and taking action in a targeted manner. The course of action decision is well suited to be expressed using a generative rule form, where "judgment of current situation" can be expressed as a rule antecedent and "action taken" can be expressed as a rule successor, and when an instance can match the antecedent of a rule, a conclusion at the back of the rule is reached. Therefore, the rule set formed by the production rules is equivalent to the brain for controlling the behavior under the game confrontation, the more fully considered the possible situation of the state, the smarter the brain is, the better the final effect of the game confrontation is, and the larger the scale of the production rule set is.
For application of large-scale generation rules in a game countermeasure scene, in the prior art, persistent storage media such as hard disks are usually adopted to store the large-scale generation rules, but rule matching is usually carried out on the basis of scanning and matching one by one of rule sets, time complexity is positively correlated with the size of the rule set, matching efficiency is greatly reduced, and timeliness of behavior decision in the game countermeasure process cannot be guaranteed.
Disclosure of Invention
The invention provides a game countermeasure behavior decision method and a game countermeasure behavior decision device based on a knowledge graph, which are used for solving the defect of poor timeliness of behavior decision in the prior art and effectively improving the behavior decision efficiency in the game countermeasure process.
The invention provides a game countermeasure behavior decision method based on a knowledge graph, which comprises the following steps:
acquiring fact data in a game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
matching the fact data with an antagonism behavior decision model to obtain a behavior decision result of the game antagonism;
wherein the countermeasure behavior decision model is constructed based on a knowledge graph, and is used for representing the generative rules of the game countermeasure.
According to the game countermeasure behavior decision method based on the knowledge graph, the game countermeasure behavior decision model based on the knowledge graph is constructed, and the game countermeasure behavior decision method based on the knowledge graph comprises the following steps:
acquiring a generative rule of the game countermeasure;
extracting predicates, conclusion primitives and logical relations in the production rule; the predicates are used for representing decision conditions of the game countermeasure, the conclusion primitives are used for representing decision results of the game countermeasure, and the logic relationship is used for representing the relationship between the two predicates and/or the predicate and the conclusion primitives;
constructing the confrontation decision model by adopting a knowledge graph based on the predicates, the conclusion primitives and the logical relationship; and creating nodes of the knowledge graph based on the predicates and the conclusion primitives, and creating edges of the knowledge graph based on the logical relations, wherein the edges of the knowledge graph are directed edges.
According to the game countermeasure behavior decision method based on the knowledge graph, the extraction of predicates, conclusion primitives and logical relations in the production formula rule comprises the following steps:
based on the game countermeasure generating formula rule, acquiring the predicate, the conclusion primitive and the attribute value of the logical relationship, and completing extraction of the predicate, the conclusion primitive and the logical relationship; wherein the attribute values are used to complete matching of the fact data to a confrontation decision model.
According to the game countermeasure behavior decision method based on the knowledge graph, the step of constructing the countermeasure behavior decision model based on the predicates, the conclusion primitives and the logic relationship by adopting the knowledge graph comprises the following steps:
and (3) node creation: receiving a predicate or a conclusion primitive in the production rule, and detecting whether a node corresponding to the predicate or the conclusion primitive exists in the knowledge graph; if not, creating a current node in the knowledge graph based on the predicate or the conclusion primitive; if yes, taking the existing node as the current node, and executing a node detection step;
and (3) node detection: detecting whether the current node is the first node of the production rule, if not, executing an edge creating step, and if so, executing the node creating step;
creating an edge: and based on the logical relationship corresponding to the current node, creating a directed edge from the previous node of the current node to the current node, and repeatedly executing the steps of node creation, node detection and edge creation until all predicates, conclusion primitives and logical relationships in the production formula rule complete the creation of the node or the edge.
According to the game countermeasure behavior decision method based on the knowledge graph, the fact data are matched with a countermeasure behavior decision model to obtain a behavior decision result of game countermeasure, and the game countermeasure decision method based on the knowledge graph comprises the following steps:
node matching: performing node matching on the fact data according to the attribute values of the nodes of the countermeasure decision model to obtain an initial node;
assertion calculation: performing assertion calculation according to the attribute value of the starting node;
edge matching: acquiring a termination node corresponding to the starting node based on an assertion calculation result of the starting node and an attribute value of an edge of the antagonistic behavior decision model;
and (3) node identification: and identifying whether the termination node is a conclusion primitive, if not, taking the termination node as an initial node, and repeatedly executing the steps of assertion calculation, edge matching and node identification, and if so, obtaining a behavior decision result of the game countermeasure according to the attribute value of the termination node.
According to the game countermeasure behavior decision method based on the knowledge graph, after the game countermeasure behavior decision model is built based on the knowledge graph, the method further comprises the following steps:
storing the countermeasure decision model by adopting a data model based on a label attribute graph; the data model respectively and independently stores the nodes of the countermeasure decision model and the labels of the nodes, and indexes are arranged in the labels and used for indicating the storage positions of the nodes.
The invention also provides a game countermeasure behavior decision device based on the knowledge graph, which comprises: the data acquisition module is used for acquiring fact data in the game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
the decision matching module is used for matching the fact data with an confrontation behavior decision model to obtain a behavior decision result of the game confrontation;
wherein the countermeasure behavior decision model is constructed based on a knowledge graph, and is used for representing the generative rules of the game countermeasure.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the game countermeasure action decision method based on the knowledge graph.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method for knowledge-graph based gambling confrontation decision making as described in any of the above.
The present invention also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method for knowledge-graph-based game play countermeasure decision-making as described in any of the above.
According to the game countermeasure behavior decision method and device based on the knowledge graph, the countermeasure behavior decision model is built through the knowledge graph, the game countermeasure generating rule is expressed through the countermeasure behavior decision model, the game countermeasure behavior decision result is obtained through matching of fact data obtained in real time and the countermeasure behavior decision model, the rule matching process is converted into the knowledge graph query search process, time complexity is only related to complexity of query sentences, and is decoupled with scale of a rule set, compared with a traditional mode of scanning and matching one by one based on a text rule set, rule matching efficiency under a large-scale rule base scene is greatly improved, and effective improvement of behavior decision efficiency in the game countermeasure process is achieved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a game countermeasure behavior decision method based on knowledge graph according to the present invention;
FIG. 2 is a schematic diagram of a knowledge graph corresponding to rule one in a military simulation game scenario provided by the present invention;
FIG. 3 is a schematic diagram of a knowledge graph corresponding to rule one and rule two in a military simulation game scenario provided by the present invention;
FIG. 4 is a schematic diagram of a knowledge graph corresponding to rules one to four in a military simulation game scenario provided by the present invention;
FIG. 5 is a schematic flow chart of rule matching in a military simulation game scenario provided by the present invention;
FIG. 6 is a schematic structural diagram of a data model based on a tag attribute map in a military simulation game scenario provided by the present invention;
FIG. 7 is a schematic diagram of the relationship between the upper and lower positions of the entity types in the scene of the military simulation game provided by the present invention;
FIG. 8 is a schematic structural diagram of a game play countermeasure decision making device based on knowledge graph provided by the invention;
fig. 9 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
The knowledge-graph-based game countermeasure behavior decision method of the present invention is described below in conjunction with fig. 1-7; the game countermeasure behavior decision method based on the knowledge graph is shown in figure 1 and comprises the following steps:
s100, acquiring fact data in a game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
s200, matching the fact data with an confrontation behavior decision model to obtain a game confrontation behavior decision result;
the game countermeasure method comprises the steps of establishing a countermeasure behavior decision model based on a knowledge graph, wherein the countermeasure behavior decision model is used for expressing a production rule of game countermeasure.
The method is characterized in that an antagonism behavior decision model is constructed based on the knowledge map, game antagonism generation rules are expressed through the antagonism behavior decision model, and game antagonism behavior decision results are obtained through matching of fact data acquired in real time and the antagonism behavior decision model, so that a rule matching process is converted into a knowledge map query search process, time complexity is only related to complexity of query sentences, and is decoupled with scale of a rule set. Meanwhile, the countermeasure behavior decision model is constructed according to the knowledge graph to express the production rules, so that the advantages of the knowledge graph underlying graph database in the aspects of storage capacity, query efficiency, expansibility, operability and the like are conveniently utilized, and the management and application of the production rules in the game countermeasure process are better supported.
In step S200, constructing an antagonistic behavior decision model based on the knowledge graph includes:
s211, acquiring a generation formula rule of game countermeasures; in the step, obtaining the generating rule of the game countermeasure comprises the following steps: acquiring a rule set, and reading production rules one by one from the rule set; the rule set may be obtained according to actual requirements, for example, by receiving a preset instruction.
S212, extracting predicates, conclusion primitives and logical relations in the production formula rule; the predicates are used for representing decision conditions of game countermeasures, namely the predicates are used for carrying out condition judgment; the conclusion primitive is used for representing a decision result of the game countermeasure, namely an instruction action executed after the rule matching is successful; logical relationships are used to represent the relationship between two predicates, and/or between a predicate and a conclusion primitive.
S213, constructing an antagonistic behavior decision model by using a knowledge graph based on predicates, conclusion primitives and logical relations; and creating nodes of the knowledge graph based on the predicates and the conclusion primitives, and creating edges of the knowledge graph based on the logical relationship, wherein the edges of the knowledge graph are directed edges.
And repeating the steps S211 to S213, and incrementally constructing the countermeasure decision model until all the production rules in the rule set are expressed in the countermeasure decision model, namely constructing the countermeasure decision model through all the production rules in the rule set.
In the prior art, knowledge graphs are generally used for decision-making systems and recommendation systems in multiple fields such as e-commerce, finance, law, medical treatment, intelligent home and the like, and due to the specific representation mode and use method of the generative rules, no mature technical means is available for representing the generative rules through the knowledge graphs. The predicate, the conclusion primitive and the logic relation in the production formula rule are extracted to construct the knowledge graph, the predicate and the conclusion primitive are used as nodes of the knowledge graph, and the logic relation is used as an edge of the knowledge graph, so that the production formula rule is effectively expressed through the knowledge graph, and a technical basis is provided for improving behavior decision efficiency in the game countermeasure process.
In step S212, extracting predicates, conclusion primitives, and logical relationships in the production formula rule includes: based on a game countermeasure production formula rule, acquiring attribute values of predicates, conclusion primitives and logic relations, and completing extraction of the predicates, the conclusion primitives and the logic relations; wherein, the attribute value is used for completing the matching of the fact data and the confrontation decision model; the attribute values of predicates, conclusion primitives, and logical relationships each include several different fields, e.g., the attribute values of predicates and conclusion primitives each include: name field, label field, parameter field, assertion/result field; the name field is used for marking the name of the predicate or conclusion primitive for easy understanding; the label field is used for marking an entity type list; the parameter field is used for marking a parameter list, the parameter list is in one-to-one correspondence with the entity type list of the label field, namely each entity type corresponds to a group of parameters, and the parameters in the parameter list are specific entities corresponding to the entity types of the label field; the assertion/result field is used for marking assertion expressions/rule conclusions in the predicate/conclusion primitives; for the predicate node, parameters in the parameter list are used for substituting the predicate expression to perform predicate calculation; for the conclusion primitive node, the parameters in the parameter list are used to indicate the specific entity in the rule conclusion for performing the instruction action, which corresponds to the entity type in the tag field. The logical relationship is represented as a directed edge in the knowledge-graph from a starting point to an ending point, and the attribute values of the logical relationship include: a true value field, a type field; the truth field is used to indicate the truth value that the starting predicate should get, and is the bar from the starting predicateAn element, e.g. a predicate of
Aircraft (x), wherein, symbol
If the predicate is negative, the true value of the true field on the edge with the starting point is False, which indicates that when x is not Aircraft, the current node can be left to continue checking the next node; the type field is used for representing the type of the entity required in the end point, and is a condition for entering the end point, for example, if the predicate is 'judge whether the distance between a certain airplane and a fighter is less than a threshold value theta', the type field value on the edge with the predicate as the end point is [ airplane, fighter]It is stated that the node can only be entered for matching if the two entities are of the type airplane and fighter plane, respectively. The attribute representations of the nodes and edges of the knowledge-graph are shown in table 1.
TABLE 1
According to the invention, through extracting the attribute values of predicates, conclusion primitives and logical relations and carrying out segmented representation on the attribute values, the searching and matching of related nodes are facilitated in the rule matching process, so that the accuracy and efficiency of rule matching are effectively improved, and the accuracy and efficiency of behavior decision in the game countermeasure process are further ensured.
In step S213, constructing an antagonistic behavior decision model using a knowledge graph based on the predicates, the conclusion primitives, and the logical relationships includes:
and (3) node creation: receiving a predicate or conclusion primitive in the production rule, and detecting whether a node corresponding to the predicate or conclusion primitive exists in the knowledge graph; if not, based on the predicate or conclusion primitive, creating a current node in the knowledge graph spectrum; if yes, taking the existing node as the current node, and executing a node detection step;
node detection: detecting whether the current node is the first node of the production rule, if not, executing an edge creating step, and if so, executing a node creating step;
creating an edge: and based on the logical relationship corresponding to the current node, creating a directed edge from the previous node of the current node to the current node, and repeatedly executing the steps of node creation, node detection and edge creation until all predicates, conclusion primitives and logical relationships in the production formula rule complete the creation of the node or the edge.
In the node creating process, whether a certain predicate or conclusion primitive in the scanning rule r appears in the previous rule r' or not is scanned, if the certain predicate or conclusion primitive appears and is constructed as the node, the rule r is constructed directly based on the node, so that different rules can share the predicate and/or conclusion primitive, data repetition in the countermeasure behavior decision model is effectively avoided, and storage overhead of the countermeasure behavior decision model is reduced.
In step S200, matching the fact data with the countermeasure behavior decision model to obtain a game countermeasure behavior decision result, including:
node matching: performing node matching on the fact data according to the attribute values of the nodes of the countermeasure behavior decision model to obtain an initial node, wherein the initial node is a node which is successfully matched;
assertion calculation: performing assertion calculation according to the attribute value of the starting node;
edge matching: acquiring a termination node corresponding to the starting node based on an assertion calculation result of the starting node and an attribute value of an edge of the countermeasure behavior decision model;
and (3) node identification: and identifying whether the termination node is a conclusion primitive, if not, taking the termination node as an initial node, and repeatedly executing the steps of assertion calculation, edge matching and node identification, and if so, obtaining a behavior decision result of the game countermeasure according to the attribute value of the termination node.
In the process of matching the fact data with the countermeasure behavior decision model, if matching fails, the behavior decision fails, so that the game countermeasure behavior decision is more accurate and effective, the required production rule is more perfect, and the scale of the production rule set is larger, therefore, the accuracy and the effectiveness of the game countermeasure decision can be effectively improved by reducing the storage overhead and improving the rule matching efficiency.
As an alternative, matching the fact data with the countermeasure behavior decision model to obtain a specific execution process of the game countermeasure behavior decision result is as follows:
node matching: for each entity in the fact data, matching nodes with the same entity type as the entity type in the fact data from the nodes without edges of the countermeasure decision model according to the entity type of the label field in the countermeasure decision model node, and taking the nodes as starting nodes;
assertion calculation: substituting the parameters corresponding to the entity type in the initial node parameter field into an assertion field, performing assertion calculation through an assertion expression in the assertion field, marking the assertion calculation value as d, and in the assertion calculation process, comparing and judging the entity in the fact data and the parameters in the parameter field, if the entity in the fact data is the same as the parameters in the parameter field, judging the entity in the fact data to be true, otherwise, judging the entity in the fact data to be false;
edge matching: traversing the outgoing edge of the initial node according to the initial node successfully matched, if the value of the true value field of the outgoing edge is the same as the assertion calculation value d, continuously determining whether the entity type in the type field of the outgoing edge is the same as the entity type in the fact data, and if so, taking the end point of the outgoing edge as a termination node;
and (3) node identification: if the termination node is a conclusion primitive, the conclusion part of the rule is reached, and a game countermeasure behavior decision result is obtained according to the rule conclusion of the result field in the termination node; if the termination node is not a conclusion primitive, the conclusion part of the rule is not reached, the termination node is used as an initial node, and the steps of assertion calculation, edge matching and node identification are repeatedly executed until a behavior decision result of the game countermeasure is obtained.
In the step of node matching and edge matching, when the entity type in the fact data is a child type, the node whose entity type is the child type or a parent type can be matched, and when the entity type in the fact data is a parent type, the child type node of the parent type cannot be matched. For example, for an entity a of a type of a fighter in the fact data, the node of the type of the fighter can be matched, and the node of the type of the fighter can also be matched, but an entity b of the type of the airplane in the fact data cannot be matched with the node of the type of the fighter; wherein, the aircraft is the father class of fighter.
The method matches the production rules according to the fact data in the game countermeasure process to support game countermeasure behavior decision, and the countermeasure behavior decision model is constructed according to the knowledge map, so that in the behavior decision process, the matching process of the original fact set and the rule set is converted into the query search process of the fact set on the rule knowledge map, the time complexity is only related to the complexity of query sentences, and is decoupled from the scale of the rule set. Meanwhile, the rule matching is carried out through the attribute values of the nodes and the edges, the matching process is very convenient and fast, the matching of the nodes and the edges without joints can be effectively avoided, the nodes and the edges related to the fact data are directly matched, the matching efficiency is greatly improved, and the timeliness of behavior decision in the game countermeasure process is effectively guaranteed.
In step S200, after constructing the countermeasure behavior decision model based on the knowledge graph, the method further includes: storing the countermeasure behavior decision model by adopting a data model based on a tag attribute diagram; the data model respectively and independently stores the node of the countermeasure behavior decision model and the label of the node, wherein the label is provided with an index which is used for indicating the storage position of the node. In the step, the labels of the nodes are label fields in the nodes, namely, an index is created in the label fields, and the labels and the nodes are stored separately, so that the corresponding nodes can be found quickly according to the index after the entity types are matched, and the labels with unmatched entity types do not need to read the contents of the corresponding nodes, thereby avoiding matching with all predicate nodes, greatly improving the rule matching efficiency and ensuring the timeliness of behavior decision in the game countermeasure process.
The game countermeasure decision method based on the knowledge graph is explained in detail below by taking the game countermeasure decision under the scene of military simulation game as an example.
The military simulation game scenario is described as follows: i pie 1 bomber b 0 Bombing enemy certain ground building c 1 Considering possible encounter with enemy plane f midway 1 1 our fighter f 0 And (5) protecting the navigation. The rule set includes four production rules for controlling bomber b 0 And fighter f 0 Four production rules are as follows:
rule one is as follows: if enemy ground building c 1 Entering a bombing range, and putting bombs for bombing;
rule two: when encountering enemy plane f 1 Get into I fighter f 0 When the shooting range is over, the bullet is struck and driven away;
rule three: if my party bomber b 0 Returning to the base for replenishment if the residual oil amount is less than 1/3;
rule four: if my party fighter f 0 And if the residual oil quantity is less than 1/3 or the residual missile quantity is less than 1/5, returning to the base for replenishment.
The scene contains four physical objects in total, namely bombers b 0 Fighter f 0 "Yiminji" for playing enemy 1 Floor building c 1 Four entity objects constitute a fact set; the fact set is changing in real time as the position, speed, etc. of the object changes as the game time advances; and matching the changed fact set with the rule set according to a certain frequency to trigger the continuous updating of the behaviors in the game countermeasure process.
The game countermeasure behavior decision method based on the knowledge graph comprises the following steps:
s310, constructing an antagonistic behavior decision model based on the knowledge graph;
the basic constituent elements of a production rule include predicates in the front-part, conclusion primitives in the back-part, and logical join relationships. Wherein, the predicates and conclusion primitives are expressed as nodes in the knowledge graph, and the logical relations are expressed as edges in the knowledge graph. The attributes of predicates, conclusion primitives, and logical relationships are shown in Table 1. The conclusion primitive is the instruction action executed after the rule matching is successful in this embodiment.
Establishing a countermeasure decision model for each production rule one by one in an increment mode according to the sequence of the production rules in the rule set; the method comprises the following steps:
firstly, constructing a knowledge graph corresponding to a rule, as shown in FIG. 2; secondly, building a knowledge graph corresponding to a rule II in an incremental manner based on the knowledge graph corresponding to the rule I, as shown in FIG. 3; and thirdly, building the knowledge maps corresponding to the rule three and the rule four in an incremental manner in sequence based on the knowledge maps corresponding to the rule one and the rule two, as shown in fig. 4.
In the process of constructing the increment of the knowledge graph, nodes shared among different rules do not need to be reconstructed, for example, if a predicate node of 'whether our party' exists in a rule I in a rule II, the node does not need to be reconstructed, and only edges need to be added according to the rule II, so that unnecessary node redundancy is reduced, and storage cost is reduced.
S320, managing and storing the countermeasure behavior decision model by using a bottom layer database of Neo4 j; neo4j adopts a data model based on a label attribute graph, as shown in fig. 6, labels and nodes are respectively and independently stored, indexes are created in the labels, and the indexes point to the positions of corresponding nodes, so that predicate nodes can be quickly and accurately positioned according to the types of entities in fact data, matching with all the predicate nodes is avoided, and matching efficiency of production formula rules is improved.
S330, performing behavior decision according to the confrontation behavior decision model; in the step, matching the fact data with an antagonistic behavior decision model at a certain frequency; for each entity in the fact data, matching is started from a predicate node without entering edges, entity type matching is firstly carried out according to a label field, after the type matching is successful, parameters corresponding to the entity type are obtained from a parameter field according to the entity type and are substituted into an assertion field for assertion calculation, and an assertion calculation value is marked as d; then, according to the successfully matched node, the exit of the node is traversedIf the value of the true value field of the edge is the same as the assertion calculated value d, continuously checking whether the entity type in the type field of the edge is consistent with the entity type of the current entity in the fact data, if so, entering the end point of the edge, if the end point of the edge is a conclusion primitive, explaining that a rule conclusion is reached, triggering a corresponding action, and finishing an action decision; and if the edge is not matched with the end point of the outgoing edge, continuing to perform edge matching and end point matching according to the end point of the outgoing edge until the action decision is completed. For example, when the entity status in the fact data satisfies the enemy ground building c 1 And my bomber b 0 Is less than b 0 The matching process is shown in the dashed area in fig. 5. In the matching process of the entity types, whether the entity types in the fact data conform to the entity types specified in the predicates or not is mainly checked, the principle that the subclasses can match the parent types and the parent types cannot match the subclasses is followed, and the upper-lower relationship of the entity types is shown in fig. 7.
The game countermeasure action decision device based on the knowledge graph provided by the invention is described below, and the game countermeasure action decision device based on the knowledge graph described below and the game countermeasure action decision method based on the knowledge graph described above can be correspondingly referred to each other. The game countermeasure action decision device based on the knowledge graph is shown in fig. 8 and comprises:
the data acquisition module 710 is used for acquiring fact data in the game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
the decision matching module 720 is used for matching the fact data with the countermeasure behavior decision model to obtain a game countermeasure behavior decision result;
the game countermeasure method comprises the steps of establishing a countermeasure behavior decision model based on a knowledge graph, wherein the countermeasure behavior decision model is used for expressing a production rule of game countermeasure.
As an alternative, the confrontation decision model is constructed based on the knowledge graph, and comprises the following steps: acquiring a production rule of game countermeasures; extracting predicates, conclusion primitives and logical relations in the production formula rule; the predicates are used for representing decision conditions of the game countermeasure, the conclusion primitives are used for representing decision results of the game countermeasure, and the logic relationship is used for representing the relationship between the two predicates and/or the relationship between the predicates and the conclusion primitives; constructing an antagonistic behavior decision model by using a knowledge graph based on predicates, conclusion primitives and logical relations; and creating nodes of the knowledge graph based on the predicates and the conclusion primitives, and creating edges of the knowledge graph based on the logical relationship, wherein the edges of the knowledge graph are directed edges.
As an alternative, the extraction of predicates, conclusion primitives, and logical relationships in the production formula rule includes: based on a game countermeasure production formula rule, acquiring attribute values of predicates, conclusion primitives and logic relations, and completing extraction of the predicates, the conclusion primitives and the logic relations; wherein the attribute values are used for completing the matching of the fact data and the confrontation decision model.
As an alternative, the construction of the confrontation decision model by using the knowledge graph based on the predicates, the conclusion primitives and the logical relations comprises the following steps: and (3) node creation: receiving a predicate or conclusion primitive in the production rule, and detecting whether a node corresponding to the predicate or conclusion primitive exists in the knowledge graph; if not, creating a current node in the knowledge graph spectrum based on the predicate or the conclusion primitive; if yes, taking the existing node as the current node, and executing a node detection step; and (3) node detection: detecting whether the current node is the first node of the production rule, if not, executing an edge creating step, and if so, executing a node creating step; creating an edge: and based on the logical relationship corresponding to the current node, creating a directed edge from the previous node of the current node to the current node, and repeatedly executing the steps of node creation, node detection and edge creation until all predicates, conclusion primitives and logical relationships in the production formula rule complete the creation of the node or the edge.
As an alternative, matching the fact data with the countermeasure behavior decision model to obtain a behavior decision result of game countermeasure, including: node matching: performing node matching on the fact data according to the attribute values of the nodes of the countermeasure behavior decision model to obtain an initial node; assertion calculation: performing assertion calculation according to the attribute value of the initial node; edge matching: acquiring a termination node corresponding to the starting node based on an assertion calculation result of the starting node and an attribute value of an edge of the countermeasure behavior decision model; and (3) node identification: and identifying whether the termination node is a conclusion primitive, if not, taking the termination node as an initial node, and repeatedly executing the steps of assertion calculation, edge matching and node identification, and if so, obtaining a behavior decision result of the game countermeasure according to the attribute value of the termination node.
As an alternative, the system further comprises a storage module, wherein the storage module is used for storing the countermeasure behavior decision model; the storage module stores the countermeasure behavior decision model by adopting a data model based on a tag attribute map; the data model respectively and independently stores the node of the countermeasure behavior decision model and the label of the node, wherein the label is provided with an index which is used for indicating the storage position of the node.
Fig. 9 illustrates a physical structure diagram of an electronic device, and as shown in fig. 9, the electronic device may include: a processor (processor) 810, a communication Interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication Interface 820 and the memory 830 communicate with each other via the communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform a knowledge-graph based gambling confrontation decision method comprising:
acquiring fact data in a game countermeasure process; wherein, the fact data is used for representing the confrontation state in the game confrontation process;
matching the fact data with the countermeasure behavior decision model to obtain a behavior decision result of game countermeasure;
the game countermeasure method comprises the steps of establishing a countermeasure behavior decision model based on a knowledge graph, wherein the countermeasure behavior decision model is used for expressing a production rule of game countermeasure.
In addition, the logic instructions in the memory 830 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the method for determining a gambling confrontation based on a knowledge graph provided by the above methods, the method comprising:
acquiring fact data in a game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
matching the fact data with the countermeasure behavior decision model to obtain a behavior decision result of game countermeasure;
the game countermeasure method comprises the steps of establishing a countermeasure behavior decision model based on a knowledge graph, wherein the countermeasure behavior decision model is used for expressing a production rule of game countermeasure.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements a method for determining a gambling behaviour based on a knowledge graph provided by the above methods, the method comprising:
acquiring fact data in a game countermeasure process; wherein the fact data is used for representing the confrontation state in the game confrontation process;
matching the fact data with the countermeasure behavior decision model to obtain a behavior decision result of game countermeasure;
the game countermeasure method comprises the steps of establishing a countermeasure behavior decision model based on a knowledge graph, wherein the countermeasure behavior decision model is used for expressing a production rule of game countermeasure.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (7)
1. A game countermeasure behavior decision-making method based on a knowledge graph is characterized by comprising the following steps:
acquiring fact data in a game countermeasure process of a military simulation game; wherein, the fact data is used for representing the confrontation state of each entity object in the game confrontation process;
matching the fact data with an confrontation behavior decision model to obtain a behavior decision result of the game confrontation, wherein the behavior decision result is used for triggering the updating of the behavior of each entity object in the military simulation game;
wherein the countermeasure behavior decision model is constructed based on a knowledge graph, and is used for representing the generative rules of the game countermeasure; the construction of the confrontation decision model based on the knowledge graph comprises the following steps:
acquiring a generative rule of the game confrontation;
extracting predicates, conclusion primitives and logical relations in the production rule; the predicates are used for representing decision conditions of the game countermeasure, the conclusion primitives are used for representing decision results of the game countermeasure, and the logic relationship is used for representing the relationship between the two predicates and/or the predicate and the conclusion primitives; the extracting predicates, conclusion primitives and logical relations in the production rule comprises: based on the game countermeasure production rule, acquiring the predicate, the conclusion primitive and the attribute value of the logical relationship, and completing extraction of the predicate, the conclusion primitive and the logical relationship;
constructing the confrontation decision model by using a knowledge graph based on the predicates, the conclusion primitives and the logical relationship; creating nodes of the knowledge graph based on the predicates and the conclusion primitives, creating edges of the knowledge graph based on the logical relations, wherein the edges of the knowledge graph are directed edges;
matching the fact data with an confrontation behavior decision model to obtain a behavior decision result of the game confrontation, wherein the behavior decision result comprises the following steps:
node matching: performing node matching on the fact data according to the attribute values of the nodes of the countermeasure decision model to obtain an initial node;
assertion calculation: performing assertion calculation according to the attribute value of the starting node;
edge matching: acquiring a termination node corresponding to the starting node based on an assertion calculation result of the starting node and an attribute value of an edge of the antagonistic behavior decision model;
and (3) node identification: and identifying whether the termination node is a conclusion primitive, if not, taking the termination node as an initial node, and repeatedly executing the steps of assertion calculation, edge matching and node identification, and if so, obtaining a behavior decision result of the game countermeasure according to the attribute value of the termination node.
2. The method for game countermeasure action decision based on knowledge graph of claim 1, wherein the construction of the countermeasure action decision model using knowledge graph based on the predicate, the conclusion primitive and the logical relationship comprises:
and (3) node creation: receiving a predicate or a conclusion primitive in the production rule, and detecting whether a node corresponding to the predicate or the conclusion primitive exists in the knowledge graph; if not, creating a current node in the knowledge graph based on the predicate or the conclusion primitive; if yes, taking the existing node as the current node, and executing a node detection step;
and (3) node detection: detecting whether the current node is the first node of the production rule, if not, executing an edge creating step, and if so, executing the node creating step;
creating an edge: and based on the logical relationship corresponding to the current node, creating a directed edge from the previous node of the current node to the current node, and repeatedly executing the steps of node creation, node detection and edge creation until all predicates, conclusion primitives and logical relationships in the production formula rule complete the creation of the node or the edge.
3. The method of any of claims 1-2, wherein after the construction of the warfare decision model based on the knowledge graph, the method further comprises:
storing the countermeasure decision model by adopting a data model based on a label attribute graph; the data model respectively and independently stores the nodes of the countermeasure decision model and the labels of the nodes, and indexes are arranged in the labels and used for indicating the storage positions of the nodes.
4. A game countermeasure behavior decision device based on a knowledge graph is characterized by comprising:
the data acquisition module is used for acquiring fact data in the game countermeasure process of the military simulation game; wherein, the fact data is used for representing the confrontation state of each entity object in the game confrontation process;
a decision matching module, configured to match the fact data with an confrontation behavior decision model to obtain a behavior decision result of the game confrontation, where the behavior decision result is used to trigger updating of behaviors of each entity object in the military simulation game;
wherein the confrontation behavior decision model is constructed based on a knowledge graph and is used for representing generative rules of the game confrontation; the construction of the confrontation decision model based on the knowledge graph comprises the following steps:
acquiring a generative rule of the game countermeasure;
extracting predicates, conclusion primitives and logical relations in the production rule; the predicates are used for representing decision conditions of the game countermeasure, the conclusion primitive is used for representing decision results of the game countermeasure, and the logic relation is used for representing the relation between the two predicates and/or the predicate and the conclusion primitive; the extracting predicates, conclusion primitives and logical relations in the production rule comprises: based on the game countermeasure generating formula rule, acquiring the predicate, the conclusion primitive and the attribute value of the logical relationship, and completing extraction of the predicate, the conclusion primitive and the logical relationship;
constructing the confrontation decision model by adopting a knowledge graph based on the predicates, the conclusion primitives and the logical relationship; creating nodes of the knowledge graph based on the predicates and the conclusion primitives, and creating edges of the knowledge graph based on the logic relations, wherein the edges of the knowledge graph are directed edges;
the step of matching the fact data with a countermeasure behavior decision model to obtain a behavior decision result of the game countermeasure comprises the following steps:
node matching: performing node matching on the fact data according to the attribute values of the nodes of the countermeasure decision model to obtain an initial node;
assertion calculation: performing assertion calculation according to the attribute value of the starting node;
edge matching: acquiring a termination node corresponding to the starting node based on an assertion calculation result of the starting node and the attribute value of the edge of the antagonistic behavior decision model;
and (3) node identification: and identifying whether the termination node is a conclusion primitive, if not, taking the termination node as an initial node, and repeatedly executing the steps of assertion calculation, edge matching and node identification, and if so, obtaining a behavior decision result of the game countermeasure according to the attribute value of the termination node.
5. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the method for knowledge-graph based gambling confrontation decision making of any of claims 1 to 3.
6. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the method for knowledge-graph based gambling confrontation decision making of any of claims 1 to 3.
7. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements the steps of a method for knowledge-graph based gambling pursuit decision making as claimed in any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111364693.9A CN114238648B (en) | 2021-11-17 | 2021-11-17 | Game countermeasure behavior decision method and device based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111364693.9A CN114238648B (en) | 2021-11-17 | 2021-11-17 | Game countermeasure behavior decision method and device based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114238648A CN114238648A (en) | 2022-03-25 |
| CN114238648B true CN114238648B (en) | 2022-11-08 |
Family
ID=80749888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111364693.9A Active CN114238648B (en) | 2021-11-17 | 2021-11-17 | Game countermeasure behavior decision method and device based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114238648B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829034B (en) * | 2023-01-09 | 2023-05-30 | 白杨时代(北京)科技有限公司 | Method and device for constructing knowledge rule execution framework |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019144039A1 (en) * | 2018-01-18 | 2019-07-25 | Risksense, Inc. | Complex application attack quantification, testing, detection and prevention |
| CN110110053A (en) * | 2018-02-01 | 2019-08-09 | 国际商业机器公司 | Logical connection is established between indirect language and affairs |
| CN110119773A (en) * | 2019-05-07 | 2019-08-13 | 中国科学院自动化研究所 | Global Situation Assessment side's method, the system, device of Strategic Games system |
| CN110277086A (en) * | 2019-06-25 | 2019-09-24 | 中国科学院自动化研究所 | Phoneme synthesizing method, system and electronic equipment based on dispatching of power netwoks knowledge mapping |
| WO2021180062A1 (en) * | 2020-03-09 | 2021-09-16 | 华为技术有限公司 | Intention identification method and electronic device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033063B (en) * | 2017-06-09 | 2022-02-25 | 微软技术许可有限责任公司 | Machine inference method based on knowledge graph, electronic device and computer readable storage medium |
| CN109857845B (en) * | 2019-01-03 | 2021-06-22 | 北京奇艺世纪科技有限公司 | Model training and data retrieval method, device, terminal and computer-readable storage medium |
| CN111046187B (en) * | 2019-11-13 | 2023-04-18 | 山东财经大学 | Sample knowledge graph relation learning method and system based on confrontation type attention mechanism |
| CN111949390A (en) * | 2020-08-13 | 2020-11-17 | 深圳市极限网络科技有限公司 | Multi-type large-scale task automatic scheduling method and system based on affair map |
| CN113297388B (en) * | 2021-04-25 | 2023-08-11 | 中国人民解放军军事科学院战争研究院 | Strategic event chained visualization method oriented to game analysis |
-
2021
- 2021-11-17 CN CN202111364693.9A patent/CN114238648B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019144039A1 (en) * | 2018-01-18 | 2019-07-25 | Risksense, Inc. | Complex application attack quantification, testing, detection and prevention |
| CN110110053A (en) * | 2018-02-01 | 2019-08-09 | 国际商业机器公司 | Logical connection is established between indirect language and affairs |
| CN110119773A (en) * | 2019-05-07 | 2019-08-13 | 中国科学院自动化研究所 | Global Situation Assessment side's method, the system, device of Strategic Games system |
| CN110277086A (en) * | 2019-06-25 | 2019-09-24 | 中国科学院自动化研究所 | Phoneme synthesizing method, system and electronic equipment based on dispatching of power netwoks knowledge mapping |
| WO2021180062A1 (en) * | 2020-03-09 | 2021-09-16 | 华为技术有限公司 | Intention identification method and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114238648A (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ilachinski | Irreducible semi-autonomous adaptive combat (ISAAC): An artificial-life approach to land combat | |
| Kewley et al. | Computational military tactical planning system | |
| CN105678030B (en) | Divide the air-combat tactics team emulation mode of shape based on expert system and tactics tactics | |
| CN112600795B (en) | Method and system for collapsing combat network under incomplete information | |
| CN109529338A (en) | Object control method, apparatus, Electronic Design and computer-readable medium | |
| CN114238648B (en) | Game countermeasure behavior decision method and device based on knowledge graph | |
| CN106844458A (en) | Show method, computing device and the storage medium of user's internet behavior track | |
| CN106874339A (en) | A kind of methods of exhibiting of circulant Digraph and its application | |
| Ramnath et al. | Scene graph based image retrieval--a case study on the CLEVR dataset | |
| Porteous et al. | Extending narrative planning domains with linguistic resources | |
| Sanchez et al. | Improving game modeling for the quoridor game state using graph databases | |
| Fay | Driving story generation with learnable character models | |
| Eisen | Simulating human game play for level difficulty estimation with convolutional neural networks | |
| Duan et al. | MOBA game item recommendation via relation-aware graph attention network | |
| CN114239834A (en) | Adversary relationship reasoning method and device based on multi-round confrontation attribute sharing | |
| Saito et al. | Grouping nodes for Monte-Carlo tree search | |
| Liu et al. | Introduction of a new dataset and method for location predicting based on deep learning in wargame | |
| CN110598182A (en) | Information prediction method and related equipment | |
| CN111723941A (en) | Rule generation method and device, electronic equipment and storage medium | |
| CN116167377A (en) | Method, device and storage medium for constructing combat ontology | |
| Hu et al. | A Neural Network-Based Intelligent Decision-Making in the Air-Offensive Campaign with Simulation | |
| Sephton et al. | Using association rule mining to predict opponent deck content in Android: Netrunner | |
| CN115952867B (en) | Processing method, processing device, processing equipment and storage medium for game countermeasure data | |
| CN117973494A (en) | Method, device and medium for enabling reinforcement learning oriented to multiple agents to be interpretable | |
| CN114239833B (en) | Military countermeasure rule confidence coefficient calculation method and device based on probability soft logic |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |