System capability requirement recommendation method and device based on confidence rule base reasoning

By decomposing action plans into tasks based on a confidence rule base and using the confidence rule base for precondition matching and evidence reasoning, the lack of recommendation technology in military requirement scenarios is solved, enabling effective recommendation of military capability requirements and supporting requirement analysis by task forces and research departments.

CN115099446BActive Publication Date: 2024-11-26NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210264952.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-17
Publication Date
2024-11-26
Estimated Expiration
2042-03-17

AI Technical Summary

Technical Problem

Existing technologies rarely combine military requirement scenarios with recommendation technologies, and there is a lack of effective methods for recommending military capability requirements.

Method used

The method employs a confidence rule base-based reasoning approach to decompose the action plan into tasks. By acquiring the input data of the tasks, the method utilizes a pre-set confidence rule base to perform precondition matching and evidence reasoning, calculates the reliability of the task's capability requirements, and achieves capability requirement recommendations.

Benefits of technology

It integrates recommendation technology with military capability requirement scenarios, providing mission-oriented forces and research departments with capability requirement recommendations adapted to missions, supporting mission execution and force development, and filling the technological gap in military requirement recommendation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115099446B_ABST
    Figure CN115099446B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for recommending system capability requirements based on confidence rule base reasoning, including: decomposing an action plan into at least one task according to a pre-built capability requirement system; obtaining input data of at least one task; performing premise matching conversion on the input data according to a preset confidence rule base to obtain a set of matched premise conditions; performing rule matching according to the set of premise conditions, and calculating the activation degree and correction factor of each matched rule in combination with the confidence rule base; performing evidence reasoning in combination with the confidence rule base according to the activation degree and the correction factor to obtain the capability requirement reasoning result of at least one task. The present invention combines recommendation technology with military capability requirement scenarios, and can recommend capability requirements suitable for tasks for task forces and scientific research departments, providing strong support for the execution of tasks, the growth of forces, and the development of military theories.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of military capability requirement recommendation, and specifically relates to a system capability requirement recommendation method and device based on confidence rule base reasoning. Background Art

[0002] In recent years, the theories and methods of requirements engineering have been continuously improved, and the application scenarios have continued to increase, such as requirement description in software development scenarios, application of requirements engineering methods in civil aircraft fuselage development, and behavior-driven requirements engineering in agile product line engineering scenarios.

[0003] Since the recommendation system was proposed in the 1990s, after decades of development, recommendation technology has achieved rapid development and produced many different types of algorithms and technologies. At present, the more widely used recommendation methods are mainly divided into recommendation based on user behavior characteristics, recommendation based on tags, recommendation based on context and recommendation technology based on knowledge. Knowledge-based recommendation technology is mainly implemented by the constructed rule base. The method based on evidence reasoning is one of them. Evidence theory, also known as DS evidence theory (Dempster-Shafer Theory of Evidence) and belief theory. The belief rule-base inference method using the evidential reasoning (RIMER) based on the evidence reasoning algorithm has the ability to model and analyze fuzzy data, subjective data, probabilistic data and ignorant information. The evidence reasoning method can be applied to evidence network reasoning of conflict measurement, power grid emergency capability evaluation, health analysis of unmanned systems, aerial target recognition, sensor failure prediction, etc., but there are few studies that combine military demand scenarios with recommendation technology. Summary of the invention

[0004] The present invention provides a system capability requirement recommendation method and device based on confidence rule base reasoning to solve the problem that military requirement scenarios are rarely combined with recommendation technologies.

[0005] Based on the above purpose, an embodiment of the present invention provides a system capability requirement recommendation method based on confidence rule base reasoning, including: decomposing an action plan into at least one task according to a pre-constructed capability requirement system; obtaining input data of the at least one task, the input data including at least one measurement scale standard of the task and the reliability corresponding to the measurement scale; performing premise matching conversion on the input data according to a preset confidence rule base to obtain a premise set matched by the input data, the confidence rule base including a task measurement scale value table and at least one rule, each rule including at least a premise and a conclusion, each element in the premise set is a tuple consisting of any one of the preconditions matched by the input data and a first matching degree corresponding to the premise; performing rule matching according to the premise set, and calculating the activation degree and correction factor of each matched rule in combination with the confidence rule base; performing evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of the at least one task.

[0006] Optionally, performing a prerequisite matching conversion on the input data according to a preset confidence rule base to obtain a set of prerequisites matched to the input data includes: calculating the degree of similarity between any of the measurement scale standards in the input data and each value corresponding to the measurement scale in the task measurement scale value table according to the task measurement scale value table in the confidence rule base; calculating the first degree of matching between any of the measurement scale standards and each of the prerequisites according to the similarity and the reliability; and combining all of the prerequisites matched to each of the measurement scale standards in the input data to form the prerequisite set.

[0007] Optionally, the rules in the confidence rule base also include weights, and the rule matching is performed according to the precondition set, and the activation degree and correction factor of each matched rule are calculated in combination with the confidence rule base, including: matching the precondition set with the preconditions of the rules in the confidence rule base to obtain at least one matched rule; for each matched rule, aggregating the first matching degree corresponding to each precondition in the rule by applying an aggregation function to obtain a second matching degree of the rule; calculating the activation degree of each rule according to the second matching degree and the weight of each rule; and calculating the correction factor of each rule according to the first matching degree of each precondition in each rule.

[0008] Optionally, the first matching degrees corresponding to each of the preconditions in the rule are aggregated by applying an aggregation function to obtain a second matching degree of the rule, including: performing an AND logical probability operation on the first matching degrees of each pair of the preconditions in the rule to obtain the second matching degree of the rule; or performing an OR logical probability operation on the first matching degrees of each pair of the preconditions in the rule to obtain the second matching degree of the rule.

[0009] Optionally, the conclusion corresponding to the rule in the confidence rule base includes capability requirements, at least one result and a corresponding result reliability, and the evidential reasoning is performed based on the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of the at least one task, including: correcting the reliability of each result in the rule according to the correction factor; determining the capability requirements according to the matched rules based on the confidence rule base; and calculating the capability requirement reliability corresponding to each capability requirement according to the activation degree of each matched rule and the corrected result reliability, to obtain the capability requirement reasoning result of the at least one task.

[0010] Optionally, the calculating the capability requirement reliability corresponding to each capability requirement according to the activation degree of each matched rule and the modified result reliability respectively includes: obtaining at least one rule corresponding to each capability requirement from each matched rule; and calculating the final result reliability of each result corresponding to each capability requirement respectively according to the activation degree of the rule corresponding to each capability requirement and the modified result reliability by applying the following relationship as the corresponding capability requirement reliability: Among them, RL is the inclusion result c l The rule set, Θ l The result c l The final reliability.

[0011] Optionally, the method further includes: when an action plan is decomposed into a plurality of the tasks, aggregating the capability requirements in the plurality of the tasks to obtain each capability requirement of the action plan and the corresponding capability requirement reliability.

[0012] Optionally, the capability requirements in the multiple tasks are aggregated to obtain the capability requirements of the action plan and the corresponding capability requirement reliability, including: combining the capability requirements in the multiple tasks as the inferred capability requirements of the action plan; for the same capability requirements in the multiple tasks, selecting the capability requirement with the highest priority and the highest final result reliability as the aggregated capability requirement.

[0013] Based on the same inventive concept, an embodiment of the present invention further proposes a system capability requirement recommendation device based on confidence rule base reasoning, comprising: an action decomposition unit, used to decompose an action plan into at least one task according to a pre-constructed capability requirement system; a data acquisition unit, used to acquire input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and the reliability corresponding to the measurement scale; a condition matching unit, used to perform premise matching conversion on the input data according to a preset confidence rule base, and acquire a premise set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each of which includes at least a premise and a conclusion, and each element in the premise set is a tuple consisting of any one of the advance conditions matched by the input data and a first matching degree corresponding to the premise; a rule matching unit, used to perform rule matching according to the premise set, and calculate the activation degree and correction factor of each matched rule in combination with the confidence rule base; an evidence reasoning unit, used to perform evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base, and acquire the capability requirement reasoning result of the at least one task.

[0014] Based on the same inventive concept, an embodiment of the present invention further proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the methods described above when executing the program.

[0015] The technical effect of the present invention is that, as can be seen from the above, an embodiment of the present invention provides a method and device for recommending system capability requirements based on confidence rule base reasoning, which decomposes an action plan into at least one task according to a pre-built capability requirement system; obtains input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and the reliability corresponding to the measurement scale; performs premise matching conversion on the input data according to a preset confidence rule base, and obtains a premise set matched to the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, wherein each rule includes at least a premise and a conclusion, and the premise Each element in the set of items is a tuple consisting of any of the advance conditions matched by the input data and a first matching degree corresponding to the prerequisite; rule matching is performed according to the prerequisite set, and the activation degree and correction factor of each matched rule are calculated in combination with the confidence rule base; evidence reasoning is performed based on the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of at least one task, which can combine the recommendation technology with the military capability requirement scenario, and can recommend capability requirements suitable for the task to the task force and scientific research department, providing strong support for the execution of the task, the growth of the troops, and the development of military theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of a method for recommending system capability requirements based on confidence rule base reasoning in an embodiment of the present invention;

[0018] Figure 2 A schematic diagram of a capability requirement system structure in an embodiment of the present invention;

[0019] Figure 3 A schematic diagram of the relationship between missions, military operations, and tasks in an embodiment of the present invention;

[0020] Figure 4 A schematic diagram of the evidential reasoning algorithm flow for a single task in an embodiment of the present invention;

[0021] Figure 5 A schematic diagram of capability requirement aggregation in an embodiment of the present invention;

[0022] Figure 6 An example diagram of capability requirement recommendation for a combat mission of a certain engineering unit in an embodiment of the present invention;

[0023] Figure 7 It is a structural schematic diagram of a system capability requirement recommendation device based on confidence rule base reasoning in an embodiment of the present invention;

[0024] Figure 8 Schematic diagram of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0026] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0027] The embodiment of the present invention also provides a system capability requirement recommendation method based on confidence rule base reasoning. Figure 1 As shown in the figure, the system capability requirement recommendation method based on confidence rule base reasoning includes:

[0028] Step S11: Decompose an action plan into at least one task according to the pre-built capability requirement system.

[0029] In the embodiment of the present invention, in order to facilitate the description and construction of the capability requirement system, the capability requirements need to be divided into multiple levels to form a tree structure of the capability requirements. Figure 2 As shown, It represents the first capability requirement of the first-level capability requirement system, and the superscript indicates the level where the capability requirement is located. lmIt represents the number of capability requirements with the same parent capability requirement at the first layer. CM is the set of capability measures of the corresponding capability requirements. In the capability requirement system, each capability requirement interacts with the corresponding weapon and equipment system in the weapon and equipment system. The capability requirements are derived from the actual situation and subsequent development of the weapon and equipment system, and the weapon and equipment system will also advance and develop in response to the continuous proposal of capability requirements. In addition, the capability requirements also interact with the military service organization system. The channel of this influence is reflected in the quality of personnel: the continuous development of existing military services and the birth of emerging military services require their combatants to have corresponding and higher capabilities and qualities, while also promoting the development and progress of the weapon and equipment system, thus giving rise to new capability requirements; and the increasing capability requirements also put forward higher requirements for the quality of personnel. Within the capability requirement system, there is also a mutual influence relationship between the various capability requirements. The bottom-level capability requirements will support the upper-level capability requirements. For the capability requirement items of non-leaf nodes in the capability requirement system, they must contain a series of lower-level capability requirements, and the lower-level capability requirements will support the upper-level capability requirements. In other words, only when the capability measurement values ​​of the lower-level capability requirements are relatively ideal, the corresponding upper-level capability requirement value level will reach the "strong" level.

[0030] See also Figure 3 , the mission is completed by a series of military actions, and a military action includes several combat tasks. Each combat task will have a series of measurement scales to provide guidance for combat troops to perform tasks. As a guide for combat troops' actions, the measurement scale will be set with certain standards. The measurement scale is an attribute that reflects the fundamental characteristics of the task and distinguishes it. The standard is used to define the acceptability of task execution and is a description of the measurement scale. Reliability is an attribute that measures the degree of credibility of the standard value. Combat tasks and their measurement scales and standards reflect the needs and expectations of superiors for the performance of subordinate actions. Therefore, mission requirements can be described by a series of measurement scales and standards. The measurement scales, standards and corresponding reliability of a series of combat tasks constitute the mission requirements and are presented in the form of a requirements list. The task requirements in the list need to determine the background of the task (various objective constraints), the time of task execution, the location of the action, the degree of task completion and other necessary information. This information will help to understand the contribution of various combat actions to the completion of the overall mission. In step S11, an action plan is decomposed into at least one task according to the aforementioned capability requirement system.

[0031] Step S12: Acquire input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and a reliability corresponding to the measurement scale.

[0032] In the embodiment of the present invention, the transformation process from mission tasks to capability requirements is the core content of completing capability requirement recommendations. The input data of a single task may include multiple measurement scales and the reliability corresponding to each measurement scale. The input data of the reasoning is the measurement scale x and the reliability ε of a single task, which can be expressed as:

[0033] (x1, ε1), (x2, ε2),…, (x P , ε P ), where ε P For the input data x P The reliability represents the certainty or possibility of the occurrence of the data value.

[0034] Step S13: Perform a prerequisite matching conversion on the input data according to a preset confidence rule base to obtain a prerequisite set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each of which includes at least a prerequisite and a conclusion, and each element in the prerequisite set is a tuple consisting of any of the prerequisites matched by the input data and a first matching degree corresponding to the prerequisite.

[0035] In the embodiment of the present invention, the existing historical experience and knowledge can be used to establish a mapping relationship between task requirements and capability requirements, and to construct a confidence rule base as the basis for evidence reasoning. The confidence rule base includes a task measurement scale value table and at least one rule. The basic structure of the confidence rule base can be expressed as follows: RB =<M,C,F> , where M = {M1, M2, ..., M P} is a set of task measurement scales, an element in the set M is the value set of the standard for measuring the scale of a task in the confidence rule base, i The meaning of each element in is the measurement scale M i Standard values ​​in the confidence rule base. The measurement scale and the corresponding standard values ​​can be used together as the prerequisites of the confidence rule base. L} represents the set of capability requirements, and the value of the element (capability requirement) is used as the reasoning result of the confidence rule base. There are three types of value results {strong, medium, weak}, representing the degree of demand for capability requirements. F is the transformation relationship from premise to conclusion.

[0036] In the confidence rule base, each rule includes at least a premise and a conclusion, and may also include a weight. The conclusion corresponding to the rule in the confidence rule base includes capability requirements, at least one result, and the corresponding result confidence. The result of each rule has confidence, which better characterizes the uncertainty of the rule. The rules with confidence structure are as follows:

[0037]

[0038]

[0039] in, Represents a set of preconditions, The measurement scale involved in this rule is is the standard value in the corresponding rule base, T k is the number of preconditions. ik In the kth rule, when the premise is met, the result is c i of reliability and meets The relative weight of this rule in the confidence rule base is defined as θ k , the relative weight of each prerequisite is

[0040] like This means that all possible outcomes are covered by the rule, and the information is called complete; It means that the result cannot be predicted at all and the output result is completely unknown; This means that the result is partially known. Please refer to Table 5 for the specific expression of the rules in the confidence rule base. For example, Rule 27 in the confidence rule base is: When a device is mounted on a mobile platform, the communication equipment is in good condition, and the communication time with a target at a distance of 50KM is not less than 2h, its anti-electromagnetic interference capability requirements are (strong, 0.7), (medium, 0.25), (weak, 0.05). The relative weight of this rule in the confidence rule base is 0.6, and the weights of the preconditions are 0.3, 0.5, and 0.2 respectively. The rule can be expressed in mathematical language as follows:

[0041] R 27 :if(M1=“Good”)^(M2=50)^(M3=2),

[0042] then{(strong, 0.7), (medium, 0.25), …, (weak, 0.05)}

[0043] θ 27 =0.6,δ 27,1 =0.3,δ 27,2 =0.5,δ 27,3 =0.2

[0044] In an embodiment of the present invention, the conversion process of input data refers to the process of matching the input data with the preconditions in the confidence rule base. The preconditions of several rules with a higher degree of matching with the input data can be found through matching calculation, so that the rules with a higher degree of matching with the input data can be better found, and the conclusion reliability can be corrected. In step S13, optionally, the similarity between any of the measurement scale standards in the input data and each value corresponding to the measurement scale in the task measurement scale value table is calculated according to the task measurement scale value table in the confidence rule base; the first matching degree of any of the measurement scale standards matching each of the preconditions is calculated according to the similarity and the reliability; all the preconditions matched by each of the measurement scale standards in the input data are collected to form the precondition set.

[0045] There are many types of similarity functions, which are generally selected according to the type and characteristics of the data. The embodiment of the present invention adopts a similarity function in which the capability attribute is a discrete value, and the expression is:

[0046]

[0047] in, It is called a similarity function and is used to characterize x i With m ij The task measurement scale standard set in the task measurement scale value table in the confidence rule base Its value is m ij ∈M i . Assume that the elements in the collection are sorted in a monotonically increasing sequence.

[0048] After obtaining the input data with confidence, the matching degree between the input data and the premise conditions can be calculated by the following relationship:

[0049]

[0050]

[0051] Among them, T(x i , ε i ) represents the input data (x i , ε i ) After data transformation, the set of matched prerequisites, the elements of the set are the task measurement scale indicator set M i The jth value m ij and the corresponding first matching degree α ij The binary group composed of ij∈[0, 1]. For example, the values ​​of the task measurement scale in the confidence rule base are {30, 60, 80}, and the input data value is (65, 1). Then the similarity between the original input data and the corresponding rule standard value in the confidence rule base is Therefore, the matching degree between the input data and the corresponding rule standard value is α 30 =0,α 60 =0.75,α 80 =0.25.

[0052] Step S14: performing rule matching according to the precondition set, and calculating the activation degree and correction factor of each matched rule in combination with the confidence rule base.

[0053] In the embodiment of the present invention, firstly, the set of prerequisites is matched with the prerequisites of the rules in the trusted rule base to obtain at least one matching rule. In this way, the original input data is converted into a series of prerequisites of the corresponding rules, that is, k means that the premise condition belongs to the kth rule.

[0054] Then, for each of the matched rules, the first matching degree corresponding to each of the preconditions in the rule is aggregated by applying an aggregation function to obtain the second matching degree of the rule. The calculation of the matching degree between the input data and the rule is essentially the process of aggregating the matching degrees of a series of rule preconditions, because the matching degree between the input and the rule depends on the sum of its matching degrees with the preconditions of the rule. The process of summarizing the matching degrees of a series of preconditions requires the use of an aggregation function, which is denoted as: There are two ways to aggregate preconditions. One is to connect all preconditions using “∧”, that is, to perform an AND logic probability operation on the first matching degree of each pair of preconditions in the rule to obtain the second matching degree of the rule. The other is to connect all of them using “∨”, that is, to perform an OR logic probability operation on the first matching degree of each pair of preconditions in the rule to obtain the second matching degree of the rule. In the case where two connection symbols appear at the same time, they can be converted into a form containing only one connection symbol through corresponding mathematical logic operations.

[0055] In the embodiment of the present invention, when the prerequisites are connected using “∧”, the aggregation method of the prerequisite matching degree is similar to the probability multiplication under independent conditions, that is, the matching degrees of all prerequisites are multiplied cumulatively, and the aggregation result is:

[0056]

[0057]

[0058] in, is the relative weight of the prerequisite, α k is the matching degree of the kth rule.

[0059] When the "∨" is used to connect the premise conditions, only one of the premise rules needs to be met. The matching degree is calculated as follows:

[0060]

[0061]

[0062]

[0063]

[0064] Then, the activation degree of each rule is calculated according to the second matching degree and the weight of each rule. The matching degree α with the kth rule is obtained according to the aggregation function of the above two cases. k After that, we can combine the weight θ of the kth rule itself k Activate and normalize the activation level ω k It can be expressed as:

[0065]

[0066] Among them, when α k = 0, there is ω k = 0. At this time, the matching degree between the input data and the rule is 0, and the activation degree is also 0, which means that the rule is not activated. In the confidence rule base, the weight of the rule is θ k Usually they are equal, but if necessary, the weights θ can be changed k The value of is adjusted.

[0067] The embodiment of the present invention further calculates the correction factor of each of the rules according to the first matching degree of each of the preconditions in each of the rules. k The calculation method is:

[0068]

[0069]

[0070] M in the formula t is a measure, R k It's a rule, T krepresents the number of metrics contained in the kth rule, k represents the rule number, k = 1, 2, ..., K. There are K rules in the confidence rule base.

[0071] Step S15: performing evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain a capability requirement reasoning result of the at least one task.

[0072] In an embodiment of the present invention, optionally, each result credibility in the rule is corrected according to the correction factor; the capability requirement is determined according to each matched rule based on the confidence rule library; the capability requirement credibility corresponding to each capability requirement is calculated according to the activation degree of each matched rule and the corrected result credibility, so as to obtain the capability requirement inference result of the at least one task.

[0073] After the actual input data is transformed, it does not completely match the capability requirements in the premise conditions in the confidence rule base. It is often possible that some premise conditions in a rule match the transformed input data. For this reason, it is necessary to make some corrections to the result reliability of each result in the conclusion of the rule using the following relationship:

[0074] in, Represents the reliability of the result after correction, μ k represents the correction factor calculated previously. l represents the number of the result in the conclusion of each rule, l = 1, 2, ..., L. Each rule in the confidence rule base has L results. If the input data contains all the prerequisites of the rule, that is, Both but It means that when the information is complete, no correction is needed to the result. So far, the conversion of input data, the activation degree of input data to the corresponding rules and the correction of the reliability of each result in the conclusion are completed.

[0075] Then, based on the confidence rule base, the capability requirements in the conclusions of the matched rules are merged to obtain the capability requirements of the task. Among the matched rules, there may be multiple rules with the same capability requirements in the conclusions. In this case, the capability requirement confidence of the capability requirement needs to be calculated based on the aggregation of the multiple rules. For a series of input data (x1, ε1), (x2, ε2), ..., (x P , ε P ), the activated rule set is RS = {R1, R2, ..., R |RS|}, the activation levels of the rules in this set are ω1, ω2, …, ω |RS| The corrected reliability of the result l of rule k is Optionally, at least one rule corresponding to each capability requirement is obtained from each matched rule; and the final result reliability of each result corresponding to each capability requirement is calculated respectively by applying the following relationship according to the activation degree of the rule corresponding to each capability requirement and the modified result reliability, as the corresponding capability requirement reliability:

[0076]

[0077] Among them, RL is the inclusion result c l The rule set, Θ l The result c l The final confidence of a single task is (c1, Θ1), (c2, Θ2), …, (c L , Θ L ). For the evidential reasoning algorithm flow of a single task in an embodiment of the present invention, see Figure 4 , perform data conversion on the requirement input of a single task to obtain a set of prerequisites, then perform rule matching and reliability correction based on the confidence rule base, and then perform evidence reasoning to obtain the capability requirement output of the single task.

[0078] In an embodiment of the present invention, when an action plan is decomposed into multiple tasks, the capability requirements in the multiple tasks are aggregated to obtain the capability requirements of each action plan and the corresponding capability requirement reliability. After completing the reasoning of the capability requirements of a single task, the capability requirements of multiple tasks need to be aggregated. Capability requirement aggregation is the process of converting the capability requirements of a single task into the capability requirements of the entire military operation. The aggregation process refers to the process of summarizing the capability requirements of multiple tasks and merging repeated capability requirements. Optionally, the capability requirements in multiple tasks are collected as the inferred capability requirements of the action plan; for the same capability requirements in multiple tasks, the capability requirement with the highest priority and the highest final result reliability is selected as the aggregated capability requirement. For example, for multiple conclusion items of the same capability C0, C0 = {(A, x1), (B, y1), (C, z1)},…, C0 i ={(A, x i ), (B, y i ), (C, z i )}. The method of pairwise aggregation of capabilities mainly relies on comparing the size of the confidence aggregation value Ji, and the calculation method of Ji is: i =ρ1x i +ρ2y i +ρ3z i (ρ1>>ρ2>>ρ3), J iThe conclusion item with the largest value is the conclusion item taken by the final aggregation. (ρ1>>ρ2>>ρ3), which means that result x has the highest priority. The capability requirement with the highest final result confidence of result x in multiple tasks is taken as the capability requirement after aggregation. Figure 5 As shown, task 1 corresponds to capability requirements 1, 2, and 5, while task 3 corresponds to capability requirements 1, 2, 3, and 5. After aggregating tasks 1-3, we obtain capability requirements 1, 2, 3, 4, and 5 corresponding to the action plan. Among them, the capability requirement 1 included in both task 1 and task 3 is aggregated to obtain the capability requirement 1 corresponding to the action plan.

[0079] The following example illustrates the combat mission of a certain engineering unit and recommends the corresponding capability requirements, such as Figure 6 As shown in the figure, it is divided into the following steps: describing typical combat tasks of engineers according to the capability requirement system; performing evidence reasoning based on the confidence rule base to complete intelligent recommendation of capability requirements for individual tasks; performing intelligent capability aggregation to complete intelligent recommendation of capability requirements for mission tasks.

[0080] According to the method of constructing the capability requirement system, the engineer capability requirement system can be obtained as follows: mobility capability requirement C1: land mobility capability requirement C2: 11 , Complex terrain passing capability requirements C 12 Firepower strike capability requirement C2: Ground firepower suppression capability requirement C 21 , Anti-armor capability requirement C 22 , Anti-fortification capability requirement C 23 , air strike capability requirement C 24 . Protection capability requirement C3: Firepower protection capability requirement C 31 , Nuclear protection capability requirement C 32 , biological protection capability requirements C 33 , Chemical protection capability requirements C 34 Engineering support capability requirement C4: Minefield laying capability requirement C 41 , minefield clearance capability requirement C 42 , bridge erection capacity requirement C 43 , road paving capacity demand C 44 , construction capacity requirements C 45 , barrier-breaking capability requirement C 46 Information perception capability requirement C5: Communication capability requirement C 51 , reconnaissance capability requirements C 52 .

[0081] The above summarizes five first-level capability requirements from C1 to C5. Each capability requirement includes C 11 to C 52 And several other secondary capability requirements.

[0082] The combat operations assigned to the engineering squad in a certain exercise are simulated as a scenario. The scenario content is as follows: Task 1. A certain unit is scheduled to launch an attack on position A at XX hours and XX minutes. The subordinate engineering squad cooperates with the reconnaissance squad to approach the front line for engineering reconnaissance, find out the situation of the enemy's defense zone, and report it. Task 2. The unit will start the attack at XX hours and XX minutes. The subordinate engineering squad is responsible for opening up a passage for the attacking squad; building bridges in areas that are difficult to pass, destroying enemy obstacles, and suppressing and attacking some firepower points to ensure that the main force advances to the offensive area. Task 3. The main attacking squad captures position A and switches to defense. The engineering squad is responsible for building fortifications at position A, assisting in defense, and ensuring that position A is held; setting up minefield obstacles quickly and at a long distance outside the position to delay possible invading enemies. According to the above scenario, based on Figure 3 The task decomposition method is used to decompose the task scenario into three tasks. The three tasks are three combat tasks under the same military operation, which are interconnected and progressive. The generated task requirement lists are shown in Tables 1 to 3.

[0083] Table 1 Task 1 Task requirements list

[0084]

[0085] Table 2 Task 2 Task Requirements List

[0086]

[0087] Table 3 Task 3 Task Requirements List

[0088]

[0089]

[0090] The possible values ​​of the standards of each measurement scale in the confidence rule base are shown in Table 4. Some examples of the confidence rule base rules are shown in Table 5. In Table 5, A in the inference conclusion represents "strong" capability requirements, B represents "medium" capability requirements, and C represents "weak" capability requirements.

[0091] Table 4 Table of value of task measurement scale in the confidence rule base

[0092]

[0093] Table 5 Examples of some rules in the trust rule base

[0094]

[0095] The rule weights θ in the confidence rule base are all the same, which has no effect on the inference results. Therefore, the weight value column is not given in the above confidence rule base rule example table. In actual applications, the weight of each rule can be added. For the convenience of demonstration and calculation, the weights θ between rules and the precondition weights δ in each rule are the same, so they can be omitted in the calculation process.

[0096] Taking Task 1 of the above tasks as an example, the input data is the input content of Table 1:

[0097] (M 11 =1200,1),(M 12 =2,1),(M 13 =10,1),

[0098] (M 14 =90,1),(M 15 =2.5,1)

[0099] Compare the input value with the value in the value table, i.e., Table 4, and calculate the degree of similarity:

[0100]

[0101]

[0102]

[0103] Calculate the matching degree with the preconditions in the confidence rule base based on the similarity The data conversion results are as follows:

[0104] (M 11 =1000,0.6),(M 11 =1500,0.4),

[0105] (M 12 =2,1),(M 13 =10,1),(M 14 =90,1),

[0106] (M 15 =2,0.5),(M 15 =3,0.5)

[0107] According to the converted input data, the preconditions of the trust rule base are matched, and the rules 1, 2, 4, 5, 7-16, 19, 20, 22-26 are activated. The value of is used to calculate the activation degree of each rule ω kThe values ​​are: [0.04, 0.04, 0, 0.04, 0.04, 0, 0.04, 0.04, 0.08, 0.03, 0.03, 0.03, 0.05, 0.05, 0.05, 0.08, 0, 0.04, 0.04, 0, 0.04, 0, 0.08, 0.04, 0.04, 0].

[0108] According to the matching of the premise conditions of each rule, the correction factor μ of each rule conclusion is calculated. k The values ​​are: [0.5, 0.5, 0, 0.5, 0.5, 0, 0.75, 0.75, 1, 0.7, 0.4, 0.4, 0.8, 0.6, 0.6, 1, 0, 0, 0.5, 0.5, 0, 0.75, 0.75, 1, 0.5, 0.5, 0].

[0109] Combined with the activation degree of each rule obtained k And the correction factor μ of the conclusion k , the result output of task one is calculated as:

[0110] C 11 ={(A, 0.146875), (B, 0.03958333), (C, 0.04270833)};

[0111] C 21 ={(A, 0.156), (B, 0.04133333), (C, 0.036)}

[0112] C 51 ={(A, 0.171875), (B, 0.04479167), (C, 0.0125)}

[0113] Similarly, the results of Task 2 and Task 3 are output.

[0114] Task 2 output:

[0115] C 21 ={(A, 0.15)(B, 0.04)(C, 0.06)}

[0116] C 42 ={(A, 0.18)(B, 0.0475)(C, 0.0225)}

[0117] C 43 ={(A, 0.17)(B, 0.03)(C, 0.05)}

[0118] C 46 ={(A, 0.165)(B, 0.05)(C, 0.035)}

[0119] Task 3 output:

[0120] C 21 ={(A, 0.12878788)(B, 0.05151515)(C, 0.03181818)}

[0121] C 31 ={(A, 0.12878788)(B, 0.05151515)(C, 0.03181818)}

[0122] C 41 ={(A, 0.3969697)(B, 0.10909091)(C, 0.06969697)}

[0123] According to the output of Task 1, the capability requirements of the engineering unit are recommended in the following order: Communication capability requirements C 51 , Ground fire suppression capability requirement C 21 , land mobility requirements C 11 The recommended result of the capability requirement for Task 2 is: Minefield Clearance Capability Requirement C 42 , bridge erection capacity requirement C 43 , barrier-breaking capability requirement C 46 , Ground fire suppression capability requirement C 21 The recommended result of the capability requirement for Task 3 is: Minefield laying capability requirement C 41 , Firepower protection capability requirement C 31 , Ground fire suppression capability requirement C 21 .

[0124] According to the capability requirement maximization aggregation model, the ground fire suppression capability requirement C 21 It appears in the three task recommendation results, so it needs to be aggregated. The final result confidence of the three tasks corresponding to the ability requirements is:

[0125] Task 1: {(A, 0.156), (B, 0.04133333), (C, 0.036)}

[0126] Task 2: {(A, 0.15), (B, 0.04), (C, 0.06)}

[0127] Task 3: {(A, 0.12878788), (B, 0.05151515), (C, 0.03181818)}

[0128] Therefore, after aggregation, the conclusion item with the highest confidence in the final result should be taken, that is, C 21={(A, 0.156),(B, 0.04133333),(C, 0.036)}. After capacity aggregation, the total capacity requirements under this scenario derived from the three tasks are shown in Table 6.

[0129] Table 6 Ability requirement reliability table

[0130]

[0131] The final recommendation results are in the following order of confidence: Minefield laying capability requirement C 41 , minefield clearance capability requirement C 42 , communication capability requirement C 51 , bridge erection capacity requirement C 43 , barrier-breaking capability requirement C 46 , Ground fire suppression capability requirement C 21 , land mobility requirements C 11 , Firepower protection capability requirement C 31 .

[0132] From the perspective of actual conditions, the capability requirement with higher reliability is mainly the minefield laying capability requirement C. 41 , minefield clearance capability requirement C 42 Such requirements are mainly for the professional and technical capabilities of engineers, which highlight the technical characteristics of engineers themselves and have irreplaceable characteristics; and the corresponding last few capability requirements, such as ground fire suppression capability requirements C 21 , land mobility requirements C 11 , Firepower protection capability requirement C 31 It is a more general capability requirement, which can be supplemented by support from other arms, so the demand is relatively low.

[0133] The above analysis results of the combat mission of a certain engineering unit are in line with the actual situation. As a technical branch of the army, engineers pay more attention to and highlight their engineering support and support business. Therefore, engineers often have a stronger demand for engineering support capabilities, which is consistent with the recommended results. In actual combat, the fire suppression and cover capability requirements of engineers can be supported by infantry, artillery, and armored troops. The air defense capability requirements are also weakened due to the cover of air defense forces. The nuclear, biological and chemical protection requirements can be provided by chemical defense troops. In the recommended results, the non-professional capability requirements of engineers are ranked at the back, which is also consistent with the actual situation. From the perspective of experimental data, in the recommended results, the reliability values ​​of various capability requirements are generally low. The reason is that the number of activated preconditions of a rule is small, resulting in a low correction factor value. Therefore, the correction factor values ​​given are mostly concentrated around 0.5, resulting in a generally low reliability of the final result. At the same time, some capability requirements that appear in the engineer capability requirement system are not reflected in the recommended results. The reason is that this method recommends corresponding capability requirements according to task requirements. Due to the limited number of task types in the examples, not all rules are activated and not all capability requirements are recommended. In actual applications, by building a large and complete trust rule library and inputting sufficient task measurement scales, the recommendation results are more likely to cover most capability requirements.

[0134] The system capability requirement recommendation method based on confidence rule base reasoning in the embodiment of the present invention can assist troops to quickly and clearly find their own capability requirements through tasks, thereby providing strong support for the execution of tasks, the growth of troops, and the development of military theories. The embodiment of the present invention combines recommendation technology with military capability requirement scenarios, realizes the function of recommending capability requirements that are suitable for tasks for task forces and scientific research departments, recommends capability requirements for military requirement scenarios, fills the technical gap of military requirement recommendation, expands the application scenarios of recommendation technology, and improves the traditional demand generation model.

[0135] The embodiment of the present invention decomposes an action plan into at least one task according to a pre-constructed capability requirement system; obtains input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and the reliability corresponding to the measurement scale; performs premise matching conversion on the input data according to a preset confidence rule base, and obtains a premise set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each rule includes at least a premise and a conclusion, and each element in the premise set is a tuple consisting of any one of the advance conditions matched by the input data and a first matching degree corresponding to the premise; performs rule matching according to the premise set, and calculates the activation degree and correction factor of each matched rule in combination with the confidence rule base; performs evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base, and obtains the capability requirement reasoning result of the at least one task, which can combine recommendation technology with military capability requirement scenarios, and can recommend capability requirements adapted to tasks for task forces and scientific research departments, and provide strong support for the execution of tasks, the growth of forces, and the development of military theories.

[0136] The above describes specific embodiments of the present invention. In some cases, the actions or steps recorded in the specification can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0137] Based on the same inventive concept, an embodiment of the present invention provides a system capability requirement recommendation device based on confidence rule base reasoning. Figure 7 As shown, the system capability requirement recommendation device based on the confidence rule base reasoning includes: an action decomposition unit, a data acquisition unit, a condition matching unit, a rule matching unit and an evidence reasoning unit. Among them,

[0138] Action decomposition unit, used to decompose an action plan into at least one task according to a pre-built capability requirement system;

[0139] A data acquisition unit, configured to acquire input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and a reliability corresponding to the measurement scale;

[0140] A condition matching unit, configured to perform a precondition matching conversion on the input data according to a preset confidence rule base, and obtain a precondition set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each of which includes at least a precondition and a conclusion, and each element in the precondition set is a tuple consisting of any precondition matched by the input data and a first matching degree corresponding to the precondition;

[0141] A rule matching unit, performing rule matching according to the set of prerequisites, and calculating the activation degree and correction factor of each matched rule in combination with the confidence rule base;

[0142] The evidence reasoning unit is used to perform evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of the at least one task.

[0143] For the convenience of description, the above device is described as various modules according to their functions. Of course, when implementing the embodiment of the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0144] The apparatus of the above-mentioned embodiment is used to implement the corresponding method in the above-mentioned embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0145] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of the above embodiments is implemented.

[0146] Figure 8 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 801, a memory 802, an input / output interface 803, a communication interface 804, and a bus 805. The processor 801, the memory 802, the input / output interface 803, and the communication interface 804 are connected to each other in communication within the device through the bus 805.

[0147] The processor 801 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present invention.

[0148] The memory 802 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 802 may store an operating system and other application programs. When the technical solution provided by the embodiment of the present invention is implemented by software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801.

[0149] The input / output interface 803 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0150] The communication interface 804 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0151] The bus 805 includes a path to transmit information between various components of the device (eg, the processor 801 , the memory 802 , the input / output interface 803 , and the communication interface 804 ).

[0152] It should be noted that, although the above device only shows the processor 801, the memory 802, the input / output interface 803, the communication interface 804 and the bus 805, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiment of the present invention, and does not necessarily include all the components shown in the figure.

[0153] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure is limited to these examples. In line with the concept of the present application, the above embodiments or technical features in different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0154] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present disclosure.

Claims

1. A system capability requirement recommendation method based on confidence rule base reasoning, characterized by: The method comprises: Break down an action plan into at least one task based on a pre-built capability requirement system; Acquiring input data of the at least one task, the input data comprising at least one measurement scale standard of the task and a reliability corresponding to the measurement scale; Performing a prerequisite matching conversion on the input data according to a preset confidence rule base, and obtaining a prerequisite set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each of which includes at least a prerequisite and a conclusion, and each element in the prerequisite set is a tuple consisting of any prerequisite matched by the input data and a first matching degree corresponding to the prerequisite; Perform rule matching according to the set of prerequisites, and calculate the activation degree and correction factor of each matched rule in combination with the confidence rule base; Performing evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain a capability requirement reasoning result of the at least one task; The conclusion corresponding to the rule in the confidence rule base includes capability requirements, at least one result, and corresponding result reliability, and performing evidential reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of the at least one task includes: Correcting each of the result credibility in the rule according to the correction factor; Determining capability requirements according to the matched rules based on the trust rule base; According to the activation degree of each of the matched rules and the modified result reliability, the capability requirement reliability corresponding to each of the capability requirements is calculated respectively to obtain the capability requirement reasoning result of the at least one task; The calculating the capability requirement reliability corresponding to each capability requirement respectively according to the activation degree of each matched rule and the modified result reliability comprises: Acquire at least one rule corresponding to each capability requirement from the matched rules; According to the activation degree of the rule corresponding to each capability requirement and the modified result reliability, the final result reliability of each result corresponding to each capability requirement is calculated by applying the following relationship as the corresponding capability requirement reliability: , in, To include the results The rule set For the result The final reliability.

2. The method according to claim 1, characterized in that: The performing a precondition matching conversion on the input data according to a preset trust rule library to obtain a precondition set matched by the input data includes: Calculate the similarity between any of the measurement scale standards in the input data and each value corresponding to the measurement scale in the task measurement scale value table according to the task measurement scale value table in the confidence rule base; Calculate the first matching degree between any one of the measurement scale standards and each of the preconditions according to the similarity degree and the reliability; All the preconditions matched by the measurement standards in the input data are collected to form the precondition set.

3. The method according to claim 2, characterized in that: The rules in the confidence rule base also include weights, and the rule matching is performed according to the precondition set, and the activation degree and correction factor of each matched rule are calculated in combination with the confidence rule base, including: Matching the precondition set with the preconditions of the rules in the trusted rule base to obtain at least one matching rule; For each of the matched rules, applying an aggregation function to aggregate the first matching degrees corresponding to the preconditions in the rule to obtain a second matching degree of the rule; Calculating the activation degree of each of the rules according to the second matching degree and the weight of each of the rules; The correction factor of each of the rules is calculated according to the first matching degree of each of the preconditions in each of the rules.

4. The method according to claim 3, characterized in that: The applying an aggregation function to aggregate the first matching degrees corresponding to the preconditions in the rule to obtain the second matching degree of the rule includes: Performing a logical probability operation on the first matching degree of each of the precondition pairs in the rule to obtain a second matching degree of the rule; or, The first matching degree of each of the precondition pairs in the rule is subjected to a probability operation of an OR logic to obtain a second matching degree of the rule.

5. The method according to claim 4, characterized in that: The method further comprises: When an action plan is decomposed into a plurality of tasks, the capability requirements in the plurality of tasks are aggregated to obtain the capability requirements of the action plan and the corresponding capability requirement reliability.

6. The method according to claim 5, characterized in that: The aggregating the capability requirements in the plurality of tasks to obtain the capability requirements of the action plan and the corresponding capability requirement reliability includes: Aggregating the capability requirements in a plurality of the tasks as the inferred capability requirements of the action plan; For the same capability requirements in a plurality of the tasks, the capability requirement with the highest final result reliability among the results with the highest priority is selected as the aggregated capability requirement.

7. A system capability requirement recommendation device based on confidence rule base reasoning according to any one of the methods of claims 1 to 6, characterized in that: The device comprises: Action decomposition unit, used to decompose an action plan into at least one task according to a pre-built capability requirement system; A data acquisition unit, configured to acquire input data of the at least one task, wherein the input data includes at least one measurement scale standard of the task and a reliability corresponding to the measurement scale; A condition matching unit, configured to perform a precondition matching conversion on the input data according to a preset confidence rule base, and obtain a precondition set matched by the input data, wherein the confidence rule base includes a task measurement scale value table and at least one rule, each of which includes at least a precondition and a conclusion, and each element in the precondition set is a tuple consisting of any precondition matched by the input data and a first matching degree corresponding to the precondition; A rule matching unit, performing rule matching according to the set of prerequisites, and calculating the activation degree and correction factor of each matched rule in combination with the confidence rule base; The evidence reasoning unit is used to perform evidence reasoning according to the activation degree and the correction factor of each matched rule in combination with the confidence rule base to obtain the capability requirement reasoning result of the at least one task.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.