User scene-oriented intelligent product service function module generation method and system

Abstract

The invention belongs to the technical field of intelligent design, and particularly relates to a user scene-oriented intelligent product service function module generation method and system, and the method comprises the steps: recognizing corresponding intelligent product service function elements according to scene experience demands; according to the intelligent product service function elements, identifying corresponding intelligent product service function components; evaluating the correlation between the intelligent product service function components; and determining an intelligent product service function module according to the correlation. By constructing the mapping relation between the scene experience requirements and the scene elements and the mapping relation between the intelligent product service function elements and the scene elements, the scene experience requirements are imported, and the intelligent product service function elements are exported. According to the method, the intelligent product service function components are identified by using the intelligent product service function elements, so that an intelligent product service function identification path oriented to a user scene can be structured, and a clear range is provided for intelligent product service function identification.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent design technology, and specifically relates to a method and system for generating intelligent product service function modules for user scenarios. Background Art

[0002] Traditional product feature design typically relies on manual effort to gather user scenarios, identify functional requirements, and match functional components. The evolving IoT landscape is driving demand for large-scale personalized service customization, necessitating the design of intelligent product and service modules. Using intelligent product and service modules as the fundamental building blocks for ecosystem configuration effectively reduces design redundancy and costs, improves design resource efficiency, simplifies the configuration process, and significantly increases the diversity of intelligent product and service offerings within the ecosystem, enriching the choices available to a wide range of users.

[0003] In related technologies, intelligent product service function modules include not only physical activities or resource components from the physical service space, but also cyber service activities or resource components that implement intelligent functions such as data monitoring, diagnosis, prediction, and guidance. These components are service activities or resources with independent functions, and are component packages constructed according to certain processes or relationships to support service functions with different attributes. For example, Chinese patent document No. CN116450102B discloses a module generation method and system for an intelligent product service ecosystem. The method includes identifying all service components, evaluating the co-intelligence relationship between service components, constructing a co-intelligence relationship evaluation matrix, and ultimately generating a service module.

[0004] However, the identification and module generation of the aforementioned intelligent product service functional components still primarily rely on the scenario service blueprint to construct a mapping relationship from intelligent product service functional elements to intelligent product service functional components. The resulting intelligent product service functional components fail to fully reflect the information of the scenario elements, which results in the generation of identified intelligent product service functional modules still having the limitations of traditional product functional design. Therefore, for the process of identifying and dividing intelligent product service functional components and modules for user scenarios, it is necessary to construct a structured logic to assist designers in more comprehensive and systematically identifying intelligent product service functions, thereby optimizing the accuracy and comprehensiveness of intelligent product service functional component identification. Summary of the Invention

[0005] Based on the above-mentioned shortcomings and deficiencies in the prior art, one of the purposes of the present invention is to provide a method for generating a customized intelligent product service function module for user scenarios that meets one or more of the aforementioned requirements.

[0006] In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical solutions:

[0007] A method for generating a user scenario-oriented intelligent product service function module includes the following steps:

[0008] Identify corresponding smart product service function element instances based on scenario experience requirements;

[0009] Based on the smart product service function element instances, identify the corresponding smart product service function components;

[0010] Evaluate the correlation between functional components of smart product services;

[0011] Determine the intelligent product service function modules based on relevance.

[0012] As a preferred solution, based on scenario experience requirements, examples of identifying corresponding smart product service functional elements include:

[0013] Construct the first mapping relationship between scene experience requirements and scene elements;

[0014] Constructing a second mapping relationship between scenario elements and smart product service function element types;

[0015] Constructing a third mapping relationship between smart product service function element types and smart product service function element instances;

[0016] Based on the first mapping relationship, the second mapping relationship and the third mapping relationship, an intelligent product service function element instance corresponding to the user experience requirement is identified.

[0017] As a preferred solution, the functional elements of smart product services include at least one of smart user perception, smart product perception, smart activity perception, smart environment perception, smart user connection, smart product connection, smart user analysis, smart product analysis, smart activity analysis, smart environment analysis, smart user support, smart product operation, smart activity execution and smart environment interaction.

[0018] As a preferred solution, the steps of identifying corresponding smart product service function components according to the smart product service function elements include:

[0019] Build a scenario service blueprint for user scenarios;

[0020] Import smart product service function element instances based on scenario service blueprint;

[0021] Identify corresponding smart product service function components based on smart product service function element instances.

[0022] As a preferred solution, the steps for building a scenario service blueprint for user scenarios include:

[0023] Determine the operating mechanism of intelligent product service functions for user scenarios based on service scope, service operation, and service resources;

[0024] Based on the operation mechanism of intelligent product service functions, generate an intelligent product service function operation model including user scenario domain, intelligent product operation domain, intelligent platform service operation domain, and service ecological resource support domain;

[0025] Run the model based on the service functions of smart products to generate a scenario service blueprint.

[0026] As a preferred solution, the steps of evaluating the dependencies between smart product components include:

[0027] Determine the correlation matrix between smart product components under multiple preset dimensions;

[0028] Aggregate all correlation matrices to obtain a comprehensive correlation matrix;

[0029] The comprehensive correlation matrix is ​​defuzzified to obtain the target comprehensive correlation matrix.

[0030] As a preferred solution, the step of defuzzifying the comprehensive correlation matrix to obtain the target comprehensive correlation matrix includes:

[0031] Get the preset cloud model scoring function;

[0032] Defuzzification is performed on the matrix elements of the comprehensive correlation matrix according to the cloud model scoring function to obtain a clear number matrix corresponding to each matrix element;

[0033] The target comprehensive correlation matrix is ​​obtained based on all explicit number matrices.

[0034] As a preferred solution, the steps of determining the intelligent product functional modules based on correlation include:

[0035] Construct a weighted complex network of intelligent product service functional components based on the target comprehensive correlation matrix;

[0036] According to the weighted complex network, the component aggregation degree is obtained based on the preset weighted complex community discovery algorithm;

[0037] Aggregation results for judging the degree of component aggregation;

[0038] Determine the intelligent product service function module based on the aggregation results.

[0039] As the preferred solution, the preset weighted complex community discovery algorithms include Louvain algorithm, weighted Girvan-Newman algorithm, and Fast-Newman algorithm; the aggregation degree is judged by modularity, module balance, and module complexity.

[0040] The present invention also provides a user scenario-oriented intelligent product service function module generation system, including:

[0041] The first identification unit is used to identify corresponding smart product service function elements according to scene experience requirements;

[0042] A second identification unit is used to identify corresponding smart product service function components according to the smart product service function elements;

[0043] A correlation evaluation unit, used to evaluate the correlation between functional components of smart product services;

[0044] A module determination / obtaining / generating unit, configured to determine a smart product service function module based on correlation;

[0045] Among them, based on the scenario experience requirements, the corresponding examples of smart product service function elements include:

[0046] Constructing a first mapping relationship between the scene experience requirements and scene elements;

[0047] Constructing a second mapping relationship between the scenario elements and the types of smart product service function elements;

[0048] Constructing a third mapping relationship between the smart product service function element type and the smart product service function element instance;

[0049] The smart product service function element instance corresponding to the user experience requirement is identified based on the first mapping relationship, the second mapping relationship, and the third mapping relationship.

[0050] Compared to existing technologies, this invention, in identifying intelligent product and service functions for user scenarios, takes an intelligent perspective on scenario experience requirements. Based on these requirements, it identifies instances of intelligent product and service function elements, including both scenario elements and service function element information. It then constructs corresponding scenario service blueprints around these instances to identify intelligent product and service function components. This proposed method for generating intelligent product and service function modules for user scenarios eliminates the need for logical deduction based on human experience and knowledge, providing a structured logic to assist designers in more comprehensive and systematic identification of intelligent product and service functions. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flowchart of a method for generating a user scenario-oriented intelligent product service function module according to an embodiment of the present invention;

[0052] Figure 2 This is a schematic diagram of deriving an example of a user scenario-oriented smart product service function element according to an embodiment of the present invention;

[0053] Figure 3 A scenario service blueprint for an embodiment of the present invention;

[0054] Figure 4 An example of a component dependency weighted complex network according to an embodiment of the present invention;

[0055] Figure 5 is a diagram of a component aggregation method based on a weighted complex network community discovery algorithm according to an embodiment of the present invention;

[0056] Figure 6 It is a sleep scene service blueprint of an embodiment of the present invention;

[0057] Figure 7 This is a weighted complex network diagram of the correlation between functional components of the sleep scenario smart product service according to an embodiment of the present invention;

[0058] Figure 8 This is a diagram showing changes in modularity during the process of aggregating functional components of smart product services for sleep scenarios using a weighted GN algorithm according to an embodiment of the present invention;

[0059] Figure 9 This is a diagram of modularity changes during the process of aggregating functional components of smart product services for sleep scenarios using the FN algorithm in an embodiment of the present invention;

[0060] Figure 10 This is a complex network diagram of the correlation between functional components of the sleep smart product service after aggregation using the weighted GN algorithm of an embodiment of the present invention;

[0061] Figure 11 This is a component aggregation modularity change diagram using the unweighted GN algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0063] Example 1

[0064] In existing technologies, the need for large-scale personalized service customization in the evolving IoT environment has led to the design of intelligent product service function modules. However, current methods for generating intelligent product service function modules still rely on human experience and logical deduction to identify intelligent product service function components and generate intelligent product service function modules, lacking a structured path to generate reliable intelligent product service function modules.

[0065] To address this issue, this embodiment provides a method for generating intelligent product service function modules based on user scenarios. Figure 1 As shown, it includes the following steps:

[0066] S1. Identify corresponding smart product service function element instances based on scenario experience requirements;

[0067] S2. Identify corresponding smart product service function components based on the smart product service function element instance;

[0068] S3. Evaluate the correlation between the functional components of the smart product service;

[0069] S4. Determine the smart product service function module based on the correlation.

[0070] Among them, based on the scenario experience requirements, the corresponding examples of smart product service function elements include:

[0071] Construct the first mapping relationship between scene experience requirements and scene elements;

[0072] Constructing a second mapping relationship between scenario elements and smart product service function element types;

[0073] Constructing a third mapping relationship between smart product service function element types and smart product service function element instances;

[0074] Based on the first mapping relationship, the second mapping relationship and the third mapping relationship, an instance of a smart product service function element corresponding to the user experience requirement is identified.

[0075] This embodiment starts from the intelligent perspective of scene experience needs, identifies corresponding smart product service function element instances based on scene experience needs, and establishes a mapping relationship from smart product service function element instances to smart product service function components based on the scene service blueprint, thereby providing a structured path for identifying smart product service function components for user scenarios.

[0076] In a possible implementation, the scenario elements include at least one of user scenario elements, product scenario elements, activity scenario elements, and environmental scenario elements.

[0077] Specifically, scenario elements are obtained by investigating user needs and extracting corresponding scenario elements.

[0078] Among them, the types of functional elements of intelligent product services cover four major categories of intelligent capabilities: intelligent perception, intelligent connection, intelligent analysis and intelligent delivery, specifically including intelligent user perception, intelligent product perception, intelligent activity perception, intelligent environment perception, intelligent user connection, intelligent product connection, intelligent user analysis, intelligent product analysis, intelligent activity analysis, intelligent environment analysis, intelligent user support, intelligent product operation, intelligent activity execution and intelligent environment interaction.

[0079] Specifically, intelligent perception capabilities include intelligent user perception, intelligent product perception, intelligent activity perception, and intelligent environment perception. Intelligent perception capabilities involve the perception and collection of data information. Intelligent perception capabilities can only be matched when scene elements possess information attributes. This means that intelligent perception capabilities can only be embedded in scene elements that can generate information and be perceived and collected. For example, users possess behavioral information and products possess operational information.

[0080] Specifically, smart connection capabilities include smart user connection and smart product connection. Smart connection capabilities involve connecting scene elements and exchanging information. Only when scene elements have connection interfaces can they match smart connection capabilities, such as product-to-product connection.

[0081] Specifically, intelligent analysis capabilities include intelligent user analysis, intelligent product analysis, intelligent activity analysis, and intelligent environmental analysis. Intelligent analysis involves the ability to mine information and knowledge from data using methods such as big data analysis and machine learning. These capabilities primarily encompass statistical analysis, diagnostic analysis, predictive analysis, and optimization analysis. This capability is closely tied to data information acquired through intelligent perception. Intelligent analysis primarily relies on the knowledge attributes of scene objects, mining patterns and features to provide decision-making insights, such as predicting user behavior, diagnosing product failures, and optimizing activity resources.

[0082] Specifically, intelligent interaction capabilities include intelligent user support, intelligent product operation, intelligent activity execution and intelligent environment interaction. Intelligent interaction capabilities involve the ability to map to the interactive attributes of scene elements, such as product remote control and autonomous operation.

[0083] In order to realize the structured path from scenario experience requirements to intelligent product service functional elements, it is necessary to build a mapping relationship from scenario experience requirements to scenario elements, a mapping relationship from scenario elements to intelligent product service functional element types, and a mapping relationship from intelligent product service functional element types to intelligent product service functional element instances. Subsequently, by importing scenario experience requirements, the corresponding intelligent product service functional elements can be identified and functional element instances can be derived, such as Figure 2 shown.

[0084] By introducing scene elements and smart product service function element types into the mapping relationship of scene experience requirements, the derived smart product service function element instances contain information of both scene elements and service function elements; using a structured recognition path instead of the recognition method of human experience, the generation of smart product service function element instances is freed from the logical deduction process of human experience knowledge, providing reliable information input for the subsequent identification of smart product service function components.

[0085] In a possible implementation, step S2 includes:

[0086] Build scenario service blueprints for user scenarios;

[0087] Import smart product service function element instances based on scenario service blueprint;

[0088] Identify corresponding smart product service function components based on smart product service function element instances.

[0089] Since the intelligent product service functions for user scenarios are built with the support of cyber-physical interactive operation activities and ecological resources, the components involved are more complex and diverse, and more difficult to identify. Therefore, as the basis for identifying the intelligent product service function components for user scenarios, it is necessary to build a scenario service blueprint for the intelligent product service function components for user scenarios. With the help of the scenario service blueprint, designers can comprehensively and intuitively display the service operation activities and resource components involved in the intelligent product service functions for user scenarios, such as Figure 3 shown.

[0090] Specifically, the scenario service blueprint includes the user scenario domain, the smart product operation domain, the platform service operation domain, and the ecological resource support domain.

[0091] The user scenario domain includes three elements: scenario goals, scenario activity journeys, and user scenario-oriented intelligent product service functions that support the activity journeys. The user activity journey includes early, mid-term, and late-stage activities. Intelligent product service functions for user scenarios are various intelligent service functions for scenario elements identified to meet user needs, including user intelligent service functions, product intelligent service functions, activity intelligent service functions, and environmental intelligent service functions. Since the intelligent product service function element instance input into the scenario service blueprint in this embodiment contains information about scenario elements, the scenario service blueprint also includes a user scenario domain to reflect the impact of scenario elements on component identification.

[0092] Among them, the smart product operation domain includes six types of activity components, including smart product network connection configuration, real-time data collection, real-time data transmission, smart sensor operation, smart actuator operation, and smart human-computer interaction interface operation. The first three components are basic activity components, and the last three components are further determined based on specific functional utilities, such as smart light sensor operation, smart air conditioning controller operation, and smart voice interaction interface operation.

[0093] The platform service operation domain encompasses two main categories of activity components: data operation components and service operation components. Data operation components primarily include four types of activity components: data storage, data processing, data analysis, and data services. Data storage and data processing activities involve the storage and processing of data related to scenario elements, enabling subsequent integration and analysis of related data. Data analysis can be further categorized into different types of analytical activities, including statistical analysis, predictive analysis, diagnostic analysis, and optimization analysis. Data services primarily include, but are not limited to, visualization services and data query. The service operation component primarily includes four activity components: service operation management, service application operation, service resource scheduling, and service task execution. Service operation management is the fundamental component for managing service operations, including service security management, service interface management, and service quality management. Service application operation is the application software operation activity related to user-facing intelligent product service functions within the platform's microservices architecture. Service resource scheduling involves the connection and use of service application operation resources, including platform system data access, external data access, expertise acquisition, invocation of specialized algorithm models, and connection to third-party platforms. Service task execution refers to activities related to service delivery and interaction, including content delivery and command delivery, such as delivering driving navigation route content and product control commands.

[0094] The ecological resource support domain encompasses resource components that support the operation of smart products and platform services, primarily categorized into four categories: data and knowledge resources, algorithmic model resources, professional and technical resources, and third-party service system / platform resource components. The resources required to build smart product and service functions for different user scenarios vary, and these resources must be determined based on the specific smart product and service functions being used.

[0095] The existing service blueprint tools for smart product services, because the input smart product service function element types do not directly reflect the scenario element information, such service blueprints can only display the components of the product service functions in a certain type of specific scenario corresponding to the input smart product service function element types. In contrast, because the smart product service function element instances of the embodiments of the present invention already contain scenario element information, the scenario service blueprint proposed in the embodiments of the present invention is able to display all the smart product service function components contained in the entire scenario rather than the components of the smart product service functions in a certain type of specific scenario. At the same time, taking into account the characteristics of cyber-physical interactive operation and ecological resources, the identified components are further aggregated according to the correlation between the components to form a smart product service function module for the user scenario. The smart product service function module thus identified has more advantages in terms of interaction and integration between components than the smart product service function modules identified by service blueprint tools for specific scenarios.

[0096] In a possible implementation, step S3 includes:

[0097] Determine the correlation matrix between smart product components under multiple preset dimensions;

[0098] Aggregate all correlation matrices to obtain a comprehensive correlation matrix;

[0099] The comprehensive correlation matrix is ​​defuzzified to obtain the target comprehensive correlation matrix.

[0100] Specifically, the normal cloud model and the multivariate fuzzy correlation matrix are used to determine the correlation between the functional components of smart product services in multiple preset dimensions.

[0101] Specifically, the correlations under the multiple dimensions include intelligent capability correlation, service process correlation, and service resource correlation.

[0102] Among them, since specific components are aggregated to form different intelligent capabilities, such as intelligent perception capability, intelligent connection capability, intelligent analysis capability and intelligent delivery capability, multiple intelligent capabilities form an intelligent product service function with relatively independent efficacy, so the correlation matrix dimension includes intelligent capability correlation.

[0103] Because components within and outside the cyber and physical operational domains interact through input and output of data, information, knowledge, and wisdom, these interactions form the operational processes of intelligent product service functions for user scenarios. Therefore, the relevance matrix dimension includes a service process relevance matrix.

[0104] Among them, since service resources are the operating basis of intelligent product service functions for user scenarios, some operating components may share one or more service resource components, so the correlation matrix dimension includes service resource correlation.

[0105] Specifically, the evaluation rules for intelligent capability relevance, service process relevance, and service resource relevance are as follows: Table 1-Table 3:

[0106] grade Description of intelligence ability correlation Very strong (VS) <![CDATA[Component C i and Component C j together build the same intelligent ability and are interdependent]]> Strong (S) <![CDATA[Component C i and Component C j together build the same intelligent ability without dependency]]> Medium (M) <![CDATA[Component C i and Component C j build different intelligent capabilities but have certain dependencies]]> Weak(W) <![CDATA[Component C i and Component C j build different intelligent capabilities and have a loose relationship]]> No <![CDATA[Component C i and Component C j build different intelligent capabilities and have nothing to do with each other]]>

[0107] Table 1 Evaluation of correlation between intelligence ability

[0108]

[0109] Table 2 Evaluation of service process relevance

[0110] grade Description of intelligence ability correlation Very strong (VS) <![CDATA[Component C i and Component C j require multiple service resources simultaneously and have a strong sharing relationship]]> Strong (S) <![CDATA[Component C i and Component C j At the same time, multiple service resources are required and there is a strong sharing relationship]]> Medium (M) <![CDATA[Component C i and Component C j simultaneously require a certain service resource and have a certain sharing relationship]]> Weak(W) <![CDATA[Component C i and Component C j simultaneously require a certain service resource but have a weak sharing relationship]]> No <![CDATA[Component C i and Component C j There is no sharing relationship of any service resources between them]]>

[0111] Table 3 Service resource relevance evaluation

[0112] In one possible implementation, based on the intelligent capability correlation, service process correlation and service resource correlation evaluation rules, and using the normal cloud model and multivariate fuzzy correlation matrix calculation method, the corresponding intelligent capability correlation matrix, service process correlation matrix and service resource matrix are obtained.

[0113] Specifically, the calculation method of the intelligence capability correlation matrix is ​​as follows:

[0114] For n intelligent product service function components FC={FC1,FC2,...,FC n}, the semantic scale in Table 2 is used to construct the intelligence capability correlation strength evaluation matrix between each component as follows:

[0115]

[0116] in, is the semantic relevance of the intelligent capabilities of the i-th component and the j-th component.

[0117] The semantic variables are quantified into normal clouds according to the rules in Table 4.

[0118] grade Normal Cloud Very strong (VS) (100，10.31，0.26) Strong (S) (69.1，6.37，0.16) Medium (M) (50，3.93，0.1) Weak(W) (30.9，6.37，0.16) No (0，0，0)

[0119] Table 4 Semantic-normal cloud transformation rules

[0120] Finally, the correlation matrix of the intelligent capability cloud is obtained as follows:

[0121]

[0122] Specifically, the service process correlation matrix is ​​calculated as follows:

[0123] The semantic scale in Table 4 is used to construct the service process correlation strength evaluation matrix between components as follows:

[0124]

[0125] in, is the semantic relevance of the service process between the i-th component and the j-th component.

[0126] The computing service process cloud dependency matrix is ​​as follows:

[0127]

[0128] Specifically, the service resource dependency matrix is ​​calculated as follows:

[0129] The semantic scale in Table 4 is used to construct the service resource correlation strength evaluation matrix between components as follows:

[0130]

[0131] The computing service resource cloud dependency matrix is ​​as follows:

[0132]

[0133] In one possible implementation, all correlation matrices are aggregated to obtain a comprehensive correlation matrix, including aggregating the cloud correlation matrices of the intelligent capability correlation matrix, the service process correlation matrix, and the service resource matrix through a cloud model weighting operator to form a cloud comprehensive correlation matrix as follows:

[0134]

[0135] The calculation method of the cloud comprehensive correlation matrix elements is as follows:

[0136]

[0137] Where k = 1, 2, 3, represents three relevance evaluation dimensions. k is the correlation weight of each dimension.

[0138] In a possible implementation, the step of defuzzifying the comprehensive correlation matrix to obtain a target comprehensive correlation matrix includes:

[0139] Get the preset cloud model scoring function;

[0140] Defuzzification is performed on the matrix elements of the comprehensive correlation matrix according to the cloud model scoring function to obtain a clear number matrix corresponding to each matrix element;

[0141] The target comprehensive correlation matrix is ​​obtained based on all explicit number matrices.

[0142] The calculation method of the cloud model score function is as follows:

[0143]

[0144] in, is the standard deviation of the cloud model. γ∈[0,1] is the standard deviation coefficient, which is generally determined based on the actual problem.

[0145] The relevance evaluation of intelligent product service functional components for user scenarios is a decision-making process based on expert experience and knowledge. The embodiment of the present invention constructs a relevance evaluation dimension for intelligent product service functional components for user scenarios based on functions, processes, and resources. Taking into account the new features of intelligent capabilities, cyberphysical interactive operation, and ecological resources, the relevance of various aspects of intelligent product service functional components for user scenarios has different connotations from product service components in a single scenario. By reasonably setting the evaluation criteria for the relevance between components, it is possible to accurately identify highly correlated components and aggregate them together to form functional modules with higher stability.

[0146] In a possible implementation, step S4 includes: the step of determining the smart product functional module according to the correlation includes:

[0147] Constructing a weighted complex network of intelligent product service functional components based on the target comprehensive correlation matrix;

[0148] According to the weighted complex network, a component aggregation degree is obtained based on a preset weighted complex community discovery algorithm;

[0149] an aggregation result for judging the aggregation degree of the components;

[0150] Based on the constructed component correlation weighted complex network, this embodiment adopts the weighted complex network community discovery method to aggregate components and create intelligent product service function modules for user scenarios. Then, the aggregation results are evaluated from three dimensions: modularity, module balance, and module complexity. The best component aggregation scheme is selected to ensure module independence and comprehensibility. Figure 5 shown.

[0151] Among them, the complex network model is G = (V, E, W), where

[0152] V={SC1,SC2,...,SC n} represents the set of identified smart product service function components;

[0153] E={e 12 ,...,e1i ,...,e ij ,...,e n(n-1)} represents the connection between two components;

[0154] W={W 12 ,...,W 1i ,...,W ij ,...,W n(n-1)} represents the strength of the component connection.

[0155] Based on the de-targeted comprehensive correlation matrix, a weighted complex network of intelligent product service functional components for user scenarios is constructed, such as Figure 4 shown.

[0156] By constructing a complex network model, we can intuitively express the relational data of complex systems and reveal cluster characteristics. It is an effective way to describe the complex and uncertain correlations between components.

[0157] In a possible implementation, the preset weighted complex community discovery algorithm includes the Louvain algorithm, the weighted Girvan-Newman algorithm, and the Fast-Newman algorithm.

[0158] The Louvain algorithm is a community discovery algorithm based on modularity optimization. Its basic idea is to first treat each node in the network as an independent community. Then, it attempts to assign the node to all neighboring communities, calculates the change in modularity ΔQ before and after assignment, and selects the neighboring node with the highest modularity to merge and form a new community. These steps are repeated until the community remains unchanged. The steps for community discovery in weighted complex networks based on the Louvain algorithm are as follows:

[0159] Step 1: Initialize each component node as a minimum community;

[0160] Step 2 assigns each component node to a neighboring node community;

[0161] Step 3: Calculate the change in modularity before and after allocation;

[0162] Step 4: merge the two communities with the largest modularity changes;

[0163] Step 5: Repeat steps 2, 3, and 4 until all communities are stable.

[0164] Among them, the basic idea of ​​the Girvan-Newman (GN) algorithm is to remove network edges based on edge betweenness values ​​to discover communities. The basic process is: first calculate the edge betweenness values ​​of all edges in the network, then try to assign nodes to all neighboring communities, calculate the modularity change ΔQ before and after the assignment, remove the edge with the largest boundary degree, update the network and recalculate the edge betweenness of the network, and repeat the above process until the predetermined number of community divisions is reached. The corresponding modularity is obtained by setting the number of community divisions, and finally the division scheme with the largest modularity is selected. However, the classic GN algorithm only considers the calculation of unweighted edge betweenness. In order to adapt to the problem of weighted complex networks, the embodiment of the present invention uses the edge betweenness weighted by the comprehensive correlation of components as the basis for GN community discovery. The steps for complex network community discovery based on the weighted GN algorithm are as follows:

[0165] Step 1: Assign distances and weights to component network nodes;

[0166] Step 2: Calculate the weighted edge betweenness between component nodes;

[0167] Step 3: Remove the component edge with the maximum weighted edge betweenness;

[0168] Step 4 calculates the edge betweenness of the remaining component edges that have not been removed;

[0169] Step 5: Calculate the modularity and repeat the above steps;

[0170] Step 6 stops when the set number of modules is reached.

[0171] The Fast-Newman (FN) algorithm is a hierarchical clustering algorithm for discovering communities in complex networks based on maximizing incremental modularity. Its basic concept is to first treat each node in the network as an independent community, then merge neighboring communities, merging communities with the largest modularity change ΔQ until all nodes are in the same community. Finally, the community partitioning scheme with the highest modularity during the merging process is found. The FN algorithm only considers nodes with connected edges, reducing computational complexity. The steps for discovering communities in complex networks based on the FN algorithm are as follows:

[0172] Step 1: Initialize the component nodes as the minimum community;

[0173] Step 2 assigns the community to the adjacent node community;

[0174] Step 3: Calculate the change in modularity before and after allocation;

[0175] Step 4: merge the communities with the largest modularity increment (or smallest modularity decrement);

[0176] Step 5: Repeat steps 2, 3, and 4 until a large community is formed.

[0177] Step 6: Find the community division result with the largest modularity during the merging process.

[0178] There are many community discovery algorithms. This embodiment of the present invention selects three algorithms suitable for weighted complex networks: the Louvain algorithm, the weighted Girvan-Newman algorithm, and the Fast-Newman algorithm. All three methods use modularity to discover communities in complex networks and are suitable for weighted complex networks. Furthermore, their community discovery mechanisms differ significantly, allowing them to complement each other and provide more reliable results.

[0179] In one embodiment, the component aggregation degree uses modularity, module balance, and module complexity to judge the aggregation results.

[0180] Modularity (Q) is a standard metric for evaluating the quality of complex network community partitioning. The modularity value range is [-1, 1]. A higher modularity indicates a better community partitioning result, satisfying the "high cohesion, loose coupling" module generation principle. The modularity calculation formula is as follows:

[0181]

[0182] Among them, A i，j is the actual value of the two nodes on both sides, m is the number of edges, k i and k j are the degrees of nodes i and j respectively, c i and c j are the communities where nodes i and j are located respectively. If nodes i and j belong to the same community, c i =c j ,δ(c i , c j )=1; if nodes i and j do not belong to the same community, then δ(c i , c j )=0.

[0183] Module balance is used to measure the size of generated modules or the evenness of the distribution of module functions. By introducing module balance, the system can avoid the emergence of "giant modules" or "fragmented modules". The calculation formula for module balance is as follows:

[0184]

[0185] Among them, S i is the size of the i-th module, specifically expressed as the number of lines of code or function points. is the average size of a module.

[0186] Module complexity is used to measure the complexity of the structure within a single module. Excessive complexity will lead to difficulties in module maintenance and increased error rates. The calculation formula for module complexity is as follows:

[0187]

[0188] Among them, G is the total number of operators, such as assignment and conditional judgment, and V is the total number of operands, such as variables and constants.

[0189] In one embodiment, the polymerization degree evaluation rules are as follows:

[0190]

[0191] Table 5. Evaluation rules for component aggregation results

[0192] Example 2

[0193] In one embodiment, a user scenario-oriented intelligent product service function module generation system is also provided, including:

[0194] The first identification unit is used to identify corresponding smart product service function element instances based on scenario experience requirements;

[0195] A second identification unit is configured to identify a corresponding smart product service function component according to the smart product service function element instance;

[0196] A correlation evaluation unit, configured to evaluate the correlation between the functional components of the intelligent product service;

[0197] a module determination unit, configured to determine a smart product service function module based on the correlation;

[0198] Among them, based on the scenario experience requirements, the corresponding examples of smart product service function elements include:

[0199] Constructing a first mapping relationship between the scene experience requirements and scene elements;

[0200] Constructing a second mapping relationship between the scenario elements and the types of smart product service function elements;

[0201] Constructing a third mapping relationship between the smart product service function element type and the smart product service function element instance;

[0202] The smart product service function element instance corresponding to the user experience requirement is identified based on the first mapping relationship, the second mapping relationship, and the third mapping relationship.

[0203] Through the above system, a structured path can be provided for the identification of intelligent product service functions for user scenarios, breaking away from the logical deduction process of human experience and knowledge, and assisting designers to identify intelligent product service functions more comprehensively and systematically.

[0204] Example 3

[0205] Taking the construction of a smart product service function module for a sleep scenario as an example, the method for generating a smart product service function module for a user scenario in the present invention includes:

[0206] S1. Identify corresponding smart product service function element instances based on scenario experience requirements;

[0207] S2. Identify corresponding smart product service function components based on the smart product service function element instance;

[0208] S3. Evaluate the correlation between the functional components of the smart product service;

[0209] S4. Determine the smart product service function module based on the correlation;

[0210] Among them, based on the scenario experience requirements, the corresponding examples of smart product service function elements include:

[0211] Constructing a first mapping relationship between the scene experience requirements and scene elements;

[0212] Constructing a second mapping relationship between the scenario elements and the types of smart product service function elements;

[0213] Constructing a third mapping relationship between the smart product service function element type and the smart product service function element instance;

[0214] The smart product service function element instance corresponding to the user experience requirement is identified based on the first mapping relationship, the second mapping relationship, and the third mapping relationship.

[0215] Next, the embodiments of the present invention are described in detail.

[0216] In step S1, to identify instances of smart product service functional elements for sleep scenarios, it is necessary to construct a multi-level mapping relationship from sleep scenario experience requirements to corresponding smart product service functional element instances. To this end, the embodiment of the present invention invited two smart sleep product service designers and two market researchers to form a decision-making team. They summarized and sorted out the scenario elements related to sleep scenarios and the corresponding smart product service functional element types and their mapping relationships. Then, they imported the scenario experience requirements and generated 32 smart product service functional element instances based on the mapping relationships, which included information on both scenario elements and service functional elements. The results are shown in Table 6.

[0217]

[0218] Table 6 Examples of functional elements of smart product services for sleep scenarios

[0219] In step S2, based on the derived sleep scene smart product service function element instance, this embodiment further carries out the identification of the sleep scene smart product service function components. To this end, this embodiment invites a cyber service platform technical director, a cyber service platform business manager, a smart home product service designer and a smart product IoT senior engineer from Enterprise A to jointly participate in the construction of a scene service blueprint for identifying smart product service function components. Combined with the actual operation process and existing ecological resources of the cyber service platform built by Enterprise A, the scene service blueprint proposed by the present invention is used to identify and visualize the sleep scene smart product service function components. The identification results are shown in Table 7 and Figure 6 As shown in the figure, a total of 70 components were identified, including 11 components in the smart product operation domain, 41 components in the platform service operation domain, and 18 components in the ecological resource support domain.

[0220]

[0221]

[0222]

[0223]

[0224] Table 7 Functional components of smart product services for sleep scenarios

[0225] In step S3, the relevance of the functional components of the intelligent sleep scenario product services was evaluated. This step further invited the decision-making team to evaluate the relevance of the 70 functional components of the intelligent sleep scenario product services listed in Table 7. The intelligent capability semantic relevance matrix, service process semantic relevance matrix, and service resource semantic relevance matrix were constructed in sequence. These matrices were then converted into an intelligent capability cloud relevance matrix, a service process cloud relevance matrix, and a service resource cloud relevance matrix. Finally, a cloud-based comprehensive relevance matrix and a target comprehensive relevance matrix for the functional components of the intelligent sleep scenario product services were obtained, as shown in Table 8.

[0226]

[0227] Table 8 Cloud comprehensive correlation matrix

[0228]

[0229] Table 8 Target comprehensive correlation matrix

[0230] According to the target comprehensive correlation matrix, an undirected weighted complex network is constructed in step S3, such as Figure 7 As shown. Step S2 aggregates the functional components of the sleep scene intelligent product service, and uses the weighted complex network community discovery algorithm comparison framework proposed by the present invention to aggregate the functional components of the sleep scene intelligent product service to form a more stable sleep scene intelligent product service function. First, Louvain algorithm, weighted GN algorithm and FN algorithm are used to Figure 7 The undirected weighted complex network shown is used for community module detection and the modularity of the module scheme is calculated.

[0231] Aggregation of functional components of intelligent product services for sleep scenarios based on the Louvain algorithm The optimal intelligent product service functional modules for sleep scenarios obtained using the Louvain algorithm are 13, with a corresponding maximum modularity of 0.3548. The components in each aggregation module are shown in Table 9.

[0232]

[0233] Table 9 Component aggregation results based on Louvain algorithm

[0234] Aggregation of sleep scene intelligent product service function components based on weighted GN algorithm The modularity changes in the process of aggregating sleep scene intelligent product service function components using weighted GN algorithm are as follows: Figure 8 The optimal number of modules is 11, corresponding to a maximum modularity of 0.4970. The components in each aggregation module are shown in Table 10.

[0235]

[0236] Table 10 Component aggregation results based on weighted GN algorithm

[0237] Aggregation of sleep scene intelligent product service function components based on FN algorithm In the process of aggregating sleep scene intelligent product service function components using FN algorithm, the modularity changes as follows Figure 9 The optimal number of modules is 6, corresponding to a maximum modularity of 0.3683. The components in each aggregation module are shown in Table 11.

[0238]

[0239] Table 11 Component aggregation results based on FN algorithm

[0240] In step S4, this embodiment compares the aggregation results of the sleep scene intelligent product service function components obtained by the three algorithms from three dimensions, as shown in Table 12. It can be found that the optimal component aggregation schemes obtained by the three algorithms are significantly different. From the modularity dimension, compared with the other two algorithms, the modularity of the optimal aggregation scheme obtained by the weighted GN algorithm is the largest, indicating that the "high cohesion, low coupling" characteristics of the scheme are the best. From the module balance comparison, the number of internal components of the module obtained by the FN algorithm is relatively balanced, but it is also due to the small number of scheme divisions, resulting in high complexity of each module. Due to the high degree of heterogeneity of scene-oriented intelligent product service function components, the small number of component modules makes the modular configurability of subsequent schemes not high. In other words, changing a component will affect the availability of the entire module. Based on the above analysis, the decision-making team believes that the component aggregation results obtained by the weighted GN algorithm are the most reasonable, and have high understandability and feasibility. Finally, the present invention visualizes the complex network of correlations of the sleep intelligent product service function components after aggregation by the weighted GN algorithm, as shown in the figure. Figure 10 shown.

[0241] Louvain Weighted GN FN Modularity 0.3548 0.4970 0.3683 Module balance generally generally high Module complexity Low Low high

[0242] Table 12 Comparison of component aggregation results of different weighted complex network community detection algorithms

[0243] In addition, the present invention also compares and analyzes the component aggregation results of the weighted GN algorithm and the unweighted GN algorithm to reveal the influence of the correlation strength on the component aggregation results. Figure 11As shown in the figure, the optimal number of component aggregation modules obtained by the unweighted GN algorithm is 16, corresponding to a modularity of 0.3239. The components contained in each module are shown in Table 13. Compared with the weighted GN algorithm, the unweighted GN algorithm obtains a smaller maximum modularity and produces less reasonable component aggregation results. For example, there is a strong correlation between components C49 and C67. The Louvain algorithm, FN algorithm, and weighted GN algorithm all place these two components in the same module, while the unweighted GN algorithm places them in different modules. It can be seen that the strength of component correlation has a significant impact on component aggregation. Therefore, compared with the unweighted complex network community discovery algorithm, the weighted complex network community discovery algorithm is more suitable for the problem of aggregating functional components of intelligent product services for user scenarios.

[0244]

[0245] Table 13 Component aggregation results based on the unweighted GN algorithm

[0246] The above description is only a detailed description of the preferred embodiments and principles of the present invention. For ordinary technicians in this field, based on the ideas provided by the present invention, there may be changes in the specific implementation methods, and these changes should also be considered as the scope of protection of the present invention.

Claims

1. A method for generating intelligent product service function modules for user scenarios, characterized in that: The method comprises the following steps: Identify corresponding smart product service function element instances based on scenario experience requirements; Identify corresponding smart product service function components based on the smart product service function element instance; Evaluate the correlation between the functional components of the smart product service; Determine the intelligent product service function module according to the correlation; The example of identifying corresponding smart product service function elements based on scenario experience requirements includes: Constructing a first mapping relationship between the scene experience requirements and scene elements; Constructing a second mapping relationship between the scenario elements and the types of smart product service function elements; Constructing a third mapping relationship between the smart product service function element type and the smart product service function element instance; The smart product service function element instance corresponding to the user experience requirement is identified based on the first mapping relationship, the second mapping relationship, and the third mapping relationship.

2. The method according to claim 1, characterized in that The scenario elements include at least one of user scenario elements, product scenario elements, activity scenario elements, and environmental scenario elements.

3. The method according to claim 1, characterized in that The types of smart product service function elements include at least one of smart user perception, smart product perception, smart activity perception, smart environment perception, smart user connection, smart product connection, smart user analysis, smart product analysis, smart activity analysis, smart environment analysis, smart user support, smart product operation, smart activity execution and smart environment interaction.

4. The method according to claim 1, wherein The step of identifying the corresponding smart product service function component according to the smart product service function element instance includes: Build scenario service blueprints for user scenarios; Importing the smart product service function element instance according to the scenario service blueprint; Identify the corresponding smart product service function component according to the smart product service function element instance.

5. The method according to claim 4, characterized in that The steps of constructing a scenario service blueprint for user scenarios include: Determine the operating mechanism of intelligent product service functions for user scenarios based on service scope, service operation, and service resources; Generate an intelligent product service function operation model including a user scenario domain, an intelligent product operation domain, an intelligent platform service operation domain, and a service ecological resource support domain based on the intelligent product service function operation mechanism; The scenario service blueprint is generated based on the smart product service function operation model.

6. The method according to claim 1, characterized in that The step of evaluating the correlation between smart product components includes: Determining a correlation matrix between components of the smart product under multiple preset dimensions; Aggregating all the correlation matrices to obtain a comprehensive correlation matrix; The comprehensive correlation matrix is ​​defuzzified to obtain a target comprehensive correlation matrix.

7. The method according to claim 6, characterized in that The step of defuzzifying the comprehensive correlation matrix to obtain a target comprehensive correlation matrix comprises: Get the preset cloud model scoring function; Defuzzifying the matrix elements of the comprehensive correlation matrix according to the cloud model scoring function to obtain a clear number matrix corresponding to each matrix element; A target comprehensive correlation matrix is ​​obtained based on all the explicit number matrices.

8. The method according to claim 1, characterized in that The step of determining the intelligent product functional module according to the correlation comprises: Constructing a weighted complex network of intelligent product service functional components based on the target comprehensive correlation matrix; According to the weighted complex network, a component aggregation degree is obtained based on a preset weighted complex community discovery algorithm; an aggregation result for judging the aggregation degree of the components; Determine the intelligent product service function module based on the aggregation result.

9. The method according to claim 8, characterized in that The preset weighted complex community discovery algorithms include the Louvain algorithm, the weighted Girvan-Newman algorithm, and the Fast-Newman algorithm; the aggregation degree evaluation uses modularity, module balance, and module complexity to evaluate the aggregation results.

10. A user scenario-oriented intelligent product service function module generation system, comprising: The first identification unit is used to identify corresponding smart product service function element instances based on scenario experience requirements; A second identification unit is configured to identify a corresponding smart product service function component according to the smart product service function element instance; A correlation evaluation unit, configured to evaluate the correlation between the functional components of the intelligent product service; a module determining / obtaining / generating unit, configured to determine a smart product service function module based on the correlation; The example of identifying corresponding smart product service function elements based on scenario experience requirements includes: Constructing a first mapping relationship between the scene experience requirements and scene elements; Constructing a second mapping relationship between the scenario elements and the types of smart product service function elements; Constructing a third mapping relationship between the smart product service function element type and the smart product service function element instance; The smart product service function element instance corresponding to the user experience requirement is identified based on the first mapping relationship, the second mapping relationship, and the third mapping relationship.

Images