Resource Search Method, Apparatus and Electronic Device

Abstract

The present invention provides a resource search method, apparatus and electronic device, which are applied to the technical field of resource discovery and scheduling. The method includes: in response to a resource search request configured with a target number of resources sent by a requesting node: performing in-domain search in a resource slice based on a set of neighborhood nodes of the requesting node to obtain a first resource result subset; in response to the number of resources in the first resource result subset not meeting the target number of resources, performing cross-domain search in the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighbor nodes of the requesting node are located to obtain a second resource result subset; merging the first resource result subset and the second resource result subset to obtain a resource result set, and comparing the relationship between the number of resources in the resource result set and the target number of resources; in response to the number of resources in the resource result set meeting the target number of resources, pushing the resource result set. This method can achieve efficient discovery of resources and improve the efficiency and accuracy of resource search.

Description

Technical Field

[0001] The present invention relates to the technical field of resource discovery and scheduling, and particularly to a resource search method, apparatus and electronic device. Background Art

[0002] Cross-domain resource modeling and discovery, as a cutting-edge comprehensive methodology, aims to deeply organize, map and retrieve digital resources scattered in different geographical locations, management domains and organizational structures. Cross-domain resources can include elements such as computing and storage resources, data resources, service resources and user resources. Modeling these cross-domain resources is a way to efficiently utilize these resources. Specifically, cross-domain resource modeling is a process of forming corresponding theoretical models according to the characteristics of different types of resources and using different theoretical modeling methods. Further, the core of cross-domain resource theoretical modeling and cross-domain resource discovery lies in developing an efficient resource search process. However, the current resource search process is difficult to use different types of resource representations and discovery requirements, resulting in low resource search efficiency and accuracy, and it is difficult to achieve efficient discovery of cross-domain resources. Summary of the Invention

[0003] In view of the above problems, the present invention provides a resource search method, apparatus and electronic device for improving resource search efficiency and accuracy.

[0004] One aspect of the present invention provides a resource search method, including: in response to a resource search request configured with a target number of resources sent by a request node, repeatedly performing the following operations until the number of resources in the resource result set meets the target number of resources: performing an in-domain search in a resource section based on a neighborhood node set of the request node to obtain a first resource result subset, where the resource section is obtained by mapping a multi-dimensional stacked network containing multi-domain and multi-type heterogeneous resources; in response to the number of resources in the first resource result subset not meeting the target number of resources, performing a cross-domain search among the cross-domain neighbors of the request node according to the attributes of the domains where the neighbor nodes of the request node are located to obtain a second resource result subset; merging the first resource result subset and the second resource result subset to obtain a resource result set, and comparing the relationship between the number of resources in the resource result set and the target number of resources; in response to the number of resources in the resource result set meeting the target number of resources, pushing the resource result set.

[0005] According to an embodiment of the present invention, the resource section is obtained by mapping a multi-dimensional stacked network containing multi-domain and multi-type heterogeneous resources, including: constructing a heterogeneous graph structure according to the multi-domain and multi-type heterogeneous resources on the multi-dimensional stacked network; constructing a mask gate array based on resource characteristic semantics and the heterogeneous graph structure; and mapping the multi-dimensional stacked network into different types of resource sections through the mask gate array.

[0006] According to an embodiment of the present invention, based on resource characteristic semantics and a heterogeneous graph structure, a masked gate array is constructed, including: obtaining resource characteristic semantics; generating a masked gate array according to the inter-node path reachability matrix of the heterogeneous graph structure under the resource characteristic semantics and the transpose matrix of the inter-node path reachability matrix.

[0007] According to an embodiment of the present invention, the resource search request is further configured with a constraint condition, and the neighborhood node set includes multiple resource nodes; an in-domain search is performed based on the neighborhood node set of the request node in the resource section to obtain a first subset of resource results, including: initializing the access preference for the resource nodes; in response to the existence of a resource that satisfies the constraint condition in the queue, adding the resource that satisfies the constraint condition to the first subset of resource results, where the queue is generated based on the access preference of the resource nodes.

[0008] According to an embodiment of the present invention, the queue is generated based on the access preference of the resource nodes, including: sorting the preference weights of the resource nodes in descending order to obtain a sorting result; adding the first K resource nodes in the sorting result to the queue, where K is a positive integer.

[0009] According to an embodiment of the present invention, in the cross-domain neighbors of the request node, a cross-domain search is performed according to the attributes of the domains where the neighbor nodes of the request node are located to obtain a second subset of resource results, including: for the request node, in the cross-domain neighbors of the request node, selecting the first K' resource nodes to join the queue according to the collaborative management mode of the domains where the neighbor nodes are located, and the collaborative management mode includes at least one of centralized, delegated, and autonomous, where K' is a positive integer; traversing each resource in the queue, dequeuing the resource, and performing an in-domain search in the resource domain where the resource is located; in response to the resource found in the in-domain search satisfying the constraint condition, adding the resource that satisfies the constraint condition to the second subset of resource results.

[0010] According to an embodiment of the present invention, the method further includes: caching the information of the resource nodes to the target node; in response to the state of network connectivity, distributing the information of the resource nodes to the request node through the target node.

[0011] According to an embodiment of the present invention, the access preference of the resource nodes is obtained in the following manner: obtaining the number of times each resource node has been searched to obtain the historical search times; obtaining the historical search rate of the resource nodes according to the ratio of the historical search times to the target resource times; generating the access preference of the resource nodes based on predetermined hyperparameters, the degree of the resource nodes, and the historical search rate of the resource nodes.

[0012] Another aspect of the present invention also provides a resource search device, which includes: a response module, configured to respond to a resource search request configured with a target number of resources sent by a requesting node, and repeatedly perform the following operations until the number of resources in the resource result set meets the target number of resources: perform an in-domain search in a resource section based on the set of neighboring nodes of the requesting node to obtain a first subset of resource results, where the resource section is obtained by mapping a multi-domain and multi-type heterogeneous resource multi-dimensional layered network; in response to the number of resources in the first subset of resource results not meeting the target number of resources, perform a cross-domain search among the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighboring nodes of the requesting node are located to obtain a second subset of resource results; merge the first subset of resource results and the second subset of resource results to obtain a resource result set, and compare the relationship between the number of resources in the resource result set and the target number of resources; a push module, configured to push the resource result set in response to the number of resources in the resource result set meeting the target number of resources.

[0013] Another aspect of the present invention also provides an electronic device, including: one or more processors; a memory for storing one or more computer programs, and the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0014] According to the resource search method, device and electronic device provided by the embodiments of the present invention, by performing an in-domain search on a resource section obtained by mapping a multi-dimensional layered network, and in the case where the number of resources in the first subset of resource results does not meet the target number of resources, performing a cross-domain search according to the neighboring nodes of the requesting node to obtain a subset of resource results that meets the number of resources. Since in the resource search process, multiple heterogeneous resources are considered, not only the same type of resources are considered, and two methods of in-domain search and cross-domain search are combined, efficient discovery of resources can be achieved, and the technical effects of improving the efficiency and accuracy of resource search are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Through the following description of the embodiments of the present invention with reference to the drawings, the above content and other objects, features and advantages of the present invention will become clearer. In the drawings:

[0016] Figure 1 A diagram showing an application scenario of the resource search method according to an embodiment of the present invention is shown;

[0017] Figure 2 A flowchart showing the resource search method according to an embodiment of the present invention is shown;

[0018] Figure 3 A schematic diagram showing the mapping of a multi-dimensional layered network to a resource section according to an embodiment of the present invention is shown;

[0019] Figure 4A flowchart of a resource search method according to another embodiment of the present invention is shown;

[0020] Figure 5 A structural block diagram of a resource search apparatus according to an embodiment of the present invention is shown;

[0021] Figure 6 A block diagram of an electronic device suitable for implementing the resource search method according to an embodiment of the present invention is shown. Detailed implementation manners

[0022] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In the following detailed description, for the sake of explanation, many specific details are set forth in order to provide a thorough understanding of the embodiments of the present invention. However, it is obvious that one or more embodiments can be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of the present invention.

[0023] The terms used herein are merely for describing specific embodiments and are not intended to limit the present invention. The terms "including", "comprising", etc. used herein indicate the presence of the described features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0024] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0025] In the case of using expressions such as "at least one of A, B, and C, etc.", generally, it should be interpreted according to the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include, but is not limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0026] In the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) are all information and data that have been authorized by the user or fully authorized by all parties. Moreover, for the processing of relevant data, such as collection, storage, use, processing, transmission, provision, disclosure, and application, etc., all comply with relevant laws, regulations, and standards, adopt necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or reject.

[0027] In the scenario of making automated decisions using personal information, the methods, devices, and systems provided by the embodiments of the present invention all provide corresponding operation entrances for users to choose to agree or reject the results of automated decisions; if the user chooses to reject, it will enter the expert decision-making process. The expression "automated decision" here refers to the activity of automatically analyzing and evaluating an individual's behavior habits, interests and hobbies, or economic, health, credit status, etc. through a computer program and making decisions. The expression "expert decision" here refers to the activity of making decisions by personnel who are engaged in work in a specific field, have specialized experience, knowledge, and skills, and have reached a certain professional level.

[0028] Currently, the academic and industrial circles have carried out extensive and in-depth research on the modeling and discovery of cross-domain resources. Existing methods mainly include cross-domain resource discovery based on ontology modeling, cross-domain resource representation based on representation learning methods, sparse cross-domain resource representation and scheduling, etc. For example, a method for discovering global resources for specific tasks mainly aims at the problems of a large amount of global resources and opaque resource information between domains. Based on the resource ontology model, it uses techniques such as ontology similarity matching, triangular fuzzy numbers, and fuzzy C-means clustering to complete the search for target resources in three stages. Another example is that when modeling resources, a cross-modal information aggregation method is adopted. First, according to the recall sorting strategy, a dual encoder is used to achieve rough recall, and then a fusion encoder is used to achieve precise sorting; secondly, according to the algorithm of constructing a dual encoder and a fusion encoder based on a multi-way Transformer pre-trained model, high-quality semantic alignment of cross-modal information is achieved, and the performance of the model is improved. Another example is that when modeling resources, starting from the usage rules of resources, a feature matrix is constructed for the resources in each domain, and a hierarchical sparse resource representation method is proposed to accurately describe the resource status of each domain in real time with low complexity. Further, a cross-domain dynamic multi-resource scheduling algorithm is designed, and by introducing a superiority factor and a policy-oriented hyperparameter, efficient cross-domain resource scheduling is achieved.

[0029] Although the above methods achieve resource scheduling, these cross-domain resource modeling techniques usually only consider the same type of resources and are difficult to adapt to the representation and discovery requirements of different types of resources. In addition, during the cross-domain discovery of resources, the above methods ignore the fact that in reality, each resource can only know the resource status within its neighborhood, and do not consider the different preferences for accessing resources within and between domains. Therefore, the embodiments of the present invention propose a method for heterogeneous resource multi-dimensional layered networking, which constructs a semantic-based mask gate array by combining the usage characteristics of various resources, maps the global resource space into resource sections in layers, and realizes the chain-like space conversion of "physical space - logical space - layered section" for hybrid topology resources. When discovering heterogeneous cross-domain resources, a segmented resource discovery method of "intra-domain - inter-domain" is proposed, and a cross-domain resource discovery technology with a dynamic preference strategy is designed according to the constraint description of the characteristics of the required resources, so as to achieve efficient discovery of cross-domain resources.

[0030] Figure 1 FIG. shows an application scenario diagram of the resource search method according to an embodiment of the present invention.

[0031] As Figure 1 shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0032] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as search applications, shopping applications, web browser applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0033] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers, etc.

[0034] Server 105 may be a server that provides various services. For example, it can be a background management server (only for example) that supports the websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server can analyze and process data such as resource search requests received, and feedback the processing results (such as resource result sets, web pages, information, or data obtained or generated according to the requests) to the terminal devices.

[0035] It should be noted that the resource search method provided by the embodiments of the present invention can generally be executed by server 105. Correspondingly, the resource search device provided by the embodiments of the present invention can generally be set in server 105. The resource search method provided by the embodiments of the present invention can also be executed by a server or a server cluster different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the resource search device provided by the embodiments of the present invention can also be set in a server or a server cluster different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0036] It should be understood that Figure 1 the numbers of terminal devices, networks, and servers in

[0037] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers. Figure 1 The following will be based on Figures 2 to 4 the described scenario, and will describe the resource search method of the embodiments of the present invention in detail through

[0038] Figure 2 shows a flowchart of the resource search method according to the embodiments of the present invention.

[0039] As Figure 2 shown, the resource search method of this embodiment includes operation S210 to operation S220.

[0040] In operation S210, in response to a resource search request configured with a target number of resources sent by a request node, the following operations of in-domain search, cross-domain search, and merging search results are repeatedly executed until the number of resources in the resource result set meets the target number of resources:

[0041] Perform an in-domain search based on the neighborhood node set of the request node in the resource section plane, and obtain a first resource result subset. Here, the resource section plane is obtained by mapping a multi-dimensional layered network containing multi-domain and heterogeneous resources;

[0042] In response to the number of resources in the first resource result subset not satisfying the target resource number, performing a cross-domain search among cross-domain neighbors of the requesting node according to the attributes of the domain where the neighboring nodes of the requesting node are located, to obtain a second resource result subset;

[0043] The first resource result subset and the second resource result subset are merged to obtain a resource result set, and the relationship between the number of resources in the resource result set and the number of target resources is compared.

[0044] In operation S220 , in response to the number of resources in the resource result set satisfying the target number of resources, the resource result set is pushed.

[0045] Optionally, the requesting node may be a node that initiates a resource search request. The resource search request may be a query instruction issued by a user or a system to obtain files or data, etc. The number of target resources may be the number of resources to be found in the resource search request, such as the number of files or data to be found. The resource result set may be files or data searched according to the resource search request.

[0046] Optionally, multiple domains and multiple heterogeneous resources, such as user resources, service resources, storage and computing resources, data resources and other resources, can be distributed in multiple domains, and the properties and functions of these resources are different.

[0047] A multi-dimensional stacked network can represent a relationship network between multiple domains and multiple heterogeneous resources. The multi-dimensional stacked network can contain multiple types of nodes and edges, such as physical nodes, logical nodes, intra-layer nodes, inter-layer nodes, physical edges, logical edges, intra-layer edges, and inter-layer edges. This multi-dimensional stacked network can describe the relationship between resources from multiple dimensions (such as users, services, storage, data, etc.).

[0048] Resource slices can be obtained by mapping a multi-dimensional stacked network. Each slice can be composed of cross-domain resources of the same type and their relationships, so as to classify and manage resources. For example, the user resource slice can include all user resources and their relationships, and the service resource slice contains all service resources and their relationships.

[0049] In one example, the set of neighborhood nodes may include resource nodes directly connected to the requesting node in the same domain or layer in a multi-dimensional layered network. An in-domain search can be performed within the domain where the resource nodes in the set of neighborhood nodes are located, and the results of the in-domain search can be added to the first subset of resource results. In one example, the resource search request may be configured with constraint conditions, which may include the target number of resources, the set of resource attributes, the set of resource types, and the set of domains where the resources are located. In one example, an in-domain search can define a queue, select the top K (where K is a positive integer) resource nodes with descending preference weights among the neighborhood nodes to enter the queue, and traverse each resource node in the queue. If the resource indicated by the resource node meets the constraint conditions, the resource is added to the first subset of resource results. The resources in the first subset of resource results can be resources whose resource attributes, resource types, and domains where the resources are located all conform to the set of resource attributes, the set of resource types, and the set of domains where the resources are located in the constraint conditions.

[0050] If the number of resources in the first subset of resource results meets the target number of resources in the constraint conditions, the first subset of resource results can be used as the resource result set pushed to the user or the system; if the number of resources in the first subset of resource results does not meet the target number of resources, for example, is less than the target number of resources, in this case, a cross-domain search can be performed.

[0051] In one example, cross-domain neighbors may be resource nodes directly connected to the requesting node in different domains or layers in a multi-dimensional layered network. The attribute of the domain, for example, is the collaborative management mode. When performing a cross-domain search, cross-domain access can be carried out according to the collaborative management mode and the clustered opportunistic distribution strategy of different domains, and the above-described in-domain resource search process is called within the domain of cross-domain access. For example, select K′ (where K′ is a positive integer) resource nodes or resource domains to join the above queue, traverse each element in the queue, such as a resource node, and repeat the above-described in-domain search process within the resource domain where the resource node is located. If the resource indicated by the resource node meets the above constraint conditions, the resource is added to the second subset of resource results.

[0052] In the case where the sum of the number of resources in the first subset of resource results and the number of resources in the second subset of resource results is equal to the target number of resources, the resources in the first subset of resource results and the resources in the second subset can be merged to obtain a resource result set, and this resource result set can be pushed to the user or the system. In one example, the resources in the first subset of resource results may carry an in-domain search identifier, and the resources in the second subset of resource results may carry a cross-domain search identifier, so that the user or the system can distinguish the resources searched within the domain and the resources searched across domains according to the identifier.

[0053] According to the resource search method, device, and electronic device provided by the embodiments of the present invention, by performing in-domain search on the resource section obtained by mapping the multi-dimensional layered network, and when the number of resources in the first resource result subset does not meet the target number of resources, cross-domain search is performed according to the neighbor nodes of the requesting node to obtain a resource result subset that meets the number of resources. Since various heterogeneous resources are considered during the resource search process, not only the same type of resources are considered, and two methods, in-domain search and cross-domain search, are combined, efficient discovery of resources can be achieved, and the technical effects of improving the efficiency and accuracy of resource search are achieved.

[0054] Figure 3 FIG. shows a schematic diagram of mapping a multi-dimensional layered network into a resource section according to an embodiment of the present invention.

[0055] As Figure 3 shown, the organization of various cross-domain resources is related to their resource types and the topological relationship of the physical nodes where the resources are located. In the actual physical topology, the central domain and its multiple slave domains form a ring structure. For example, the central domain and slave domains A, B, C, D, and E form a ring structure. The slave domain resource service center can serve as the central node to provide computing and storage resources for its affiliated nodes. Therefore, the global computing and storage resource directory is organized in a hybrid topology of ring and star. Services between multiple nodes within a single system domain can access each other, and the service directories between multiple nodes present a mesh structure. The data resources within a single domain are usually organized according to a tree topology. Therefore, the global service resource directory is organized in a hybrid topology of cloud-edge layered multi-level ring and single-domain mesh, and the global data resource directory is organized in a hybrid topology of cloud-edge layered multi-level ring and single-domain tree. Since there can be an organizational hierarchy relationship between user resources, the user resources can be distributed according to the tree topology organization. These resources, such as user resources, service resources, computing and storage resources, and data resources, can rely on the self-organizing and reloadable distributed resource management service to logically organize various resources distributed over a wide area.

[0056] Based on this, the embodiments of the present invention can map various heterogeneous resources to the corresponding resource sections to realize the conversion of the hybrid topology resources from "physical space - logical space - layered section"; then, according to the segmented resource discovery method of "in-domain - inter-domain", efficient discovery of cross-domain resources based on the dynamic preference strategy can be realized. In one example, the resource section can be as Figure 3 shown, and in each resource section, it can be the same type of resources and the relationships between the resources.

[0057] Optionally, Figure 3The resource cross-section can be obtained in the following manner: construct a heterogeneous graph structure based on multi-domain and multi-type heterogeneous resources on a multi-dimensional stacked network; construct a masked gate array based on resource characteristic semantics and the heterogeneous graph structure; and map the multi-dimensional stacked network into different types of resource cross-sections through the masked gate array.

[0058] In some embodiments, a semantic-based masked gate array can be defined for mapping a multi-dimensional stacked network composed of multi-domain and multi-type heterogeneous resources into different types of resource cross-sections. In one example, the semantic-based masked gate array can be constructed based on resource characteristic semantics and the heterogeneous graph structure. Optionally, this process can include the following operations: obtain resource characteristic semantics; generate a masked gate array according to the inter-node path reachability matrix of the heterogeneous graph structure under the resource characteristic semantics and the transpose matrix of the inter-node path reachability matrix.

[0059] In some embodiments, various types of resources rely on a self-organizing and reloadable distributed resource management service to logically organize various types of resources distributed over a wide area.

[0060] Heterogeneous resources can be represented as a heterogeneous graph structure , and each node in the graph structure can represent a resource. can represent the set of all resource nodes, and the nodes in the set can include multiple types, such as users, services, computing and storage, and data, etc.; can represent the set of relationships between resource nodes, and this relationship can include multiple relationship types, such as the relationship between a user and a service, the relationship between a service and data, etc. When performing multi-dimensional stacked mapping on heterogeneous resources, resource characteristic semantics can be utilized. In the heterogeneous graph , can represent a certain meta-path or meta-structure. A meta-path can be used to describe the path pattern between resource nodes. A meta-path can be composed of a series of node types and edge types, representing the path from one node type to another node type. For example, user service data can be a meta-path. A meta-structure can be an extension of a meta-path, describing the structure pattern between resource nodes, such as the hierarchical structure or network structure between multiple resource nodes, such as user service data service user, etc. A meta-structure can represent not only a linear path but also a tree shape, a ring shape, etc.

[0061] Under semantics the masked gate array can be calculated , where , where Representing heterogeneous graph structures exist The node path reachability matrix below can indicate whether there is a certain path between resource nodes, that is, whether they are reachable. Can be The transposed matrix of is a mask gate array, which can be expressed in semantic The reachability between resource nodes.

[0062] According to the above solution process, the reachability matrix of user resources, service resources, storage resources and data resources can be obtained, such as the reachability matrix of user resources , the reachability matrix of service resources , the reachability matrix of storage and computing resources , and the reachability matrix of data resources So far, it has been achieved to map the multi-dimensional stacked network composed of multi-domain and multi-heterogeneous resources into different types of resource slices, each of which is composed of the same type of cross-domain resources and the relationships between them.

[0063] According to an embodiment of the present invention, by constructing a method based on a mask gate array and using a semantic-based mask gate array to map a multi-dimensional stacked network into a resource slice, multi-domain and multi-type heterogeneous resources can be effectively organized and classified, facilitating resource management and access.

[0064] Optionally, the resource search request described above may be configured with constraints, and the neighborhood node set may include multiple resource nodes; the process of performing an intra-domain search based on the neighborhood node set of the requesting node in the resource aspect to obtain a first subset of resource results may include the following operations: initializing the access preferences of the resource nodes; in response to the presence of resources satisfying the constraints in the queue, adding the resources satisfying the constraints to the first subset of resource results, wherein the queue is generated based on the access preferences of the resource nodes.

[0065] Optionally, the queue is generated based on the access preference of the resource nodes. In one embodiment, it can be generated in the following manner: the preference weights of the resource nodes are arranged in descending order to obtain a sorting result; the first K resource nodes in the sorting result are added to the queue, where K is a positive integer.

[0066] In some embodiments, the present invention proposes a biased breadth-first resource access strategy based on resource historical discovery rate and topology. In one example, for the current requesting node, given a set of constraints on the required resources , the number of target resources, that is, the number of resources requested , a collection of properties for each resource , a collection of resource types , the set of domains where resources are located .

[0067] Under the condition that each resource can only obtain the status of its neighboring resources, for the current node , initialize the access preference of the resource nodes in its neighboring node set: , where , represents the resource node , degree of represents the resource node neighboring node set of. The degree of a node can refer to the number of edges directly connected to a node. The higher the degree of a node, the more neighboring nodes it can represent, usually meaning that the node is more important in the network. In the above process, the degree of the node is higher, its access preference can be larger, which can mean that the node has a higher probability of being accessed.

[0068] Define a queue queue, and select the top K resource nodes with the neighboring node preference weights sorted in descending order to enter the queue, that is . Traverse each resource in the queue queue. If the resource meets the constraint condition , then add the resource to the first resource result subset.

[0069] Optionally, the access preference of the resource nodes in the above process can be obtained in the following way: Obtain the number of times each resource node has been searched to get the historical search times; According to the ratio of the historical search times to the target resource times, obtain the historical search rate of the resource node; Based on the predetermined hyperparameters, the degree of the resource node and the historical search rate of the resource node, generate the access preference of the resource node.

[0070] Optionally, after multiple tasks of searching for resources within a domain are executed, the historical search rate of each resource can be calculated , which can be obtained by the ratio of the historical search times of the resource being searched to the target resource times, that is, the total number of tasks of the task of searching for resources within the domain, that is . The historical search rate can reflect the popularity of the resource node . By introducing the historical discovery rate, the access order can be dynamically adjusted to preferentially access more popular resources.

[0071] Based on this, the access preference of the resource node can be calculated as: , where is a hyperparameter that can be used to balance the influence of the topological structure and the historical discovery rate. When When approaching 1, the preference can be mainly determined by the topology. When approaching 0, the preference can be mainly determined by the historical search rate.

[0072] According to an embodiment of the present invention, the above in-domain search method adds preference weights when searching for resources, which can make the order of searching for resources more intelligent. The above in-domain search method dynamically adjusts the preference weights of resource search through the degree of nodes and the historical discovery rate, selects the top K nodes with the highest preference weights to enter the queue, filters the resources that meet the constraint conditions, and balances the influence of the topology and the historical discovery rate through hyperparameters, making the in-domain search method more flexible and intelligent, improving the efficiency of resource discovery and search, and can be applicable to scenarios where resource distribution is uneven or network topology is complex.

[0073] Optionally, if the above in-domain search finds resources that meet the constraint conditions and adds them to the first resource result subset, but the number of resources in the first resource result subset is less than the target number of resources, in this case, cross-domain search for heterogeneous resources can be performed, and cross-domain access can be carried out according to the collaborative management mode and the cluster opportunistic distribution rule of different domains, and the in-domain resource discovery process is called in the new domain.

[0074] Optionally, the process of performing cross-domain search in the cross-domain neighbors of the requesting node to obtain the second resource result subset according to the attributes of the domains where the neighbor nodes of the requesting node are located may include the following operations: for the requesting node, in the cross-domain neighbors of the requesting node, select the top K' resource nodes to join the queue according to the collaborative management mode of the domains where the neighbor nodes are located, the collaborative management mode includes at least one of centralized, delegated, and autonomous, and K' is a positive integer; traverse each resource in the queue, dequeue the resource, and perform in-domain search in the resource domain where the resource is located; in response to the resources found in the in-domain search meeting the constraint conditions, add the resources that meet the constraint conditions to the second resource result subset.

[0075] Optionally, there are different collaborative management modes and different cross-domain access policies between different domains, which may include centralized 、delegated and autonomous . Among them, the centralized mode adopts the direct access method, which means that the requesting node can directly access the target resource without intermediate steps. The delegated mode adopts the conditional access method, and the requesting node needs to meet the conditions to access the target resource and may need to go through an intermediary or authorization. The autonomous mode adopts the non-direct access method, and the requesting node cannot directly access the target resource and may need to indirectly access through other nodes.

[0076] For the resource nodes of the requesting node , in its cross-domain neighbors According to different collaborative management modes of neighbor nodes in different domains, select resource domains to join the queue. This process can be represented as , where represents the resource domain where the resource node is located.

[0077] Traverse each element in the queue queue, such as a resource, dequeue it, and repeat the in-domain resource discovery process described above in the resource domain where the resource is located. If the resource indicated by the resource node meets the constraint conditions , then add it to the second resource result subset.

[0078] According to the embodiments of the present invention, the process of cross-domain resource search can discover resources in multiple domains, no longer only considering the same type of resources, improving the comprehensiveness and efficiency of resource discovery and search.

[0079] Optionally, in the case of cross-domain search, there may be a situation where the network is temporarily unavailable. Based on this, the method provided by the embodiments of the present invention, on the basis of the above operations, may further include the following operations: caching the information of the resource node to the target node; in response to the network connectivity state, distributing the information of the resource node from the target node to the requesting node.

[0080] Optionally, there may be a situation where the network is temporarily unavailable for cross-domain connection. When the network connectivity condition is insufficient, the embodiments of the present invention can cache the resource node information to the target node, such as the central node. The central node can be a global management node responsible for storing and distributing resource information.

[0081] According to the opportunistic distribution strategy, continue to access after the network is connected to achieve the global discovery of resources and their search paths. When the network resumes connectivity, the central node can distribute the resource node information to the requesting node according to the cached information. This distribution strategy can be "opportunistic", meaning that the distribution behavior depends on the network connectivity state. Resource distribution is only performed when the network is connected.

[0082] According to the embodiments of the present invention, through caching and opportunistic distribution, the global discovery of resources and their search paths can be achieved. Even in the case of temporarily unavailable network, resources can still be obtained, improving the reliability of resource access and the global coverage ability.

[0083] In one example, the above process can be described as the following process:

[0084] Input: Requesting node , constraint set of requested resources , number of target resources requested , attribute set of each resource , the set of types corresponding to the resource , the set of domains where the resource is located .

[0085] Output: a resource result set that meets the constraints . .

[0086] Obtain the resource types described by the constraint, perform an in-domain search, and add the results found in the domain to the resource result set If the number of resources in the resource result set is less than , perform a cross-domain search.

[0087] During in-domain search, the top K nodes with the neighborhood preference weights of the nodes can be selected and put into the queue, and the resources that meet the constraints in the queue are inserted into the resource result set.

[0088] During cross-domain search, obtain the cross-domain relationships of the resources in the current resource group, select the top K' cross-domain nodes according to the collaborative management mode and the clustered opportunistic distribution, and determine whether the resources in the queue meet the constraint conditions, and add the resources that meet the constraint conditions to the resource result set.

[0089] Figure 4 shows a flowchart of a resource search method according to another embodiment of the present invention.

[0090] An embodiment of the present invention also provides another resource search method, as Figure 4 shown, this method may include operations S410 to S450.

[0091] In operation S410, map the multi-domain and multi-type heterogeneous resource composed multi-dimensional layered network into different types of resource sections.

[0092] In operation S420, search for resources that meet the constraint condition set within the domain where the request node is located.

[0093] In operation S430, determine whether the current resource group meets the quantity of the requested resources. If it meets, operation S450 can be executed; if it does not meet, operation S440 can be executed.

[0094] In operation S440, perform cross-domain access according to the collaborative management mode of different domains and the clustered opportunistic distribution rules.

[0095] In operation S450, return the obtained resource result set.

[0096] In one example, operations S410 to S440 can refer to operation S210, and operation S450 can refer to operation S220.

[0097] The resource search method provided by the embodiments of the present invention combines the multi-dimensional layered networking based on a semantic mask array and the segmented cross-domain resource discovery algorithm based on a dynamic preference strategy. By aiming at the difficulties of a large variety of resources and cross-domain distribution, a multi-dimensional layered networking method based on a semantic mask array is proposed. A semantic-based mask array is constructed by combining the usage characteristics of various resources, and the global resource space is hierarchically mapped into resource sections, realizing the chain-like space conversion of "physical space - logical space - hierarchical section" for hybrid topology resources. By aiming at the difficulty that only the local neighborhood resource status can be detected for real cross-domain resources and it is difficult to achieve global resource discovery, a "within-domain - between-domains" segmented resource discovery method is proposed. According to the constraint description of the characteristics of the required resources, a cross-domain resource discovery technology with a dynamic preference strategy is designed to achieve efficient discovery of cross-domain resources.

[0098] It should be noted that, unless it is clearly stated that there is a sequential execution order between different operations in the flowcharts shown in the embodiments of the present invention, or there is a sequential execution order between different operations in the technical implementation, the execution order between multiple operations can be unordered, and multiple operations can also be executed simultaneously.

[0099] Based on the above resource search method, the present invention also provides a resource search device. The following will be combined with Figure 5 to describe this device in detail.

[0100] Figure 5 The structural block diagram of the resource search device according to the embodiments of the present invention is shown.

[0101] As Figure 5 shown, the resource search device 500 of this embodiment includes a response module 510 and a push module 520.

[0102] The response module 510 is configured to, in response to a resource search request configured with the number of target resources sent by a requesting node, loop to execute the following operations until the number of resources in the resource result set meets the number of target resources: perform an in-domain search in the resource section based on the set of neighborhood nodes of the requesting node to obtain a first resource result subset, where the resource section is obtained by mapping a multi-dimensional layered network containing multi-domain and heterogeneous resources; in response to the number of resources in the first resource result subset not meeting the number of target resources, perform a cross-domain search among the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighbor nodes of the requesting node are located to obtain a second resource result subset; merge the first resource result subset and the second resource result subset to obtain a resource result set, and compare the relationship between the number of resources in the resource result set and the number of target resources.

[0103] The push module 520 is configured to, in response to the number of resources in the resource result set meeting the number of target resources, push the resource result set.

[0104] According to the resource search method, device, and electronic device provided by the embodiments of the present invention, by performing in-domain search on the resource section obtained by mapping the multi-dimensional layered network, and when the number of resources in the first resource result subset does not meet the target number of resources, cross-domain search is performed according to the neighbor nodes of the requesting node to obtain a resource result subset that meets the number of resources. Since various heterogeneous resources are considered during the resource search process, not only the same type of resources are considered, and two methods, in-domain search and cross-domain search, are combined, efficient discovery of resources can be achieved, and the technical effects of improving the resource search efficiency and accuracy are achieved.

[0105] According to an embodiment of the present invention, the response module 510 may include a first construction sub-module, a second construction sub-module, and a mapping sub-module.

[0106] The first construction sub-module is used to construct a heterogeneous graph structure according to multi-domain and multi-type heterogeneous resources on the multi-dimensional layered network.

[0107] The second construction sub-module is used to construct a mask gate array based on the resource characteristic semantics and the heterogeneous graph structure.

[0108] The mapping sub-module is used to map the multi-dimensional layered network into different types of resource sections through the mask gate array.

[0109] According to an embodiment of the present invention, the second construction sub-module may include an acquisition unit and a first generation unit.

[0110] The acquisition unit is used to acquire the resource characteristic semantics.

[0111] The first generation unit is used to generate a mask gate array according to the node-to-node path reachability matrix of the heterogeneous graph structure under the resource characteristic semantics and the transposed matrix of the node-to-node path reachability matrix.

[0112] According to an embodiment of the present invention, the response module 510 may further include an initialization sub-module and a first addition sub-module.

[0113] The initialization sub-module is used to initialize the access preference of the resource nodes.

[0114] The first addition sub-module is used to add the resources that meet the constraint conditions to the first resource result subset in response to the existence of resources that meet the constraint conditions in the queue, where the queue is generated based on the access preference of the resource nodes.

[0115] According to an embodiment of the present invention, the first addition sub-module may include a sorting unit and an addition unit.

[0116] The sorting unit is used to perform a descending order arrangement on the preference weights of the resource nodes to obtain a sorting result.

[0117] An adding unit, configured to add the top K resource nodes in the sorting result to a queue, where K is a positive integer.

[0118] According to an embodiment of the present invention, the response module 510 may further include a selection sub-module, a traversal sub-module, and a second adding sub-module.

[0119] The selection sub-module is configured to, for a request node, select the top K' resource nodes from the cross-domain neighbors of the request node and add them to the queue according to the collaborative management mode of the domain where the neighbor nodes are located. The collaborative management mode includes at least one of a centralized mode, a delegated mode, and an autonomous mode, and K' is a positive integer.

[0120] The traversal sub-module is configured to traverse each resource in the queue, dequeue the resource, and perform an in-domain search in the resource domain where the resource is located.

[0121] The second adding sub-module is configured to, in response to the resource found in the in-domain search satisfying the constraint condition, add the resource satisfying the constraint condition to the second resource result subset.

[0122] According to an embodiment of the present invention, the resource search device may further include a caching module and a distribution module.

[0123] The caching module is configured to cache the information of the resource nodes to a target node.

[0124] The distribution module is configured to, in response to the state of network connectivity, distribute the information of the resource nodes to the request node through the target node.

[0125] According to an embodiment of the present invention, the first adding sub-module may further include an obtaining unit, a result unit, and a second generating unit.

[0126] The obtaining unit is configured to obtain the number of times each resource node has been searched, so as to obtain the historical search times.

[0127] The result unit is configured to obtain the historical search rate of the resource node according to the ratio of the historical search times to the target resource times.

[0128] The second generating unit is configured to generate the access preference of the resource node based on a predetermined hyperparameter, the degree of the resource node, and the historical search rate of the resource node.

[0129] According to an embodiment of the present invention, any number of the response module 510 and the push module 520 may be combined and implemented in one module, or any one of them may be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules may be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the response module 510 and the push module 520 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or any other reasonable way of integrating or packaging circuits, etc., implemented by hardware or firmware, or implemented in any one of the three implementation manners of software, hardware, and firmware, or in an appropriate combination of any several of them. Alternatively, at least one of the response module 510 and the push module 520 may be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0130] Figure 6 FIG. shows a block diagram of an electronic device suitable for implementing a resource search method according to an embodiment of the present invention.

[0131] As Figure 6 shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) 602 or a program loaded from a storage section 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general microprocessor (such as a CPU), an instruction set processor, and / or a related chipset, and / or a dedicated microprocessor (such as an application specific integrated circuit (ASIC)), etc. The processor 601 may also include on-board memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0132] In the RAM 603, various programs and data required for the operation of the electronic device 600 are stored. The processor 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. The processor 601 performs various operations of the method flow according to an embodiment of the present invention by executing the programs in the ROM 602 and / or the RAM 603. It should be noted that the program may also be stored in one or more memories other than the ROM 602 and the RAM 603. The processor 601 may also perform various operations of the method flow according to an embodiment of the present invention by executing the programs stored in the one or more memories.

[0133] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, and the input / output (I / O) interface 605 is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. The drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0134] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist separately without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiments of the present invention is implemented.

[0135] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the above-described ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603.

[0136] An embodiment of the present invention further includes a computer program product, which includes a computer program, and the computer program includes program code for executing the method shown in the flowchart. When the computer program product runs in a computer system, the program code is used to cause the computer system to implement the resource search method provided by the embodiments of the present invention.

[0137] When the computer program is executed by the processor 601, the above functions defined in the system / apparatus of the embodiments of the present invention are performed. According to an embodiment of the present invention, the above-described system, apparatus, module, unit, etc. can be implemented by computer program modules.

[0138] In one embodiment, the computer program can rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program can also be transmitted and distributed in the form of signals on a network medium, and be downloaded and installed through the communication part 609, and / or be installed from the removable medium 611. The program code included in the computer program can be transmitted by any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0139] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or be installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiments of the present invention are performed. According to an embodiment of the present invention, the above-described system, device, apparatus, module, unit, etc. can be implemented by computer program modules.

[0140] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include but are not limited to, such as Java, C++, python, the "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0141] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code, and the above-mentioned module, segment of a program, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as combinations of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0142] Those skilled in the art can understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of the present invention.

[0143] The above describes the embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although the embodiments are described separately above, this does not mean that the measures in each embodiment cannot be used advantageously in combination. Without departing from the scope of the present invention, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications should fall within the scope of the present invention.

Claims

1. A resource search method, characterized in that, Including: In response to a resource search request configured with a target number of resources sent by a requesting node, the following operations are cyclically executed until the number of resources in the resource result set meets the target number of resources: Perform an in-domain search in a resource section based on the set of neighboring nodes of the requesting node to obtain a first subset of resource results, where the resource section is obtained by mapping a multi-dimensional layered network containing multi-domain and heterogeneous resources; In response to the number of resources in the first subset of resource results not meeting the target number of resources, perform a cross-domain search among the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighboring nodes of the requesting node are located to obtain a second subset of resource results; Merge the first subset of resource results and the second subset of resource results to obtain the resource result set, and compare the relationship between the number of resources in the resource result set and the target number of resources; In response to the number of resources in the resource result set meeting the target number of resources, push the resource result set; Among them, the resource section is obtained by mapping a multi-dimensional layered network containing multi-domain and heterogeneous resources, including: constructing a heterogeneous graph structure according to the multi-domain and heterogeneous resources on the multi-dimensional layered network; constructing a mask gate array based on resource characteristic semantics and the heterogeneous graph structure; mapping the multi-dimensional layered network into different types of resource sections through the mask gate array; Among them, constructing the mask gate array based on resource characteristic semantics and the heterogeneous graph structure includes: obtaining resource characteristic semantics; generating the mask gate array according to the inter-node path reachability matrix of the heterogeneous graph structure under the resource characteristic semantics and the transposed matrix of the inter-node path reachability matrix; 2. The method according to claim 1, wherein, The resource search request is also configured with constraint conditions, and the set of neighboring nodes includes multiple resource nodes; Performing an in-domain search in a resource section based on the set of neighboring nodes of the requesting node to obtain a first subset of resource results includes: Initializing the access preferences of the resource nodes; In response to the existence of a resource in the queue that meets the constraint conditions, adding the resource that meets the constraint conditions to the first subset of resource results, where the queue is generated based on the access preferences of the resource nodes.

3. The method according to claim 2, wherein The queue is generated based on the access preferences of the resource nodes, including: Sorting the preference weights of the resource nodes in descending order to obtain a sorting result; Adding the first K resource nodes in the sorting result to the queue, where K is a positive integer.

4. The method according to claim 3, characterized in that, Performing a cross-domain search among the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighboring nodes of the requesting node are located to obtain a second subset of resource results includes: For the requesting node, among the cross-domain neighbors of the requesting node, select the first K' resource nodes to join the queue according to the collaborative management mode of the domains where the neighboring nodes are located, where the collaborative management mode includes at least one of centralized, delegated, and autonomous, and K' is a positive integer; Traverse each resource in the queue, dequeue the resource, and perform the in-domain search in the resource domain where the resource is located; In response to the resources found within the domain satisfying the constraint condition, add the resources that satisfy the constraint condition to the second resource result subset.

5. The method according to claim 1, wherein The method further includes: Caching the information of the resource node to the target node; In response to the network connectivity status, distribute the information of the resource node to the requesting node through the target node.

6. The method according to claim 2, characterized in that, The access preference of the resource node is obtained in the following manner: Obtain the number of times each resource node has been searched to obtain the historical search times; Obtain the historical search rate of the resource node according to the ratio of the historical search times to the target resource times; Generate the access preference of the resource node based on predetermined hyperparameters, the degree of the resource node, and the historical search rate of the resource node.

7. A resource search device, characterized in that The apparatus includes: A response module, configured to, in response to a resource search request configured with a target resource number sent by a requesting node, loop through the following operations until the number of resources in the resource result set satisfies the target resource number: Perform an in-domain search based on the set of neighborhood nodes of the requesting node in the resource section to obtain a first resource result subset, where the resource section is obtained by mapping a multi-dimensional stacked network including multi-domain and multi-type heterogeneous resources; In response to the number of resources in the first resource result subset not satisfying the target resource number, perform a cross-domain search among the cross-domain neighbors of the requesting node according to the attributes of the domains where the neighbor nodes of the requesting node are located to obtain a second resource result subset; Merge the first resource result subset and the second resource result subset to obtain the resource result set, and compare the relationship between the number of resources in the resource result set and the target resource number; A push module, configured to push the resource result set in response to the number of resources in the resource result set satisfying the target resource number; Wherein, the response module includes a first construction sub-module, a second construction sub-module, and a mapping sub-module. The first construction sub-module is configured to construct a heterogeneous graph structure according to the multi-domain and multi-type heterogeneous resources on the multi-dimensional stacked network; the second construction sub-module is configured to construct a mask gate array based on the resource characteristic semantics and the heterogeneous graph structure; the mapping sub-module is configured to map the multi-dimensional stacked network into different types of resource sections through the mask gate array; The second construction sub-module includes an acquisition unit and a first generation unit. The acquisition unit is configured to acquire resource characteristic semantics; the first generation unit is configured to generate the mask gate array according to the inter-node path reachability matrix of the heterogeneous graph structure under the resource characteristic semantics and the transpose matrix of the inter-node path reachability matrix.

8. An electronic device, including: One or more processors; A memory for storing one or more computer programs, Characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 6.

Images