Method for performing task execution on basis of knowledge graph obtained by means of entity alignment

Abstract

Disclosed in the present disclosure is a method for performing task execution on the basis of a knowledge graph obtained by means of entity alignment. The method specifically comprises: acquiring a knowledge graph pair including a first knowledge graph and a second knowledge graph; screening entity information of the first knowledge graph, so as to select target entities; on the basis of the target entities, adjacent entities of the target entities, and alignment probabilities of the target entities with respect to some entities in the second knowledge graph, determining centralities and uncertainties of the target entities; then, constructing sample entity pairs on the basis of the centralities and uncertainties of the target entities; using the sample entity pairs to train an entity alignment model; and finally, using the trained entity alignment model to perform entity alignment on knowledge graphs, and executing a target task by using a knowledge graph obtained after entity alignment. The method greatly reduces the sample annotation cost during a training process, improves the efficiency of a model training process, and also significantly improves the overall task execution efficiency.

Description

A method for task execution based on knowledge graph obtained by entity alignment Technical Field

[0001] The present disclosure relates to the field of knowledge graph construction, and in particular to a method for performing tasks based on a knowledge graph obtained by entity alignment. Background Art

[0002] In the construction of knowledge graphs, entity alignment can effectively solve the heterogeneous problem when multiple knowledge graphs interoperate, so that heterogeneous knowledge graphs can improve each other's coverage.

[0003] Typically, entity alignment methods rely on manually selecting and labeling entities in knowledge graphs, and then using these labeled entities to train entity alignment models. This method is both time-consuming and labor-intensive, and the labeled entities often fail to effectively train untrained entity alignment models. Entity alignment models trained using this method also generally have low alignment accuracy. Therefore, it is crucial to select entities with high training value to train the entity alignment model, thereby achieving a highly accurate model.

[0004] Summary of the Invention

[0005] The present disclosure provides a method for performing tasks based on a knowledge graph obtained by entity alignment, in order to partially solve the above-mentioned problems existing in the prior art.

[0006] This disclosure adopts the following technical solutions:

[0007] The present disclosure provides a method for executing a task based on a knowledge graph obtained by entity alignment, comprising:

[0008] Obtaining a knowledge graph pair, wherein the knowledge graph pair includes a first knowledge graph and a second knowledge graph;

[0009] Filtering a target entity from a plurality of entity information contained in the first knowledge graph;

[0010] Determine the centrality corresponding to the target entity based on entity information corresponding to neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph;

[0011] Determining the uncertainty corresponding to the target entity based on an alignment probability between at least some entities included in the second knowledge graph and the target entity, and an alignment probability between the entity information corresponding to the neighboring node and at least some entities included in the second knowledge graph;

[0012] Constructing sample entity pairs according to the centrality and uncertainty corresponding to the target entity;

[0013] Training an entity alignment model to be trained based on the sample entity pairs;

[0014] According to the trained entity alignment model, the knowledge graph to be entity aligned is entity aligned to obtain a knowledge graph fused after entity alignment, and the target task is performed based on the fused knowledge graph.

[0015] Optionally, filtering out a target entity from a plurality of entity information included in the first knowledge graph includes:

[0016] Determining the out-degree of each entity information in the plurality of entity information according to the connection relationship between the plurality of entity information included in the first knowledge graph;

[0017] The plurality of entity information are filtered according to the out-degree of each entity information in the plurality of entity information to obtain the target entity.

[0018] Optionally, determining the centrality corresponding to the target entity based on entity information corresponding to neighboring nodes within a preset adjacency range of the node corresponding to the target entity in the first knowledge graph includes:

[0019] Determining the monomer centrality of the target entity according to the out-degree of the target entity;

[0020] Determine the monomer centrality of the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph according to the out-degree of the entity information corresponding to the neighboring nodes;

[0021] The centrality of the target entity is determined according to the individual centrality of the target entity and the individual centralities of the entities corresponding to the neighboring nodes.

[0022] Optionally, determining the uncertainty corresponding to the target entity based on the alignment probability between at least some entities included in the second knowledge graph and the target entity, and the alignment probability between the entity information corresponding to the neighboring node and at least some entities included in the second knowledge graph, includes:

[0023] Determining an alignment probability between the target entity and at least some of the entities included in the second knowledge graph;

[0024] Determining a monomer uncertainty of the target entity based on an alignment probability between the target entity and at least some entities included in the second knowledge graph;

[0025] Determining, based on the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph, an alignment probability between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph;

[0026] Determining a monomer uncertainty of the entity information corresponding to the neighboring node based on an alignment probability between the entity information corresponding to the neighboring node and at least some entities included in the second knowledge graph;

[0027] The uncertainty of the target entity is determined according to the individual uncertainty of the target entity and the individual uncertainty of the entity information corresponding to the neighboring node.

[0028] Optionally, determining the monomer uncertainty of the target entity according to the alignment probability between the target entity and at least some entities included in the second knowledge graph includes:

[0029] Determining a probability difference between the alignment probabilities between the target entity and at least some of the entities included in the second knowledge graph based on the alignment probabilities between the target entity and at least some of the entities included in the second knowledge graph;

[0030] Determining a monomer uncertainty of the target entity based on a probability difference between the target entity and at least some entities included in the second knowledge graph;

[0031] Determining the monomer uncertainty of the entity information corresponding to the neighbor node according to the alignment probability between the entity information corresponding to the neighbor node and at least part of the entities included in the second knowledge graph includes:

[0032] Determining a probability difference between the alignment probabilities between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph based on the alignment probabilities between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph;

[0033] The monomer uncertainty of the entity information corresponding to the neighbor node is determined based on the probability difference between the entity information corresponding to the neighbor node and at least part of the entities included in the second knowledge graph.

[0034] Optionally, constructing a sample entity pair according to the centrality and uncertainty corresponding to the target entity includes:

[0035] Determining a centrality weight corresponding to the centrality of the target entity, and determining an uncertainty weight corresponding to the uncertainty of the target entity;

[0036] Determining a representative value of the target entity according to the centrality corresponding to the target entity, the centrality weight, the uncertainty corresponding to the target entity, and the uncertainty weight;

[0037] The target entities are screened according to the representative values ​​of the target entities, and sample entity pairs are constructed based on the screened target entities.

[0038] Optionally, determining a centrality weight corresponding to the centrality of the target entity and determining an uncertainty weight corresponding to the uncertainty of the target entity include:

[0039] According to the difference between the centrality and the uncertainty corresponding to the target entity, the centrality weight corresponding to the centrality of the target entity is determined, and the uncertainty weight corresponding to the uncertainty of the target entity is determined.

[0040] Optionally, determining a centrality weight corresponding to the centrality of the target entity and determining an uncertainty weight corresponding to the uncertainty of the target entity include:

[0041] According to the current training round of the entity alignment model, the center weight corresponding to the center of the target entity is determined, and the uncertainty weight corresponding to the uncertainty of the target entity is determined, wherein, the larger the current training round is, the smaller the center weight corresponding to the center of the target entity is, and the larger the uncertainty weight corresponding to the uncertainty of the target entity is.

[0042] The present disclosure provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for performing tasks based on the knowledge graph obtained by entity alignment.

[0043] The present disclosure provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for executing tasks based on the knowledge graph obtained by entity alignment is implemented.

[0044] At least one of the above technical solutions adopted in the present disclosure can achieve the following beneficial effects:

[0045] It can be seen from the above method that in the method for performing tasks based on the knowledge graph obtained by entity alignment provided by the present disclosure, a knowledge graph pair including a first knowledge graph and a second knowledge graph is obtained, and each target entity is selected by screening the entity information of each target entity and each target entity's neighboring nodes, and the alignment probability of each target entity with the entities contained in the second knowledge graph are respectively determined. Then, each sample entity pair is constructed according to the centrality and uncertainty of each target entity, and each sample entity pair is used to train the entity alignment model to be trained. Finally, the trained entity alignment model is used to perform entity alignment on each knowledge graph, and the knowledge graph after entity alignment fusion is used to perform the target task.

[0046] From the above content, it can be seen that the method for executing tasks based on the knowledge graph obtained by entity alignment provided by the present disclosure can evaluate the target entities with higher training value in the training process of the entity alignment model according to the centrality and uncertainty of each target entity in the knowledge graph, thereby using the above target entities with higher training benefits in the training process of the entity alignment model, and then obtaining an entity alignment model with higher alignment accuracy to perform the target task. This method can not only obtain an entity alignment model with higher alignment accuracy, but also greatly reduce the sample annotation cost in the training process, improve the efficiency of the model training process, and significantly improve the efficiency of the overall task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The exemplary embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation of the present disclosure. In the drawings:

[0048] FIG1 is a flow chart of a method for executing a task based on a knowledge graph obtained by entity alignment provided in the present disclosure;

[0049] FIG2 is a schematic diagram of a knowledge graph provided by the present disclosure;

[0050] FIG3 is a schematic diagram of a knowledge graph fusion through entity alignment provided by the present disclosure;

[0051] FIG4 is a schematic diagram of a device for performing tasks based on a knowledge graph obtained by entity alignment, provided by the present disclosure;

[0052] FIG5 is a schematic structural diagram of an electronic device provided in the present disclosure corresponding to FIG1 . DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions, and advantages of the present disclosure more clear, the technical solutions of the present disclosure will be clearly and completely described below in conjunction with the specific embodiments of the present disclosure and the corresponding drawings. The described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present disclosure.

[0054] The technical solutions provided by various embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0055] FIG1 is a flow chart of a method for performing a task based on a knowledge graph obtained by entity alignment provided in the present disclosure, which includes the following steps:

[0056] S101: Obtain a knowledge graph pair, wherein the knowledge graph pair includes a first knowledge graph and a second knowledge graph.

[0057] In the process of knowledge graph construction, the entity alignment method can be used to fuse heterogeneous knowledge graphs and improve the coverage of the knowledge graph. Training the entity alignment model iteratively through model training can effectively improve the efficiency of entity alignment. Therefore, it is particularly important to select entities with high training returns for the training process of the entity alignment model and then obtain an entity alignment model with high alignment accuracy.

[0058] The execution entity of the method for executing tasks based on the knowledge graph obtained by entity alignment provided in this disclosure can be a terminal device such as a desktop computer or laptop computer, or a server. In addition, the execution entity of this disclosure can also be a software entity, such as a client installed in the terminal device, or an online platform that can be logged in online. For ease of explanation, this disclosure only uses the terminal device as the execution entity to illustrate the provided task execution method.

[0059] Based on this, a terminal device that applies the method for performing tasks based on the knowledge graph obtained by entity alignment provided by the present disclosure can fuse multiple knowledge graphs through the entity alignment model, and use the fused knowledge graph to perform the target task, wherein the target task performed by the terminal device can be determined according to the actual scenario. For example, in the information recommendation scenario, when recommending tourist attraction related information to the user, the knowledge graph containing tourist attraction information and the knowledge graph containing local food information can be fused into a knowledge graph containing tourist attraction information and local food information, and pushed to the user, so that when the user views the tourist attraction related information, he can understand and choose the tourist attractions and local food at the same time; for another example, in the risk control scenario, the risk control security of a single risk type is usually low. The terminal device can fuse multiple knowledge graphs containing a single risk type, and use the fused knowledge graph with multiple risk types to perform risk control tasks to improve the efficiency and security of the overall risk control task.

[0060] In the present disclosure, a terminal device may obtain a knowledge graph pair including a first knowledge graph and a second knowledge graph. The terminal device may obtain multiple sets of knowledge graph pairs containing known entity alignment results and select the first knowledge graph and the second knowledge graph from each of the above knowledge graph pairs.

[0061] It should be noted that the selection of the first knowledge graph and the second knowledge graph is not strictly set, but relative. For example, for knowledge graph A and knowledge graph B in a certain knowledge graph pair, if the target entity is selected in knowledge graph A, knowledge graph A is selected as the first knowledge graph and knowledge graph B is the second knowledge graph. Conversely, if the target entity is selected in knowledge graph B, knowledge graph B is selected as the first knowledge graph and knowledge graph A is the second knowledge graph.

[0062] S102: Filter out a target entity from the multiple entity information included in the first knowledge graph.

[0063] In the present disclosure, the terminal device can filter out each target entity from the entity information contained in the first knowledge graph of the knowledge graph pair.

[0064] Among them, the terminal device can determine the core entity in the first knowledge graph based on the number of first-order adjacent entities of each entity information in the first knowledge graph and whether the order of the farthest adjacent entity of the first-order adjacent entity meets the preset value, and use the above core entity as the target entity. Specifically, for each entity information in the first knowledge graph, when the number of first-order adjacent entities of the entity information is greater than or equal to the preset value of the preset number of first-order adjacent entities, the terminal device will judge the order of the farthest adjacent entity of all first-order adjacent entities of the entity information. If there are one or more orders of the farthest adjacent entities of all first-order adjacent entities of the entity information that are greater than or equal to the preset value of the order of the farthest adjacent entity of the preset first-order adjacent entity, then the entity information can be determined as a core entity, and the core entity can be used as the target entity. In order to facilitate the explanation of how to determine the core entity in the knowledge graph, a schematic diagram of a knowledge graph will be used for introduction, as shown in Figure 2.

[0065] FIG2 is a schematic diagram of a knowledge graph provided by the present disclosure.

[0066] In the knowledge graph shown in Figure 2, entities B, C, F, and E are all first-order neighboring entities of entity A, meaning that entity A has four first-order neighboring entities. Entity O is entity B's third-order neighboring entity and its furthest neighboring entity. Therefore, when the preset value for the number of first-order neighboring entities is 4 and the preset value for the order of the furthest neighboring entity of a first-order neighboring entity is 3, the core entity of the knowledge graph shown in Figure 2 is entity A, and entity A is used as the target entity when the knowledge graph is used as the first knowledge graph.

[0067] It should be noted that, in actual applications, the above preset values ​​can be determined according to actual needs. The specific values ​​of the preset values ​​in the example of FIG2 are listed for ease of understanding. Therefore, the present disclosure does not limit the specific values ​​of the preset values.

[0068] In addition to the above method, the terminal device may also determine each core entity according to the out-degree of each entity information in the plurality of entity information, thereby determining each target entity.

[0069] Among them, the terminal device can remove the entity information with an out-degree of 1 in the first knowledge graph. After the removal, entity information with an out-degree of 1 may be generated again in the first knowledge graph, and it will be removed again. This process is iterated repeatedly to peel off the entity information of the first knowledge graph layer by layer until the entity information with an out-degree of 1 no longer appears in the first knowledge graph, and then the core entities of the first knowledge graph are determined.

[0070] Continuing with the previous example, in the knowledge graph shown in Figure 2, entities K, P, and I all have an out-degree of 1. After these entities with an out-degree of 1 are removed one by one, entities E, M, and D will become out-degree 1 entities. Similarly, these out-degree 1 entities are removed layer by layer. Ultimately, only entity A remains in the knowledge graph shown in Figure 2. Therefore, entity A is the core entity of the knowledge graph shown in Figure 2, and the terminal device can also use entity A as the target entity when the knowledge graph is the first knowledge graph.

[0071] S103: Determine the centrality corresponding to the target entity based on the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph, and determine the uncertainty corresponding to the target entity based on the alignment probability between at least some entities included in the second knowledge graph and the target entity, and the alignment probability between the entity information corresponding to the neighboring nodes and at least some entities included in the second knowledge graph.

[0072] In the present disclosure, in order to select target entities with higher training benefits for the training process of the entity alignment model, the terminal device needs to determine the centrality and uncertainty of each target entity. The centrality of the target entity can be used to reflect whether the target entity is a central alignment entity in the entity alignment process of the knowledge graph to which it belongs. If a target entity with high centrality is used in the training process, the training results of the entity alignment model training process can be effectively compared with the standard results to improve the model training efficiency; the uncertainty of the target entity can be used to reflect the difficulty of alignment when the knowledge graph to which the target entity belongs performs the entity alignment process. If a target entity with high uncertainty is used in the training process, the entity alignment capability of the entity alignment model can be effectively trained and improved, thereby improving the alignment accuracy of the entity alignment model.

[0073] It should be noted that this disclosure takes into account the influence of the target entity's neighbor entities on the target entity. Therefore, the centrality and uncertainty of the target entity are jointly determined by the target entity itself and its neighbor entities. The centrality of the target entity can be specifically reflected by determining the centrality of the target entity, and the uncertainty of the target entity can be specifically reflected by determining the uncertainty of the target entity.

[0074] For each target entity, the terminal device determines the centrality of the target entity based on the entity information corresponding to the target entity and the neighboring nodes of the target entity's corresponding node within a preset adjacent range. Specifically, the terminal device can determine the individual centrality of the target entity based on the out-degree of the target entity, and determine the individual centrality of the entity information corresponding to the neighboring nodes of the target entity's corresponding node based on the out-degree of the entity information corresponding to the neighboring nodes. Then, the centrality of the target entity is determined based on the individual centrality of the target entity and the individual centrality of the entity information corresponding to the neighboring nodes of the target entity. Specifically, refer to the following formula:

[0075] Among them, e i Represented as the i-th target entity, R n (e i ) is expressed as the centrality of the i-th target entity, k s (e i ) is represented as the monomer centrality of the i-th target entity, k s (e j ) represents the monomer centrality of the j-th entity information corresponding to the neighboring node of the target entity corresponding node within the preset adjacent range, and p represents the preset constant coefficient in the interval [0,1]. It is represented as the nth-order adjacent entity of the i-th target entity.

[0076] From the above formula, we can know that when the target entity e i The larger the adjacency order n, the smaller the impact of the neighboring node's entity information on the target entity's centrality. This disclosure considers that when the adjacency order n is too large, it will generate noise interference in the determination of the target entity's centrality. Therefore, the adjacency order n is generally preset to 3. Of course, the adjacency order n can also be dynamically adjusted according to actual conditions.

[0077] In the present disclosure, for each target entity, the terminal device can determine the uncertainty of the target entity based on the alignment probability between the target entity and at least some entities contained in the second knowledge graph in the knowledge graph pair, and the alignment probability between the entity information corresponding to the neighboring nodes of the target entity and at least some entities contained in the above-mentioned second knowledge graph.

[0078] Among them, the terminal device can determine the probability difference of the alignment probability between the target entity and at least part of the entities contained in the second knowledge graph by utilizing the alignment probability between the target entity and at least part of the entities contained in the second knowledge graph in the knowledge graph pair, and determine the probability difference of the alignment probability between the entity information corresponding to the neighboring node of the target entity and at least part of the entities contained in the second knowledge graph by utilizing the alignment probability between the entity information corresponding to the neighboring node of the target entity and at least part of the entities contained in the second knowledge graph, specifically referring to the following formula:

[0079] Among them, bt(e i ) is expressed as the probability difference between the alignment probability of the i-th target entity and at least part of the entities contained in the second knowledge graph, and can also be expressed as the monomer uncertainty of the i-th target entity. Represented as the i-th target entity and the entities contained in the second knowledge graph The probability of alignment between .

[0080] It should be noted that bt(e i ) is negatively correlated with the uncertainty of the target entity. i ) indicates a greater singleton uncertainty for the target entity. This indicates that, among at least some entities included in the second knowledge graph, there may be multiple entity information that can be aligned with the target entity. The singleton uncertainty of the entity information corresponding to the target entity's neighbor nodes can also be determined using the above formula.

[0081] In some embodiments, when determining the probability difference of the alignment probability between the target entity and at least some of the entities included in the second knowledge graph, the two entities in the second knowledge graph with the highest probability of alignment with the target entity can be selected for calculation. That is, the difference between the first alignment probability and the second alignment probability is determined, where the first alignment probability is the alignment probability between the entity with the highest probability of alignment with the target entity in the second knowledge graph and the target entity, and the second alignment probability is the alignment probability between the entity with the second highest probability of alignment with the target entity in the second knowledge graph and the target entity. The smaller the difference between the first alignment probability and the second alignment probability, the greater the monomer uncertainty of the target entity. The monomer uncertainty of the entity information corresponding to the neighboring nodes of the target entity can also be determined similarly.

[0082] Furthermore, the terminal device may determine the uncertainty of the target entity by using the uncertainty of the target entity and the uncertainty of the entity information corresponding to the neighboring nodes of the target entity. Specifically, refer to the following formula:

[0083] Among them, ei Represented as the i-th target entity, I n (e i ) is expressed as the uncertainty of the i-th target entity, bt(e i ) is expressed as the uncertainty of the individual target entity i, bt(e j ) represents the uncertainty of the j-th entity information corresponding to the neighboring node of the target entity, and p represents the preset constant coefficient in the interval [0,1]. It is represented as the nth-order adjacent entity of the i-th target entity.

[0084] From the above formula, we can know that when the target entity e i The larger the adjacency order n, the smaller the impact of the individual uncertainty of the entity information of the neighboring node on the uncertainty of the target entity. The reason is the same as the above-mentioned formula used to determine the centrality of the target entity. This disclosure takes into account that when the value of the adjacency order n is too large, it will generate noise interference in the determination process of the uncertainty of the target entity. Therefore, the adjacency order n is generally preset to 3. Of course, the adjacency order n can also be dynamically adjusted according to actual conditions.

[0085] S104: Construct each sample entity pair according to the centrality and uncertainty corresponding to each target entity.

[0086] In the present disclosure, the terminal device determines the representative value of each target entity based on the centrality and uncertainty corresponding to each target entity, and selects the target entities used to construct each sample entity pair from these target entities based on the representative value.

[0087] After determining the target entity for constructing a sample entity pair from the first knowledge graph, the entity in the second knowledge graph corresponding to the target entity can be determined based on the correspondence between the entity in the first knowledge graph and the entity in the second knowledge graph, thereby determining the sample entity pair.

[0088] The terminal device may select target entities with larger representative values ​​to construct each sample entity pair. A larger representative value indicates better centrality of the target entity, which facilitates entity alignment, and stronger uncertainty, which improves the alignment capability of the entity alignment model. Specifically, the terminal device may sum the centrality and uncertainty corresponding to the target entity to determine the representative value of the target entity.

[0089] Of course, when determining the above representative value, the terminal device can also introduce the centrality weight corresponding to the centrality and the uncertainty weight corresponding to the uncertainty, so as to determine the representative value of the target entity through the centrality, centrality weight, uncertainty and uncertainty weight corresponding to the target entity.

[0090] The centrality weight and uncertainty weight are mainly used to measure the contribution of uncertainty and centrality when determining the representative value of the target entity. The determination of centrality weight and uncertainty weight is closely related to centrality and uncertainty itself. The specific determination method can be: for any target entity, the terminal device can determine the centrality weight corresponding to the centrality of the target entity and the uncertainty weight corresponding to the uncertainty of the target entity based on the difference between the centrality and uncertainty corresponding to the target entity. For details, please refer to the following formula: dif(s) = f R (s)-f I (s) AC(s)=αf R (s)+(1-α)f I (s)

[0091] Among them, f R (s) represents the centrality of the target entity, f I (s) represents the uncertainty of the target entity, dif(s) represents the difference between the centrality and uncertainty of the target entity, α and (1-α) represent the centrality weight corresponding to the centrality of the target entity and the uncertainty weight corresponding to the uncertainty of the target entity, respectively, p represents a preset constant coefficient in the interval [0,1], [-θ, θ] represents the preset difference range interval, AC(s) represents the weight distribution of the centrality and uncertainty of the target entity in each sample entity pair, that is, the representative value of the target entity.

[0092] It can be seen from the above formula that the terminal device can determine the centrality weight and uncertainty weight of the target entity according to the interval range of the difference between the centrality and uncertainty of each target entity, and then determine the representative value of each target entity.

[0093] Furthermore, it can be seen from the above formula that when the centrality corresponding to a target entity is significantly greater than the uncertainty corresponding to the target entity, then when determining the representative value of the target entity, only the centrality corresponding to the target entity can be referred to, without considering the uncertainty corresponding to the target entity. When there is a certain deviation between the centrality corresponding to the target entity and the uncertainty corresponding to the target entity, but it is within the preset deviation range, then when determining the representative value of the target entity, both the centrality corresponding to the target entity and the uncertainty corresponding to the target entity should be considered. When the centrality corresponding to the target entity is significantly less than the uncertainty corresponding to the target entity, then when determining the representative value of the target entity, only the uncertainty corresponding to the target entity can be considered, while ignoring the centrality corresponding to the target entity.

[0094] Of course, in the early stages of the entity alignment model training process, the training gain value brought by constructing each sample entity pair based on the centrality of the target entity is far greater than the training gain value brought by constructing each sample entity pair based on the uncertainty of the target entity.

[0095] Therefore, in the present disclosure, the centrality weight corresponding to the centrality of a target entity and the uncertainty weight corresponding to the uncertainty of the target entity can be determined through the current training round of the entity alignment model, wherein the larger the current training round, the smaller the centrality weight corresponding to the centrality of the target entity, and the larger the uncertainty weight corresponding to the uncertainty of the target entity.

[0096] That is to say, in the early stage of the entity alignment model training process, the terminal device can construct each sample entity pair based on the centrality of each target entity alone, using each target entity with a higher centrality in the first knowledge graph and each aligned entity corresponding to each target entity in the second knowledge graph to form each sample entity pair constructed based on the centrality of the target entity. Specifically expressed in the above formula, the centrality f of the target entity in the representative value AC(s) of the target entity is R The weight α of (s) is 1, and the uncertainty f of the target entity I The weight (1-α) of (s) is 0.

[0097] In the mid-stage of the entity alignment model training process, although the entity alignment model has undergone a certain degree of training, the overall entity alignment capability of the model is still unstable and the alignment accuracy is not high, making it difficult to meet the standard requirements. At this time, the training value of the target entity's centrality and uncertainty for the current training process is not significantly different, and both have a certain training gain for the entity alignment model training process. Therefore, constructing each sample entity pair based on both the target entity's centrality and uncertainty will bring higher training benefits to the entity alignment model than constructing each sample entity pair based on either the target entity's centrality or the target entity's uncertainty alone. Therefore, in the mid-stage of the entity alignment model training process, the terminal device can construct each sample entity pair based on both the target entity's centrality and uncertainty.

[0098] Specifically in the above formula, the centrality f of the target entity is R (s) is slightly larger than the uncertainty f of the target entity I (s), the centrality f of the target entity in the representative value AC(s) of the target entity R The weight α of (s) is the preset value p, and the uncertainty f of the target entity I The weight (1-α) of (s) is 1-p; the uncertainty of the target entity f I(s) is slightly larger than the centrality Δf of the target entity R (s), the uncertainty f of the target entity in the representative value AC(s) of the target entity I The weight (1-α) of (s) is the preset value p, and the centrality of the target entity is f R The weight α of (s) is 1-p.

[0099] In the later stages of the entity alignment model training process, when the entity alignment model has a certain accuracy of entity alignment capability after multiple rounds of training, constructing each sample entity pair based on the uncertainty of the target entity will bring higher training value to the model training process than constructing each sample entity based on the center of the target entity. Using target entities with higher uncertainty to train the entity alignment model can effectively improve the entity alignment capability of the entity alignment model, and thus obtain an entity alignment model with higher alignment accuracy. Therefore, in the later stages of the entity alignment model training process, the terminal device can construct each sample entity pair based on the uncertainty of each target entity alone, and use each target entity with higher uncertainty in the first knowledge graph and each alignment entity corresponding to each target entity in the second knowledge graph to form each sample entity pair constructed based on the uncertainty of the target entity. Specifically expressed in the above formula, the center of the target entity f in the representative value AC(s) of the target entity R The weight α of (s) is 0, and the uncertainty f of the target entity I The weight (1-α) of (s) is 1.

[0100] S105: Training an entity alignment model to be trained according to the sample entity pairs.

[0101] In the present disclosure, the terminal device may use each sample entity pair in the training process of the entity alignment model to be trained, and obtain an entity alignment model with entity alignment capability after multiple rounds of training iterations.

[0102] It should be noted that at the end of each round of training, the terminal device will recalculate the centrality weight and uncertainty weight of each target entity according to the above formula for constructing each sample entity pair, and then construct each sample entity pair for the new round of training process.

[0103] S106: Perform entity alignment on the knowledge graphs to be entity aligned according to the trained entity alignment model to obtain a knowledge graph fused after entity alignment, and execute the target task according to the fused knowledge graph.

[0104] In the present disclosure, the terminal device will perform entity alignment on multiple knowledge graphs to be entity aligned involved in the target task by using a trained entity alignment model with entity alignment capabilities, and fuse the different information about the same entity in different knowledge graphs, so that multiple knowledge graphs can be fused into a knowledge graph with a wider coverage of entity information, and then use the fused knowledge graph to perform the target task. In order to facilitate the explanation of the fusion process of entity alignment of multiple knowledge graphs, the following will be introduced with a schematic diagram of the fusion of a knowledge graph through entity alignment, as shown in Figure 3.

[0105] FIG3 is a schematic diagram of a knowledge graph fusion through entity alignment provided by the present disclosure.

[0106] In the graph shown in Figure 3, entity a, entity c, entity d, etc. are all entity information shared by knowledge graph 1 and knowledge graph 2, while entity information such as entity b and entity h belong to the different entity information in knowledge graph 1 and knowledge graph 2. It can be seen from Figure 3 that after entity alignment of knowledge graph 1 and knowledge graph 2, knowledge graph 3 is obtained. Knowledge graph 3 contains all the shared entity information in knowledge graph 1 and knowledge graph 2, and also integrates the different entity information between knowledge graph 1 and knowledge graph 2. The coverage of the final knowledge graph 3 covers both knowledge graph 1 and knowledge graph 2.

[0107] From the above content, it can be seen that the method for executing tasks based on the knowledge graph obtained by entity alignment provided by the present disclosure can evaluate the target entities with higher training value in the training process of the entity alignment model according to the centrality and uncertainty of each target entity in the knowledge graph, thereby using the above target entities with higher training benefits in the training process of the entity alignment model, and then obtaining an entity alignment model with higher alignment accuracy to perform the target task. This method can not only obtain an entity alignment model with higher alignment accuracy, but also greatly reduce the sample annotation cost in the training process, improve the efficiency of the model training process, and significantly improve the efficiency of the overall task execution.

[0108] The above are one or more implementation methods of the present disclosure. Based on the same idea, the present disclosure also provides a corresponding device for performing tasks based on the knowledge graph obtained by entity alignment, as shown in Figure 4.

[0109] FIG4 is a schematic diagram of a device for performing tasks based on a knowledge graph obtained by entity alignment, provided by the present disclosure, comprising:

[0110] An acquisition module 401 is configured to acquire a knowledge graph pair, wherein the knowledge graph pair includes a first knowledge graph and a second knowledge graph;

[0111] A screening module 402 is configured to screen out a target entity from a plurality of entity information items included in the first knowledge graph;

[0112] A calculation module 403 determines the centrality corresponding to the target entity based on entity information corresponding to neighboring nodes within a preset adjacency range of the node corresponding to the target entity in the first knowledge graph, and determines the uncertainty corresponding to the target entity based on the alignment probability between at least some entities included in the second knowledge graph and the target entity, and the alignment probability between the entity information corresponding to the neighboring nodes and at least some entities included in the second knowledge graph;

[0113] A construction module 404 is used to construct a sample entity pair according to the centrality and uncertainty corresponding to the target entity;

[0114] A training module 405 is configured to train an entity alignment model to be trained based on the sample entity pairs;

[0115] The execution module 406 is used to perform entity alignment on the knowledge graph to be entity aligned according to the trained entity alignment model to obtain a knowledge graph fused after entity alignment, and execute the target task according to the fused knowledge graph.

[0116] Optionally, the filtering module 402 is specifically used to determine the out-degree of each entity information in the multiple entity information based on the connection relationship between the multiple entity information contained in the first knowledge graph; and filter the multiple entity information based on the out-degree of each entity information in the multiple entity information to obtain the target entity.

[0117] Optionally, the calculation module 403 is specifically used to determine the monomer centrality of the target entity based on the out-degree of the target entity; determine the monomer centrality of the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph based on the out-degree of the entity information corresponding to the neighboring nodes; determine the centrality of the target entity based on the monomer centrality of the target entity and the monomer centrality of the entity corresponding to the neighboring nodes.

[0118] Optionally, the calculation module 403 is specifically used to determine the alignment probability between the target entity and at least some of the entities contained in the second knowledge graph; determine the monomer uncertainty of the target entity based on the alignment probability between the target entity and at least some of the entities contained in the second knowledge graph; determine the alignment probability between the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph based on the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph; determine the monomer uncertainty of the entity information corresponding to the neighbor node based on the alignment probability between the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph; determine the uncertainty of the target entity based on the monomer uncertainty of the target entity and the monomer uncertainty of the entity information corresponding to the neighbor node.

[0119] Optionally, the calculation module 403 is specifically configured to determine, based on the alignment probability between the target entity and at least some of the entities included in the second knowledge graph, a probability difference between the alignment probabilities of the target entity and at least some of the entities included in the second knowledge graph; and determine the monomer uncertainty of the target entity based on the probability difference between the alignment probabilities of the target entity and at least some of the entities included in the second knowledge graph;

[0120] The calculation module 403 is specifically used to determine the probability difference between the alignment probabilities of the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph based on the alignment probability between the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph; and determine the monomer uncertainty of the entity information corresponding to the neighbor node based on the probability difference between the alignment probabilities of the entity information corresponding to the neighbor node and at least some of the entities contained in the second knowledge graph.

[0121] Optionally, the construction module 404 is specifically used to determine the centrality weight corresponding to the centrality of the target entity, and determine the uncertainty weight corresponding to the uncertainty of the target entity; determine the representative value of the target entity based on the centrality corresponding to the target entity, the centrality weight, the uncertainty corresponding to the target entity, and the uncertainty weight; screen the target entity based on the representative value of the target entity, and construct a sample entity pair based on the screened target entities.

[0122] Optionally, the construction module 404 is specifically used to determine the centrality weight corresponding to the centrality of the target entity and determine the uncertainty weight corresponding to the uncertainty of the target entity based on the difference between the centrality and uncertainty corresponding to the target entity.

[0123] Optionally, the construction module 404 is specifically used to determine the center weight corresponding to the center of the target entity and the uncertainty weight corresponding to the uncertainty of the target entity based on the current training round of the entity alignment model, wherein the larger the current training round, the smaller the center weight corresponding to the center of the target entity, and the larger the uncertainty weight corresponding to the uncertainty of the target entity.

[0124] The present disclosure also provides a computer-readable storage medium storing a computer program, which can be used to execute a method for performing a task based on a knowledge graph obtained by entity alignment as provided in FIG. 1 above.

[0125] The present disclosure also provides a schematic structural diagram of an electronic device corresponding to FIG1 , as shown in FIG5 . As shown in FIG5 , at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the method for performing tasks based on the knowledge graph obtained by entity alignment as described in FIG1 above.

[0126] In the 1990s, technological improvements could be clearly distinguished as either hardware improvements (for example, improvements to circuit structures like diodes, transistors, and switches) or software improvements (improvements to process flows). However, with the advancement of technology, many process flow improvements today can now be considered direct improvements to hardware circuit structures. Designers almost always create the corresponding hardware circuit structure by programming the improved process flow into the hardware circuit. Therefore, it cannot be said that a process flow improvement cannot be implemented using hardware modules. For example, a programmable logic device (PLD), such as a field programmable gate array (FPGA), is an integrated circuit whose logical function is determined by user programming. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to hire a chip manufacturer to design and manufacture a dedicated integrated circuit chip. Moreover, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly done using "logic compiler" software. This is similar to the software compiler used when developing programs. Before compilation, the original code must also be written in a specific programming language, called a hardware description language (HDL). There is not just one HDL, but many, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art will also understand that by simply programming the method flow in one of these hardware description languages ​​and then programming it into an integrated circuit, a hardware circuit that implements the logic method flow can be easily obtained.

[0127] The controller can be implemented in any suitable manner. For example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that in addition to implementing the controller in a purely computer-readable program code format, the controller can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered as structures within the hardware component. Or even, the devices for implementing various functions can be considered as both software modules that implement the method and structures within the hardware component.

[0128] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0129] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing the present disclosure, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0130] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0132] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0134] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0135] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0136] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0137] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0138] Those skilled in the art will appreciate that embodiments of the present disclosure may be provided as methods, systems, or computer program products. Thus, the present disclosure may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] The present disclosure may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present disclosure may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0140] The various embodiments of this disclosure are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so their description is relatively simple. For relevant portions, refer to the description of the method embodiments.

[0141] The foregoing is merely an embodiment of the present disclosure and is not intended to limit the present disclosure. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure are intended to be included within the scope of the claims of the present disclosure.

Claims

1. A method for performing tasks based on a knowledge graph obtained by entity alignment, characterized in that: include: Obtain a knowledge graph pair, wherein the knowledge graph pair includes a first knowledge graph and a second knowledge graph; Filtering a target entity from a plurality of entity information contained in the first knowledge graph; Determine the centrality corresponding to the target entity according to the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph; Determine the uncertainty corresponding to the target entity according to the alignment probability between at least some entities included in the second knowledge graph and the target entity, and the alignment probability between the entity information corresponding to the neighbor node and at least some entities included in the second knowledge graph; Constructing sample entity pairs according to the centrality and uncertainty corresponding to the target entity; According to the sample entity pairs, training the entity alignment model to be trained; According to the trained entity alignment model, the knowledge graph to be entity aligned is entity aligned to obtain a knowledge graph fused after entity alignment, and the target task is performed according to the fused knowledge graph.

2. The method according to claim 1, characterized in that Filtering out a target entity from a plurality of entity information contained in the first knowledge graph includes: Determine the out-degree of each entity information in the plurality of entity information according to the connection relationship between the plurality of entity information included in the first knowledge graph; The plurality of entity information are screened according to the out-degree of each entity information in the plurality of entity information to obtain the target entity.

3. The method according to claim 2, characterized in that Determining the centrality corresponding to the target entity according to the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph includes: Determining the monomer centrality of the target entity according to the out-degree of the target entity; Determine the monomer centrality of the entity information corresponding to the neighboring nodes within a preset adjacent range of the node corresponding to the target entity in the first knowledge graph according to the out-degree of the entity information corresponding to the neighboring nodes; The centrality of the target entity is determined according to the monomer centrality of the target entity and the monomer centrality of the entity corresponding to the neighboring node.

4. The method according to claim 1, characterized in that Determining the uncertainty corresponding to the target entity according to the alignment probability between at least some entities included in the second knowledge graph and the target entity, and the alignment probability between the entity information corresponding to the neighbor node and at least some entities included in the second knowledge graph, including: Determining an alignment probability between the target entity and at least some of the entities included in the second knowledge graph; Determining a monomer uncertainty of the target entity according to an alignment probability between the target entity and at least some entities included in the second knowledge graph; Determine, according to the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph, an alignment probability between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph; Determining the monomer uncertainty of the entity information corresponding to the neighbor node according to the alignment probability between the entity information corresponding to the neighbor node and at least part of the entities included in the second knowledge graph; The uncertainty of the target entity is determined according to the individual uncertainty of the target entity and the individual uncertainty of the entity information corresponding to the neighboring node.

5. The method according to claim 4, characterized in that Determining the monomer uncertainty of the target entity according to the alignment probability between the target entity and at least some entities included in the second knowledge graph includes: Determining a probability difference of the alignment probability between the target entity and at least some of the entities included in the second knowledge graph according to the alignment probability between the target entity and at least some of the entities included in the second knowledge graph; Determining a monomer uncertainty of the target entity according to a probability difference of an alignment probability between the target entity and at least some entities included in the second knowledge graph; Determining the monomer uncertainty of the entity information corresponding to the neighbor node according to the alignment probability between the entity information corresponding to the neighbor node and at least part of the entities included in the second knowledge graph includes: Determine a probability difference between the alignment probabilities between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph according to the alignment probabilities between the entity information corresponding to the neighbor node and at least some of the entities included in the second knowledge graph; The monomer uncertainty of the entity information corresponding to the neighbor node is determined according to the probability difference between the entity information corresponding to the neighbor node and at least part of the entities included in the second knowledge graph.

6. The method according to claim 1, characterized in that According to the centrality and uncertainty corresponding to the target entity, a sample entity pair is constructed, including: Determining a centrality weight corresponding to the centrality of the target entity, and determining an uncertainty weight corresponding to the uncertainty of the target entity; Determining a representative value of the target entity according to the centrality corresponding to the target entity, the centrality weight, the uncertainty corresponding to the target entity, and the uncertainty weight; According to the representative value of the target entity, the target entity is screened, and according to the screened target entity, Construct sample entity pairs.

7. The method according to claim 6, characterized in that Determining a centrality weight corresponding to the centrality of the target entity, and determining an uncertainty weight corresponding to the uncertainty of the target entity, including: According to the difference between the centrality and the uncertainty corresponding to the target entity, the centrality weight corresponding to the centrality of the target entity is determined, and the uncertainty weight corresponding to the uncertainty of the target entity is determined.

8. The method according to claim 6, characterized in that Determining a centrality weight corresponding to the centrality of the target entity, and determining an uncertainty weight corresponding to the uncertainty of the target entity, including: According to the current training round of the entity alignment model, the centrality weight corresponding to the centrality of the target entity is determined, and the uncertainty weight corresponding to the uncertainty of the target entity is determined, wherein, the larger the current training round is, the smaller the centrality weight corresponding to the centrality of the target entity is, and the larger the uncertainty weight corresponding to the uncertainty of the target entity is.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

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

Images