Node task generation method, electronic device, readable storage medium and product

Abstract

The application discloses a node task generation method, an electronic device, a readable storage medium and a product, relates to the technical field of artificial intelligence, and comprises the following steps: constructing a task knowledge graph of a business scenario, wherein the task knowledge graph comprises task nodes and semantic relationship information between the task nodes; extracting a dependency weight matrix from the semantic relationship information between the task nodes by using a preset neural network model, wherein the dependency weight matrix is used for representing the dependency relationship between the task nodes; and performing sorting processing on the task nodes by using the dependency weight matrix, so as to obtain a target task flow of the business scenario. The application realizes automatic generation of the task flow, solves the problems of subjectivity and low efficiency in the task flow construction mode, and better meets actual business requirements.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a node-based task generation method, an electronic device, a readable storage medium and a product. BACKGROUND

[0002] At present, in many business scenarios, for example, government service scenarios and enterprise internal processes, it is mostly necessary to build standard task processes. Traditional task processes mostly rely on manual operation, for example, relevant personnel can build task processes through manual dragging on a zero-code business platform. However, because it is necessary to rely on manual understanding of actual business scenarios and building of processes, the efficiency of process building is relatively slow, and when business requirements or relevant systems frequently change, the manual building method cannot respond to changes in time. Therefore, the current traditional task process building method has the problem of being difficult to meet actual business requirements.

[0003] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide a node-based task generation method, an electronic device, a readable storage medium and a product, aiming to solve the technical problem that the current traditional task process building method is difficult to meet actual business requirements.

[0005] To achieve the above purpose, the present application provides a node-based task generation method, which comprises:

[0006] Building a task knowledge graph of a business scenario, wherein the task knowledge graph comprises task nodes and semantic relationship information between task nodes;

[0007] Extracting a dependency weight matrix from the semantic relationship information between the task nodes through a preset neural network model, wherein the dependency weight matrix is used to represent the dependency relationship between the task nodes;

[0008] Sorting processing the task nodes through the dependency weight matrix to obtain a target task process of the business scenario.

[0009] Optionally, the node-based task generation method further comprises:

[0010] Extracting entity relationships from new information of the business scenario and / or execution feedback of the target task process to obtain graph update content;

[0011] Updating the task knowledge graph through the graph update content to obtain a new task knowledge graph, and recording a task knowledge graph cumulative update amount.

[0012] In a case where the accumulated update amount of the task knowledge graph is greater than a preset accumulated threshold, returning to perform the step of extracting a dependency weight matrix from the semantic relationship information between each task node by using the preset neural network model based on the new task knowledge graph; or, based on the new task knowledge graph, performing incremental learning on the preset neural network model to obtain a new preset neural network model, and returning to perform the step of extracting a dependency weight matrix from the semantic relationship information between each task node by using the preset neural network model based on the new task knowledge graph and the new preset neural network model.

[0013] Optionally, the step of sorting each task node based on the dependency weight matrix to obtain a target task flow of the business scenario comprises:

[0014] sorting each task node based on the dependency weight matrix to generate a candidate task flow set;

[0015] filtering, from each candidate task flow of the task flow set, a candidate task flow without conflict as the target task flow.

[0016] Optionally, the dependency weight matrix comprises a dependency weight between any two task nodes in each task node, and the step of sorting each task node based on the dependency weight matrix to generate a candidate task flow set comprises:

[0017] filtering out a dependency weight less than a preset threshold in the dependency weight matrix to obtain a target dependency weight matrix;

[0018] constructing each directed task node sequence composed of each task node based on each dependency weight in the target dependency weight matrix;

[0019] determining a candidate task flow based on each directed task node sequence, and combining each candidate task flow into the candidate task flow set.

[0020] Optionally, the step of determining a candidate task flow based on each directed task node sequence comprises:

[0021] detecting any directed task node sequence;

[0022] in a case where the result of the detection is that the directed task node sequence is a ring sequence, disconnecting a position corresponding to a lowest dependency weight in the directed task node sequence to form a directed acyclic task node sequence;

[0023] Sort the directed acyclic task node sequence based on a topology algorithm to obtain a candidate task flow.

[0024] Optionally, the step of screening the candidate task flow without conflict from the candidate task flows in the task flow set as the target task flow comprises:

[0025] Detecting conflict between each candidate task flow in the candidate task flow set through a preset task constraint rule;

[0026] Screening the candidate task flow without conflict based on the detection result of the conflict detection as the target task flow.

[0027] Optionally, the step of extracting the dependency weight matrix from the semantic relationship information between each task node through a preset neural network model comprises:

[0028] Inputting the task knowledge graph including the semantic relationship information into the preset neural network model to obtain the dependency weight matrix of each task node, wherein the preset neural network model is used to aggregate the adjacent node features of each task node in the task knowledge graph according to the semantic relationship information, and identify the dependency weight matrix based on the adjacent node features of each task node.

[0029] In addition, to achieve the above-mentioned purposes, the present application also proposes an electronic device, which comprises a memory, a processor and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the node task generation method as described above.

[0030] In addition, to achieve the above-mentioned purposes, the present application also proposes a storage readable storage medium, which is a computer readable storage medium, and the storage readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the node task generation method as described above.

[0031] In addition, to achieve the above-mentioned purposes, the present application also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the node task generation method as described above.

[0032] The one or more technical solutions proposed in the present application have at least the following technical effects:

[0033] In the embodiment of the present application, a task knowledge graph of a business scenario is constructed, wherein the task knowledge graph comprises task nodes and semantic relationship information between the task nodes; a dependency weight matrix is extracted from the semantic relationship information between the task nodes by a preset neural network model, wherein the dependency weight matrix is used to represent the dependency relationship between the task nodes; and the task nodes are sorted by the dependency weight matrix to obtain a target task process of the business scenario. That is, in the embodiment, the original semantic relationship information between the task nodes in the business scenario is reflected by constructing a task knowledge graph, and the dependency weight matrix between the task nodes is extracted from the semantic relationship information included in the task knowledge graph by a preset neural network model, so as to quantify the hidden dependency relationship between the task nodes. The target task process can be built based on the dependency weight matrix and the task nodes. Thus, the automatic generation of the task process is realized, the problems of subjectivity and inefficiency in the task process construction mode are solved, and the actual business requirements are better met. BRIEF DESCRIPTION OF DRAWINGS

[0034] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor.

[0036] Figure 1 A flowchart of a first embodiment of the node-based task generation method of the present application;

[0037] Figure 2 A flowchart of a second embodiment of the node-based task generation method of the present application;

[0038] Figure 3 A flowchart of a third embodiment of the node-based task generation method of the present application;

[0039] Figure 4 A flowchart of the overall process of the node-based task generation method of the present application;

[0040] Figure 5 A device structure diagram of the hardware running environment involved in the node-based task generation method in the embodiment of the present application.

[0041] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are merely exemplary of the application and do not limit the application.

[0043] In order to better understand the technical solutions of the present application, the following will be described in detail in conjunction with the drawings and specific embodiments.

[0044] At present, in many business scenarios, for example, government service scenarios and enterprise internal processes, it is often necessary to build standard task processes. Traditional task processes mostly rely on manual operation, for example, relevant personnel can build task processes through manual drag-and-drop on a zero-code business platform. However, since it is necessary to rely on manual understanding of actual business scenarios and construction of processes, the efficiency of process construction is relatively slow, and when business requirements or related systems change frequently, manual construction methods cannot respond to changes in time. Therefore, the current traditional task process construction method has the problem of being difficult to meet actual business requirements.

[0045] The main solution of the embodiment of the present application is to construct a task knowledge graph of a business scenario, wherein the task knowledge graph includes task nodes and semantic relationship information between the task nodes; a dependency weight matrix is extracted from the semantic relationship information between the task nodes by a preset neural network model, wherein the dependency weight matrix is used to represent the dependency relationship between the task nodes; and each task node is sorted by the dependency weight matrix to obtain a target task process of the business scenario.

[0046] That is, in the present application, the task knowledge graph is constructed to reflect the original relationship between the task nodes in the business scenario, and the dependency weight matrix between the task nodes is extracted from the task knowledge graph by a preset neural network model, so as to quantify the hidden dependency relationship between the task nodes. Based on the dependency weight matrix and the task nodes, a target task process can be built. Thus, the automatic generation of the task process is realized, the problems of subjectivity and inefficiency in the task process construction method are solved, and the actual business requirements are better met.

[0047] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a server, a computer, a mobile phone, etc., or an electronic device capable of realizing the above functions.

[0048] Based on the above introduction, the embodiment of the present application provides a node task generation method, which refers to Figure 1 The flowchart of the first embodiment of the node task generation method of the present application is shown in the figure.

[0049] In this embodiment, the node task generation method includes steps S10-S30:

[0050] Step S10, constructing a task knowledge graph of the business scenario, wherein the task knowledge graph includes task nodes and semantic relationship information between the task nodes;

[0051] It should be noted that the present embodiment mainly replaces manual generation by automatically generating a task flow, so the execution subject of the above node task generation method is usually a server. In addition, the specific type of the above business scenario is determined by actual demand, for example, the business scenario can be a government service scenario, or a business service for internal and external enterprises, etc. In order to facilitate description, the government service scenario will be taken as an example in the subsequent content.

[0052] For example, in this embodiment, the task knowledge graph corresponding to the business scenario can be constructed based on the Neo4j graph database, and in the task knowledge graph, each task node (i.e. entity) and semantic relationship information (i.e. edge) between the task nodes can be included. The attributes of the task node can include node type, state, and update time, etc., and the semantic relationship information can be a dependency or a precondition. It should be noted that the semantic relationship information can refer to the semantic relationship between two task nodes, or a set of semantic relationships between any two adjacent nodes based on actual demand.

[0053] Optionally, the Schema (i.e. metadata framework, entity-edge-entity) can be created by the Cypher language (i.e. query language of the graph database) of the Neo4j graph database to build the task knowledge graph corresponding to the business scenario. For example, in the government service scenario, the types of the task nodes can include policy provisions, approval departments, material lists, etc. An example of combining the task nodes and the semantic relationship information can be that the policy provisions affect the approval departments, the approval departments require the material lists, the material lists are preceded by the business affairs, and the business affairs are applicable to the reporting subjects, etc. In addition, the dynamic attributes of the task nodes can be marked by the node attribute last_updated to update the time, so as to support the dynamic incremental update of the knowledge graph, thereby adapting to the dynamic changes of demand and policy.

[0054] Step S20, extracting a dependency weight matrix from the semantic relationship information between the task nodes by a preset neural network model, wherein the dependency weight matrix is used to represent the dependency relationship between the task nodes;

[0055] Exemplarily, after the knowledge graph is constructed, the knowledge graph can be input into a preset neural network model, and a dependency weight matrix of each task node is extracted from the semantic relationship information included in the task knowledge graph by the preset neural network model. The preset neural network model can be a graph neural network (GNN, Graph Neural Network), and in a feasible implementation, an initial graph neural network can be constructed by PyTorch Geometric, and the initial model is trained to obtain the above-mentioned preset neural network model. It can be understood that the above-mentioned preset neural network model is mainly used to extract the dependency relationship between any two task nodes from the task knowledge graph, and the strength of the dependency relationship can be represented by the dependency weight, for example, the greater the dependency weight between two task nodes, the stronger the dependency relationship between the two task nodes, so the dependency weight matrix can represent the dependency relationship between the task nodes.

[0056] In step S30, each task node is sorted by the dependency weight matrix to obtain a target task process of the business scenario.

[0057] Exemplarily, by the dependency weight matrix, the dependency relationship between each task node can be determined, and each task node is sorted according to the direction and size of the dependency relationship to construct a directed acyclic graph (e.g., DAG, Directed Acyclic Graph). For example, if the dependency weight of task node A to node B is 0.9 (strong dependency), and the dependency weight of node B to node C is 0.8, then A→B→C can constitute a directed acyclic graph. It can be understood that the arrangement order of each task node in the directed acyclic graph can constitute a task process, and the task process can also be used as a target task process.

[0058] In this embodiment, a task knowledge graph of a business scenario is constructed, where the task knowledge graph includes task nodes and semantic relationship information between the task nodes; a dependency weight matrix is extracted from the semantic relationship information between the task nodes by a preset neural network model, where the dependency weight matrix is used to represent the dependency relationship between the task nodes; and each task node is sorted by using the dependency weight matrix to obtain a target task process of the business scenario. That is, in this embodiment, the original semantic relationship information between the task nodes in the business scenario is reflected by constructing the task knowledge graph, and the dependency weight matrix between the task nodes is extracted from the semantic relationship information included in the task knowledge graph by using the preset neural network model, so that the hidden dependency relationship between the task nodes is quantified. The target task process can be built based on the dependency weight matrix and the task nodes. Therefore, the automatic generation of the task process is realized, the problems of subjectivity and inefficiency in the task process construction mode are solved, and the actual business requirements are better met.

[0059] In a feasible implementation, the step of extracting the dependency weight matrix of each task node from the task knowledge graph by using the preset neural network model includes step S21:

[0060] In step S21, the task knowledge graph including the semantic relationship information is input into the preset neural network model to obtain the dependency weight matrix of each task node, where the preset neural network model is used to aggregate the adjacent node features of each task node in the task knowledge graph according to the semantic relationship information, and identify the dependency weight matrix based on the adjacent node features of each task node.

[0061] For example, the task knowledge graph including the semantic relationship information between the task nodes can be input into the preset neural network model (for example, the node feature matrix and the adjacent matrix in the knowledge graph can be input into the model), and the preset neural network model outputs the dependency weight matrix of each task node. Wherein, the node feature matrix is used to describe the attribute set of each node in the graph, for example, in a feasible implementation, the node feature matrix can be a two-dimensional matrix of [node number x feature dimension], and each node corresponds to a row and each feature corresponds to a column. The adjacent matrix can be a [node number x node number] square matrix describing the specific connection relationship between the nodes, which can be constructed by using the semantic relationship information included in the task knowledge graph.

[0062] Optionally, in the preset neural network model, a message passing mechanism can be used, and the semantic relationship information included in the task knowledge graph is used to aggregate the adjacent node features of each task node, and then the GCNConv or GraphSAGE layer in the preset neural network model is used to learn the node embedding (i.e., the aggregated adjacent node features) to output the dependency weight matrix of the task node, and the higher the weight value between two task nodes in the matrix, the stronger the dependency between the two nodes. In a feasible implementation, the dependency weight between two task nodes can be obtained by calculating the similarity between the adjacent node features of the two task nodes, and used to construct the dependency weight matrix.

[0063] Referring to Figure 2 For the first embodiment of the node-based task generation method based on the present application, the flowchart of the second embodiment of the node-based task generation method of the present application is presented. In this embodiment, the same or similar content as the above embodiment can be referred to the above introduction, and will not be described in detail. The node-based task generation method further includes steps S1-S3:

[0064] Step S1, extracting entity relationships from the new information of the business scenario and / or the execution feedback of the target task flow to obtain graph update content;

[0065] Step S2, updating the task knowledge graph by the graph update content to obtain a new task knowledge graph, and recording the cumulative update amount of the task knowledge graph;

[0066] Step S3, in the case where the cumulative update amount of the task knowledge graph is greater than a preset cumulative threshold, returning to execute the step of extracting the dependency weight matrix from the semantic relationship information between the task nodes by the preset neural network model based on the new task knowledge graph; or, based on the new task knowledge graph, incrementally learning the preset neural network model to obtain a new preset neural network model, and based on the new task knowledge graph and the new preset neural network model, returning to execute the step of extracting the dependency weight matrix from the semantic relationship information between the task nodes by the preset neural network model.

[0067] It should be noted that in actual application, the content in the task knowledge graph may need to be dynamically changed according to actual needs.

[0068] For example, the business scenario may have new information, for example, in the government service scenario, the relevant service policy may change, and the new information can be the changed policy or the newly added policy. During the execution of the generated target task flow, the personnel performing the execution can also give the corresponding feedback, i.e., the execution feedback, such as the flow is unreasonable, the reason why it is unreasonable, etc.

[0069] Exemplarily, the entity relationship can be extracted from the new information and / or the execution feedback, so as to obtain the graph update content. Optionally, the new information can be extracted by an ALBERT-BiLSTM-CRF model. Taking the new information as an example, if the new information is a new policy text, the ALBERT-BiLSTM-CRF model can be used to encode the new policy text to generate a context vector, while the legality of the entity label is constrained, so as to extract the entity relationship in the new policy text, and obtain the graph update content. According to the graph update content, the task knowledge graph is updated to obtain a new task knowledge graph. Optionally, the APOC plug-in (Awesome Procedures On Cypher, which is a third-party extension library of the Neo4j graph database) can be used to incrementally update the graph to obtain the new task knowledge graph. At the same time, the cumulative update amount of the task knowledge graph is recorded. The cumulative update amount of the task knowledge graph can be obtained by recording the graph update content of the knowledge graph each time the knowledge graph is updated, and then quantitatively accumulating the update content. The cumulative update amount of the task knowledge graph is a judgment parameter for triggering the subsequent steps.

[0070] Exemplarily, the cumulative amount of the task knowledge graph can be compared with a preset cumulative threshold. If the cumulative amount of the task knowledge graph is greater than the preset cumulative threshold, the step of returning to execute the extraction of the dependency weight matrix of each task node from the task knowledge graph by the preset neural network model based on the new task knowledge graph can be triggered, and the cumulative amount of the task knowledge graph is reset to zero at the same time. Alternatively, the preset neural network model is incrementally learned based on the new task knowledge graph, to obtain a new preset neural network model. For example, in a feasible implementation, part of the model parameters in the preset neural network model can be fixed or isolated, and then the preset neural network model with the fixed or isolated parameters is trained using the new task knowledge graph, so as to obtain a new preset neural network model. Based on the new task knowledge graph and the new preset neural network model, the step of returning to execute the extraction of the dependency weight matrix of each task node from the task knowledge graph by the preset neural network model can be executed. Optionally, when the new information is the applicable version or the applicable time limit of a certain node, the target task process can also be directly regenerated, that is, the step of returning to execute the extraction of the dependency weight matrix from the semantic relationship information between the task nodes by the preset neural network model based on the new task knowledge graph.

[0071] It can be understood that in the present embodiment, the knowledge graph can be automatically updated according to the newly added information of the business scenario and / or the execution feedback of the target task process. At the same time, the cumulative update amount of the knowledge graph is recorded, and when the cumulative update amount reaches a certain degree, the preset neural network model is triggered to re-extract the dependency weight matrix, or the preset neural network model is triggered to incrementally learn, and then the dependency weight is re-extracted, and finally the target task process is regenerated. Thus, the target process is matched with the dynamic scene demand, and the actual business demand is met.

[0072] Referring to Figure 3 For the first and second embodiments of the node-based task generation method of the present application, the flowchart of the third embodiment of the node-based task generation method of the present application is presented. In the present embodiment, the same or similar contents as the above embodiments can be referred to the above introduction, and will not be described in detail. The steps of generating the target task process based on the dependency weight matrix and each task node include steps S31-S32:

[0073] Step S31, sorting each task node based on the dependency weight matrix, and generating a candidate task process set;

[0074] Step S32, screening a conflict-free candidate task process from each candidate task process in the task process set as the target task process.

[0075] For example, the task nodes with dependency relationship can be connected and sorted according to the direction of the dependency relationship between the task nodes, thereby forming a task node sequence and obtaining a candidate task process. The candidate task process is used to form a candidate task process set. It can be understood that the candidate task process set obtained in the usual case will include multiple candidate task processes. Therefore, it is necessary to screen the target task process from the task process set, such as screening a conflict-free task process from the task process set as the target task process.

[0076] In a feasible implementation, the dependency weight matrix includes the dependency weight between any two task nodes in the task nodes, and the step of sorting each task node based on the dependency weight matrix and generating a candidate task process set includes steps S311-S313:

[0077] Step S311, screening out the dependency weight less than the preset threshold in the dependency weight matrix to obtain a target dependency weight matrix;

[0078] Step S312, constructing each directed task node sequence composed of each task node based on each dependency weight in the target dependency weight matrix;

[0079] Step S313, determining a candidate task process based on each directed task node sequence, and combining each candidate task process into a candidate task process set.

[0080] For example, in order to ensure the quality of the candidate task flow set, the dependency weights in the dependency weight matrix can be screened. That is, the dependency weights in the dependency weight matrix that are less than a preset threshold can be screened out to obtain a target dependency weight matrix. For example, the dependency weights less than the preset threshold θ (θ = 0.7) are screened out. It should be noted that screening out a dependency weight also means that the dependency relationship between the two task nodes corresponding to the dependency weight is discarded. In this way, a part of the task node pairs with weak dependency relationships can be ignored, and the quality of the candidate task flow set can be ensured. Further, based on the dependency weights in the target dependency weight matrix, each directed task node sequence composed of task nodes can be constructed, such as connecting the task nodes with dependency relationships to form a directed task node sequence. The sequence direction of the directed task node sequence also represents the dependency direction between the task nodes. Alternatively, a directed task node sequence corresponds to a candidate task flow, and each candidate task flow can constitute a candidate task flow set.

[0081] In a feasible implementation, the step of determining the candidate task flow based on each directed task node sequence includes steps S301-S303:

[0082] Step S301, detecting any directed task node sequence;

[0083] Step S302, in the case where the detection result is that the directed task node sequence is a loop sequence, disconnecting the position corresponding to the lowest dependency weight in the directed task node sequence to form a directed acyclic task node sequence;

[0084] Step S303, sorting the directed acyclic task node sequence based on a topological algorithm to obtain a candidate task flow.

[0085] It should be noted that since the processing process for each directed task node is similar, in this embodiment, a directed task node sequence is taken as an example for description, and other directed task node sequences are processed accordingly.

[0086] For any one of the directed task nodes, the directed task node sequence can be detected by a preset detection algorithm (such as a depth-first search algorithm). It can be understood that the detection algorithm is mainly used to ensure that the directed task node sequence is a linear execution sequence, while ensuring that all dependency relationships are strictly followed. Among them, the detection of the directed task node sequence by the detection algorithm is mainly to judge whether the directed task node sequence is a loop sequence, for example, marking the state of the task node (not visited, in visit, completed) when traversing. If a node with a state of "in visit" is encountered in the traversal process, it means that there is a back edge, that is, a loop is found. Correspondingly, the detection result can be that the directed task node sequence is a loop sequence, or the directed task node sequence is not a loop sequence. Among them, if the detection result is that the directed task node sequence is not a loop sequence, then the directed task node sequence can actually be a directed acyclic task node sequence, so the subsequent sorting of the candidate task process can be directly based on the directed task node sequence. Conversely, if the detection result is that the directed task node sequence is a loop sequence, then the position corresponding to the lowest dependency weight in the directed task node sequence can be selected to break, thereby forming a directed acyclic task node sequence. Then, the directed acyclic task node sequence obtained is sorted based on the topology algorithm, thereby obtaining the candidate task process. Alternatively, the sorting process based on the topology algorithm can be as follows:

[0087] 1. Calculate the in-degree of each node in the directed acyclic task node sequence. Among them, the in-degree refers to the number of edges pointing to a certain node, for example, if node C depends on node A and node B, then the in-degree of node C is 2.

[0088] 2. Initialize a queue (or stack) and add all nodes with an in-degree of 0 to it, where the in-degree of 0 can be considered as the "starting point" of the task process.

[0089] 3. Take a node (such as node U) from the queue. Output this node U to the topology sequence.

[0090] 4. "Simulate" the completion of the task, that is, traverse all neighbor nodes of node U. Reduce the in-degree of these neighbor nodes V by 1. If the in-degree of a neighbor node V is thus reduced to 0, then V is added to the queue.

[0091] 5. Loop the above steps until the queue is empty. Check whether the topology sequence contains all nodes. If the topology sequence contains all nodes, the sorting is successful, that is, the candidate task process is obtained.

[0092] It can be understood that by the above method, the problem of the candidate task process appearing in a dead loop can be avoided, and the quality of the candidate task process is also guaranteed.

[0093] In an implementable embodiment, the step of screening out the conflict-free candidate task flow from the candidate task flows in the candidate task flow set as the target task flow comprises steps S321-S322:

[0094] In step S321, conflict detection is performed between the candidate task flows in the candidate task flow set according to the preset task constraint rule.

[0095] In step S322, the conflict-free candidate task flow is screened out based on the detection result of the conflict detection as the target task flow.

[0096] It should be noted that after the candidate task flow set is generated, the candidate task flows in the candidate task flow set can be further screened.

[0097] For example, the candidate task flows in the candidate task flow set can be subjected to conflict detection according to the preset task constraint rule. For example, in the scenario of government service, the preset task constraint rule can include the policy effective period. Accordingly, in the screening process, if the policy effective period corresponding to the candidate task flow in the candidate task flow set does not match the current time, it can be considered that the detection result of the conflict detection is conflict-free, and accordingly, the conflict-free candidate task flow can be used as the target task flow.

[0098] It can be understood that after the candidate task flows are screened by the above conflict detection, the effectiveness of the finally obtained target task flow can be further ensured.

[0099] Referring to Figure 4 The overall flowchart of the node task generation method of the present application is shown in FIG. 1. As shown in the figure, knowledge modeling is performed by Neo4j to obtain a knowledge graph (i.e., the task knowledge graph described above). The dependency weight matrix is extracted from the knowledge graph using a preset neural network model. After the dependency weights with low weights in the dependency weight matrix are screened out, the task flow is obtained in combination with topological sorting. After the task flow is executed, execution feedback can be generated for graph incremental updating, and the content of real-time policy change can also be used for graph incremental updating. The graph incremental updating can obtain a new knowledge graph and trigger retraining. The retraining refers to retraining of the preset neural network model. After the retraining, the new dependency weight matrix can be extracted from the updated knowledge graph based on the new preset neural network model obtained by the retraining, so that the finally generated task flow matches the actual scene demand.

[0100] Reference will now be made to Figure 5 which shows a structural schematic diagram of an electronic device suitable for implementing the embodiments of the present application. Figure 5The electronic device shown is merely an example and should not impose any limitation on the functions and usage range of the embodiments of the present application.

[0101] As shown in Figure 5 The electronic device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. Various programs and data required for operation of the electronic device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the electronic device having various systems is shown in the figure, it should be understood that all the systems shown are not required to be implemented or possessed. More or less systems can be alternatively implemented or possessed.

[0102] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable storage medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are performed.

[0103] The electronic device provided in the present application adopts the node task generation method in the above embodiments, and can solve the technical problem that the current traditional task flow construction method cannot meet the actual business requirements. Compared with the prior art, the electronic device provided in the present application has the same beneficial effects as the node task generation method provided in the above embodiments, and other technical features in the electronic device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0104] It should be understood that various parts of the present application can be realized in hardware, software, firmware, or a combination thereof. In the above description of embodiments, specific functional, structural, material or characteristic features are combined in a manner appropriate for the particular example or embodiment. However, each feature can also be provided separately or in any appropriate sub-combination.

[0105] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of the changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0106] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer programs) for performing the node-wise task generation method in the above embodiments.

[0107] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM or flash memory), an optical fiber, a portable compact disk read only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination of the above. In the present embodiment, the computer readable storage medium can be any tangible readable storage medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer readable storage medium can be transmitted in any appropriate readable storage medium, including but not limited to an electrical wire, an optical cable, an RF (Radio Frequency), etc., or any appropriate combination of the above.

[0108] The above computer readable storage medium can be contained in an electronic device; or can exist separately without being assembled into an electronic device.

[0109] The above computer readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to:

[0110] construct a task knowledge graph of the business scenario, wherein the task knowledge graph comprises task nodes and semantic relationship information between the task nodes;

[0111] extract a dependency weight matrix of each task node from the semantic relationship information comprised in the task knowledge graph by a preset neural network model, wherein the dependency weight matrix is used to represent the dependency relationship between the task nodes;

[0112] sort the task nodes by the dependency weight matrix to obtain a target task flow.

[0113] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0114] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a procedure, or a part of code, which comprises one or more executable instructions for implementing the specified functions. It should also be noted that, in some alternative implementations, the functions noted in the blocks can occur in a different order than that noted in the figure. For example, two blocks noted in succession can in fact be executed substantially concurrently or in the opposite order, depending on the functionality involved. It should also be noted that each block in the block diagrams and / or flow diagrams, and combinations of blocks in the block diagrams and / or flow diagrams, can be implemented by dedicated hardware-based systems that perform the specified functions or operations, or can be implemented by a combination of dedicated hardware-based systems and computer instructions.

[0115] The modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0116] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the above-mentioned node task generation method, and can solve the technical problem that the current traditional task flow construction method is difficult to meet the actual business requirements. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the node task generation method provided by the above-mentioned embodiments, and will not be described here.

[0117] The present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the node task generation method as described above.

[0118] The computer program product provided by the present application can solve the technical problem that the current traditional task flow construction method is difficult to meet the actual business requirements. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the node task generation method provided by the above-mentioned embodiments, and will not be described here.

[0119] The above is only some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A node-based task generation method, characterized in that, The node-based task generation method includes: Construct a task knowledge graph for business scenarios, wherein the task knowledge graph includes each task node and the semantic relationship information between task nodes; A dependency weight matrix is ​​extracted from the semantic relationship information between the task nodes using a preset neural network model, wherein the dependency weight matrix is ​​used to characterize the dependency relationship between the task nodes. The task nodes are sorted using the dependency weight matrix to obtain the target task flow of the business scenario. The node-based task generation method further includes: Entity relationships are extracted from the new information in the business scenario and / or the execution feedback of the target task process to obtain the updated graph content; The task knowledge graph is updated by updating the graph content to obtain a new task knowledge graph, and the cumulative update amount of the task knowledge graph is recorded. If the cumulative update amount of the task knowledge graph exceeds a preset cumulative threshold, the step of extracting the dependency weight matrix from the semantic relationship information between the task nodes using a preset neural network model is returned based on the new task knowledge graph; or, the preset neural network model is incrementally learned based on the new task knowledge graph to obtain a new preset neural network model, and the step of extracting the dependency weight matrix from the semantic relationship information between the task nodes using a preset neural network model is returned based on the new task knowledge graph and the new preset neural network model.

2. The node-based task generation method as described in claim 1, characterized in that, The step of sorting each task node using the dependency weight matrix to obtain the target task flow of the business scenario includes: Based on the dependency weight matrix, each task node is sorted to generate a candidate task process set; From the candidate task flows in the task flow set, the candidate task flows without conflicts are selected as the target task flow.

3. The node-based task generation method as described in claim 2, characterized in that, The dependency weight matrix includes the dependency weights between any two task nodes in each task node, and the step of sorting the task nodes based on the dependency weight matrix to generate a candidate task process set includes: The dependency weights in the dependency weight matrix that are less than a preset threshold are filtered out to obtain the target dependency weight matrix; Based on the dependency weights in the target dependency weight matrix, construct a sequence of directed task nodes composed of the task nodes. Candidate task flows are determined based on the sequence of each directed task node, and the candidate task flows are combined to form the candidate task flow set.

4. The node-based task generation method as described in claim 3, characterized in that, The step of determining the candidate task process based on each of the directed task node sequences includes: Detect any of the directed task node sequences; If the detection result indicates that the directed task node sequence is a circular sequence, the position corresponding to the lowest dependency weight in the directed task node sequence is disconnected to form a directed acyclic task node sequence. The directed acyclic task node sequence is sorted using a topology algorithm to obtain candidate task flows.

5. The node-based task generation method as described in claim 2, characterized in that, The step of selecting conflict-free candidate task flows from each candidate task flow in the task flow set as the target task flow includes: Conflict detection is performed between the candidate task processes in the candidate task process set by using preset task constraint rules. Based on the conflict detection results, conflict-free candidate task flows are selected as the target task flow.

6. The node-based task generation method as described in claim 1, characterized in that, The step of extracting the dependency weight matrix from the semantic relationship information between the task nodes using a preset neural network model includes: The task knowledge graph, including the semantic relationship information, is input into the preset neural network model to obtain the dependency weight matrix of each task node. The preset neural network model is used to aggregate the neighbor node features of each task node in the task knowledge graph based on the semantic relationship information, and to identify the dependency weight matrix based on the neighbor node features of each task node.

7. An electronic device, characterized in that, The electronic device includes: a processor, a memory, and a node-based task generation program stored in the memory and executable on the processor, wherein the node-based task generation program, when executed, implements the steps of the node-based task generation method as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a node-based task generation program, which, when executed, implements the steps of the node-based task generation method as described in any one of claims 1-6.

9. A computer program product, characterized in that, The computer program product includes a node-based task generation program, which, when executed by a processor, implements the steps of the node-based task generation method as described in any one of claims 1-6.

Images