Business process management method and system, electronic equipment and storage medium

Abstract

The invention discloses a business process management method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring target user information and business process application information; performing data preprocessing on the target user information and the business process application information to obtain business feature information; inputting the service feature information into a preset neural network model for identity prediction to obtain identity recommendation information; obtaining user selection data, and determining target identity information from the identity recommendation information through the user selection data; and dynamically constructing approval process data according to the target identity information. According to the embodiment of the invention, the efficiency of business process management can be effectively improved, and the accuracy and safety of process management are improved. The method can be widely applied to the field of informatization management.

Description

Technical Field

[0001] The present application relates to the field of information management technology, and in particular to a business process management method, system, electronic device and storage medium. Background Art

[0002] In some business processes, such as travel approval processes, relevant personnel are often required to manually fill out business trip applications, collect reimbursement vouchers, and go through multiple levels of approval. The process is cumbersome and prone to errors. In addition, paper receipts are easy to lose and fade, which also increases the difficulty of reimbursement. With the expansion of enterprise scale and the diversification of business, relevant personnel often need to schedule work and travel activities between different departments, subsidiaries or affiliated companies, and different business scenarios perform different business processes. In related technologies, business process management efficiency is low, and it is difficult to achieve accurate and secure process management.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a business process management method, system, electronic device and storage medium, which can effectively improve the efficiency of business process management and improve the accuracy and security of process management.

[0005] To achieve the above object, an embodiment of the present application provides a business process management method, which includes the following steps:

[0006] Obtain target user information and business process application information;

[0007] Performing data preprocessing on the target user information and the business process application information to obtain business feature information;

[0008] Inputting the service feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information;

[0009] Acquiring user selection data to determine target identity information from the identity recommendation information through the user selection data;

[0010] The approval process data is dynamically constructed according to the target identity information.

[0011] In some embodiments, obtaining target user information and business process application information includes:

[0012] Acquire the identity status information of the target object to obtain the target user information; wherein the target user information includes department information, rank information and business standard information;

[0013] Acquire the business process application information; wherein the business process application information includes time information, location information and historical business data.

[0014] In some embodiments, the data preprocessing of the target user information and the business process application information to obtain business feature information includes:

[0015] Performing data segmentation on the target user information and the business process application information to obtain target feature data; wherein the target feature data includes numerical feature data, category feature data and time feature data;

[0016] Performing feature normalization processing on the numerical feature data to obtain first feature data;

[0017] The category feature data is embedded and encoded by a hash feature algorithm to obtain second feature data;

[0018] Performing one-hot encoding on the time feature data to obtain third feature data;

[0019] The first feature data, the second feature data, and the third feature data are combined to obtain the service feature information.

[0020] In some embodiments, before inputting the service feature information into a preset neural network model for identity prediction to obtain identity recommendation information, the method further includes:

[0021] Constructing an initial neural network model; wherein the initial neural network model includes an attention mechanism layer, and the attention mechanism layer is used to determine feature weight data;

[0022] Constructing a preset data set; wherein the preset data set includes input feature data and identity tag information, and the input feature data corresponds to the identity tag information;

[0023] According to the preset data set, the cross entropy loss function is used as the model optimization target to perform model training on the initial neural network model to obtain the preset neural network model.

[0024] In some embodiments, inputting the service feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information includes:

[0025] Inputting the service feature information into an input layer of a preset neural network model to perform feature calculation to obtain fourth feature data;

[0026] Adding the fourth feature data to the embedding layer for averaging to obtain fifth feature data; wherein the embedding layer is connected to the input layer in a fully connected manner;

[0027] Inputting the fifth feature data into the first hidden layer to extract feature association information through a first activation function to obtain a first association feature;

[0028] Input the first associated feature into the attention mechanism layer to perform weight distribution calculation to obtain a weighted feature vector;

[0029] Inputting the weighted feature vector into a second hidden layer to extract feature association information through a second activation function to obtain a second association feature;

[0030] The second association feature is input into the output layer to output the identity recommendation information through a first normalization function; wherein the identity recommendation information includes recommendation category information and recommendation index information, and the recommendation index information corresponds to the recommendation category information.

[0031] In some embodiments, the step of inputting the first associated feature into the attention mechanism layer for weight distribution calculation to obtain a weighted feature vector includes:

[0032] Performing dot product calculation on each of the first associated features to obtain preset importance score data;

[0033] Normalizing the preset importance score data by a second normalization function to obtain feature weight data;

[0034] The feature weight data is weighted and summed to obtain the weighted feature vector.

[0035] In some embodiments, after performing the step of acquiring the user selection data to determine the target identity information from the identity recommendation information through the user selection data, the method further includes:

[0036] Parameters of the preset neural network model are optimized through a preset reward mechanism according to the target identity information and the identity recommendation information.

[0037] To achieve the above object, another aspect of the embodiment of the present application provides a business process management system, the system comprising:

[0038] The first module is used to obtain target user information and business process application information;

[0039] The second module is used to perform data preprocessing on the target user information and the business process application information to obtain business feature information;

[0040] The third module is used to input the business feature information into a preset neural network model to perform identity prediction and obtain identity recommendation information;

[0041] A fourth module is used to obtain user selection data, so as to determine target identity information from the identity recommendation information through the user selection data;

[0042] The fifth module is used to dynamically construct approval process data according to the target identity information.

[0043] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising:

[0044] at least one processor;

[0045] at least one memory for storing at least one program;

[0046] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0047] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0048] The embodiments of the present application include at least the following beneficial effects: The present application provides a business process management method, system, electronic device and storage medium. The scheme obtains target user information and business process application information, performs data preprocessing on the target user information and business process application information, obtains business feature information, and then inputs the business feature information into a preset neural network model for identity prediction to obtain identity recommendation information. Then, the embodiment of the present invention obtains user selection data to determine the target identity information from the identity recommendation information through the user selection data, and then dynamically constructs the approval process data according to the target identity information, thereby realizing business process management. It is easy to understand that the embodiment of the present invention obtains business feature information through the target user information and business process application information extraction, and inputs the preset neural network model to obtain the corresponding identity recommendation information by performing identity prediction through the preset neural network model, and then determines the target identity information in combination with the user selection data, so that the corresponding approval process data can be dynamically constructed according to the target identity information, effectively improving the efficiency of business process management, and improving the accuracy and security of process management. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flow chart of a business process management method provided by an embodiment of the present invention;

[0050] Figure 2is a flow chart of obtaining target user information and business process application information provided by an embodiment of the present invention;

[0051] Figure 3 It is a flow chart of performing data preprocessing on target user information and business process application information to obtain business feature information provided by an embodiment of the present invention;

[0052] Figure 4 is a flow chart of constructing a preset neural network model provided by an embodiment of the present invention;

[0053] Figure 5 is a schematic diagram of the structure of a preset neural network model provided by an embodiment of the present invention;

[0054] Figure 6 It is a flow chart of inputting business feature information into a preset neural network model for identity prediction to obtain identity recommendation information provided by an embodiment of the present invention;

[0055] Figure 7 It is a flowchart of inputting the first correlation feature into the attention mechanism layer to perform weight distribution calculation to obtain a weighted feature vector provided by an embodiment of the present invention;

[0056] Figure 8 is a flow chart of optimizing parameters of a preset neural network model provided by an embodiment of the present invention;

[0057] Fig. 9 This is a flow chart of dynamic approval of mobile terminal business processes provided by an embodiment of the present invention;

[0058] Fig.10 It is a schematic diagram of the structure of the business process management system provided by the embodiment of the present application;

[0059] Fig.11 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0061] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0062] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0063] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0064] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0065] Business travel refers to the related activities of "air tickets, train tickets, car use, and accommodation" when corporate employees travel.

[0066] Approval flow: refers to the approval process that corporate employees go through based on their identities before traveling. After the process is completed, an approval form can be generated for business travel planning and control.

[0067] Dynamic identity recognition: refers to the system intelligently identifying the optimal identity through a deep learning model based on the employee's behavioral data, authority level, and historical usage, in order to match the most appropriate approval process.

[0068] Fully Connected Neural Network (FCNN): is a basic neural network structure, also known as Multilayer Perceptron (MLP), which consists of multiple fully connected layers (also called dense layers).

[0069] In some business processes, such as travel approval processes, relevant personnel are often required to manually fill out business trip applications, collect reimbursement vouchers, and go through multiple levels of approval. The process is cumbersome and prone to errors. In addition, paper receipts are easy to lose and fade, which also increases the difficulty of reimbursement. With the expansion of enterprise scale and the diversification of business, relevant personnel often need to schedule work and travel activities between different departments, subsidiaries or affiliated companies, and different business scenarios perform different business processes. In related technologies, business process management efficiency is low, and it is difficult to achieve accurate and secure process management.

[0070] In view of this, the embodiments of the present application provide a business process management method, system, electronic device and storage medium. The scheme obtains target user information and business process application information, performs data preprocessing on the target user information and business process application information, obtains business feature information, and then inputs the business feature information into a preset neural network model for identity prediction to obtain identity recommendation information. Then, the embodiment of the present invention obtains user selection data to determine the target identity information from the identity recommendation information through the user selection data, and then dynamically constructs the approval process data according to the target identity information, thereby realizing business process management, which can effectively improve the efficiency of business process management and improve the accuracy and security of process management.

[0071] The business process management method provided in the embodiment of the present application relates to the field of information management technology. The business process management method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements the business process management method, etc., but is not limited to the above forms.

[0072] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can 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, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0073] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0074] Figure 1 is an optional flowchart of the business process management method provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S110 to S150.

[0075] Step S110: Acquire target user information and business process application information.

[0076] Step S120: pre-process the target user information and the business process application information to obtain business feature information.

[0077] Step S130: Input the business feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information.

[0078] Step S140: Acquire user selection data to determine target identity information from the identity recommendation information through the user selection data.

[0079] Step S150: Dynamically construct approval process data according to the target identity information.

[0080] During the working process of this specific embodiment, the embodiment of the present invention first obtains the target user information and the business process application information to perform data preprocessing on the target user information and the business process information to obtain the business feature information. Specifically, the target user information in the embodiment of the present invention refers to the relevant information of the target user who needs to apply for the business process, such as identity information, occupation information, etc. Correspondingly, the business process application information in the embodiment of the present invention refers to the business process related request information initiated by the target user who needs to apply for the business process, such as the time, frequency, historical record, etc. of the business request. Accordingly, the embodiment of the present invention performs data preprocessing on the target user information and the business process application information, such as data standardization, feature normalization, etc., so as to extract the corresponding business feature information. Then, the embodiment of the present invention inputs the business feature information into the preset neural network model for identity prediction to obtain identity recommendation information. Specifically, the preset neural network model in the embodiment of the present invention refers to a deep neural network model pre-trained by relevant business feature data. Correspondingly, the identity recommendation information refers to the identity information of the target user in the currently applied business process predicted according to the corresponding business feature information. The embodiment of the present invention obtains the identity recommendation information by inputting the business feature information into the preset neural network model to predict the identity of the target user in the currently applied business process. For example, a user has multiple different identities in different subsidiaries or partners. The embodiment of the present invention analyzes and predicts the input business feature information through a preset neural network model, thereby obtaining the predicted situation of different identities of the target user in the current application business process, that is, identity recommendation information, such as sorting according to the matching degree of the predicted different identities, thereby obtaining the corresponding identity recommendation information. Further, the embodiment of the present invention obtains user selection data to determine the target identity information from the identity recommendation information through the user selection data, and then dynamically constructs the approval process data according to the target identity information. Specifically, the user selection data in the embodiment of the present invention refers to the identity data selected by the target user according to the identity recommendation information. Among them, after the preset neural network model predicts the identity recommendation information, the target user selects the identity to be adopted based on the identity recommendation information, thereby determining the target identity information. Correspondingly, after the target identity information is determined, the embodiment of the present invention generates the corresponding approval process and business approval form according to the determined target identity information and the corresponding business process application. The embodiment of the present invention uses a preset neural network model to analyze and predict business feature information, quickly predict and identify the identity information of the target user in the current business process, and then automatically generate approval process data based on the corresponding identity information, helping the target user to quickly and conveniently generate corresponding approval process data, effectively improving the efficiency of business process management, and improving the accuracy and security of process management.

[0081] Reference Figure 2 In some embodiments of the present invention, obtaining target user information and business process application information includes but is not limited to the following steps:

[0082] Step S210: Acquire the identity status information of the target object to obtain the target user information, wherein the target user information includes department information, job level information and business standard information.

[0083] Step S220: Acquire business process application information, wherein the business process application information includes time information, location information, and historical business data.

[0084] During the working process of this specific embodiment, the embodiment of the present invention first obtains the identity status information of the target object and obtains the target user information. Specifically, the target user information in the embodiment of the present invention includes the department information, rank information and business standard information of the target user. Among them, the department information in the embodiment of the present invention refers to the unit department to which the target user belongs, the rank information refers to the position grade information of the target user, and the business standard information refers to the standard corresponding to the business applied by the target user, such as the difference standard, meal standard, project application standard, etc. For example, when the target user logs in to the corresponding business process management system, the embodiment of the present invention obtains the identity status information of the target user, such as the user department, rank, difference standard, etc., and obtains the target user information. At the same time, the embodiment of the present invention obtains the business process application information. Specifically, the business process application information in the embodiment of the present invention includes time information, location information and historical business information. Accordingly, the time information in the embodiment of the present invention includes the time information of the target user logging into the system, the time of executing the corresponding business (such as business trip time), etc. In addition, the location information in the embodiment of the present invention includes the location where the target user is when logging into the system (such as login city information), the location where the business is executed (such as business trip city, project landing city, etc.), etc. Accordingly, the historical business data in the embodiment of the present invention refers to the historical business data of the target user's application, such as the commonly used business execution locations, identity usage frequency, identity weight, and historical business data (such as historical order average amount data).

[0085] Reference Figure 3 In some embodiments of the present invention, data preprocessing is performed on target user information and business process application information to obtain business feature information, including but not limited to the following steps:

[0086] Step S310: Data segmentation is performed on target user information and business process application information to obtain target feature data, wherein the target feature data includes numerical feature data, category feature data and time feature data.

[0087] Step S320: performing feature normalization processing on the numerical feature data to obtain first feature data.

[0088] Step S330: embedding and encoding the category feature data through a hash feature algorithm to obtain second feature data.

[0089] Step S340: Perform one-hot encoding on the time feature data to obtain third feature data.

[0090] Step S350: Merge the first feature data, the second feature data, and the third feature data to obtain service feature information.

[0091] In this specific embodiment, the embodiment of the present invention first divides the acquired target user information and business process application information to obtain numerical feature data, category feature data and time feature data. Specifically, the embodiment of the present invention divides the target user information and business process application information according to different dimensions, including numerical type data, category type data and time type data. For example, in the embodiment of the present invention, data such as difference standard, identity usage frequency, average order amount, etc. are divided into numerical feature data, and information such as department, rank, business trip city, and frequently used city are divided into category feature data, and information such as target user login time and business trip time are divided into time feature data. Then, the embodiment of the present invention performs feature normalization processing on the numerical feature data to obtain the first feature data. Specifically, the embodiment of the present invention generates the first feature data by normalizing the numerical feature data obtained by the division to map the corresponding numerical features to a distribution with a mean of 0 and a variance of 1. For example, the embodiment of the present invention uses the SoftMax function to standardize the numerical features such as identity weight, difference standard, historical order amount, and identity usage frequency to generate 4-dimensional feature data.

[0092] Next, the embodiment of the present invention embeds and encodes the category feature data through a hash feature algorithm to obtain the second feature data. Specifically, the embodiment of the present invention embeds and encodes the category feature data to convert it into a dense vector of fixed dimension to obtain the second feature data. For example, the embodiment of the present invention embeds and encodes the category feature information such as department, rank, business trip city, and frequently used city, and generates 4, 4, 32, and 32 dimensional features after embedding the department, rank, business trip city, and frequently used city through the hash feature method. At the same time, the embodiment of the present invention performs one-hot encoding on the time feature data to obtain the third feature data. Specifically, the embodiment of the present invention encodes the time type data, such as the login time, business trip time and other fields, through the one-hot encoding algorithm to generate the third feature data. For example, the embodiment of the present invention splits the login time and the business trip time according to the month and day, and performs one-hot encoding on the month and day to generate 12-dimensional and 31-dimensional encodings, and obtains an 86-bit time feature. Finally, the embodiment of the present invention merges the first feature data, the second feature data, and the third feature data to obtain business feature information. Specifically, the embodiment of the present invention combines the first feature data, the second feature data, and the third feature data to construct input data for the network model, namely, service feature information. For example, the embodiment of the present invention combines the above-generated 4, 4, 4, 32, 32-dimensional features and 86-bit time features to generate 162-dimensional service feature information.

[0093] Reference Figure 4 In some embodiments of the present invention, before inputting the business feature information into a preset neural network model to perform identity prediction and obtain identity recommendation information, the business process management method provided by the embodiment of the present invention further includes but is not limited to the following steps:

[0094] Step S410: construct an initial neural network model, wherein the initial neural network model includes an attention mechanism layer, and the attention mechanism layer is used to determine feature weight data.

[0095] Step S420: construct a preset data set, wherein the preset data set includes input feature data and identity tag information, and the input feature data and the identity tag information correspond to each other.

[0096] Step S430: According to the preset data set, the cross entropy loss function is used as the model optimization target to train the initial neural network model to obtain the preset neural network model.

[0097] In this specific embodiment, the embodiment of the present invention first constructs an initial neural network model. Specifically, the initial neural network model in the embodiment of the present invention includes an attention mechanism layer to determine feature weight data through the attention mechanism layer. Among them, the embodiment of the present invention explores the correlation between model features through the attention mechanism layer in the initial neural network model, assigns weights to different features, and determines the relationship between important feature values ​​and user identity selection. At the same time, the embodiment of the present invention constructs a preset data set. Specifically, the preset data set in the embodiment of the present invention includes input feature data and identity tag information, and the input feature data corresponds to the identity tag information. Among them, the input feature data in the embodiment of the present invention includes feature information such as corresponding department information, rank information, business standard information, time information, location information, and historical business data. Correspondingly, the identity tag information refers to the actual identity tag of the user corresponding to the corresponding input feature data. At the same time, the example of the present invention divides the preset data set into a training set and a test set. Further, the embodiment of the present invention uses the cross entropy loss function as the model optimization target according to the preset data set to train the initial neural network model and obtain the preset neural network model. Specifically, in order to ensure that the model has good generalization ability, the embodiment of the present invention uses the cross-entropy loss function as the model optimization target to minimize the gap between the predicted identity and the actual identity. At the same time, the optimizer uses Adam to ensure the training efficiency of the model under complex feature combinations, and trains to obtain a preset neural network model. For example, Figure 5 As shown, the preset neural network model constructed in the embodiment of the present invention includes an input layer, an embedding layer, a first hidden layer (hidden layer 1), an attention mechanism layer (attention layer), a second hidden layer (hidden layer 2) and an output layer. Among them, the input layer in the embodiment of the present invention includes 162 neuron nodes for receiving preset feature information. In the embodiment of the present invention, the embedding layer includes 128 neuron nodes, which are connected to the input layer nodes in a fully connected manner to eliminate the influence of parameter positions on model discrimination. In addition, in the embodiment of the present invention, the first hidden layer includes 128 neuron nodes, and a ReLU activation function is used to extract high-order associations between features. At the same time, the self-attention mechanism layer in the embodiment of the present invention is used to assign weights to different features. Among them, the second hidden layer in the embodiment of the present invention includes 32 neuron nodes, and each layer includes a ReLU activation function to extract high-order associations between features. Finally, in the embodiment of the present invention, the output layer outputs the category and recommendation confidence value of the priority identity through Softmax, that is, the identity recommendation information.

[0098] Reference Figure 6 In some embodiments of the present invention, the embodiments of the present invention input the service feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information, including but not limited to the following steps:

[0099] Step S510: inputting the service feature information into the input layer of the preset neural network model to perform feature calculation to obtain fourth feature data.

[0100] Step S520: Add the fourth feature data to the embedding layer for averaging to obtain the fifth feature data. The embedding layer is connected to the input layer in a fully connected manner.

[0101] Step S530: input the fifth feature data into the first hidden layer to extract feature association information through the first activation function to obtain the first association feature.

[0102] Step S540: Input the first associated feature into the attention mechanism layer for weight distribution calculation to obtain a weighted feature vector.

[0103] Step S550: input the weighted feature vector into the second hidden layer to extract feature association information through a second activation function to obtain a second association feature.

[0104] Step S560: Input the second association feature into the output layer to output the identity recommendation information through the first normalization function. The identity recommendation information includes recommendation category information and recommendation index information, and the recommendation index information corresponds to the recommendation category information.

[0105] In this specific embodiment, the embodiment of the present invention first inputs the business feature information into the input layer of the preset neural network model for feature calculation to obtain the fourth feature data, and adds the fourth feature data to the embedding layer for averaging processing to obtain the fifth feature data. Specifically, the neuron nodes of the embedding layer in the embodiment of the present invention are connected to the neuron nodes of the input layer in a fully connected manner. Among them, the embodiment of the present invention adds the business feature information to the input layer, the input layer performs feature calculation, and adds the obtained fourth feature data to the embedding layer to generate feature data of 128 neuron nodes, that is, the fifth feature data. Then, the embodiment of the present invention inputs the fifth feature data into the first hidden layer to extract feature association information through the first activation function to obtain the first association feature. Specifically, the embodiment of the present invention sends the fifth feature data into the first hidden layer, extracts the high-order association between features through the ReLU activation function, and obtains the first association feature. Further, the embodiment of the present invention inputs the first association feature into the attention mechanism layer for weight distribution calculation to obtain a weighted feature vector, and inputs the weighted feature vector into the second hidden layer for feature association information extraction to obtain the second association feature. Specifically, the embodiment of the present invention uses the attention mechanism layer to explore the correlation between the first associated features output by the first hidden layer, and assigns weights to different features to obtain a weighted feature vector. Then, the embodiment of the present invention inputs the weighted feature vector into the second hidden layer, extracts the associated information between the features through the ReLU activation function, and obtains the second associated feature. Finally, the embodiment of the present invention inputs the second associated feature into the output layer to output the identity recommendation information through the first normalization function. Specifically, the first normalization function in the embodiment of the present invention includes the Softmax function. Accordingly, the identity recommendation information in the embodiment of the present invention includes the recommended category information and the recommended index information, and the recommended index information corresponds to the recommended category information one by one. Among them, the recommended category information in the embodiment of the present invention refers to the identity information of the target user predicted by the model, and the recommended index information refers to the confidence or confidence value corresponding to the identity information of the target user predicted by the model. For example, the model predicts that the target user may match three identity labels A, B, and C, respectively, and the recommendation index of identity label A is 5 stars, the recommendation index of identity label B is 3 stars, and the recommendation index of identity label C is 2 stars. Correspondingly, the embodiment of the present invention outputs the category and recommendation confidence value of the priority identity through the Softmax function at the output layer.

[0106] Reference Figure 7 In some embodiments of the present invention, the first associated feature is input into the attention mechanism layer for weight distribution calculation to obtain a weighted feature vector, including but not limited to the following steps:

[0107] Step S610: performing dot product calculation on each first correlation feature to obtain preset importance score data.

[0108] Step S620: normalizing the preset importance score data using a second normalization function to obtain feature weight data.

[0109] Step S630: Perform weighted summation on the feature weight data to obtain a weighted feature vector.

[0110] In this specific embodiment, the embodiment of the present invention first performs dot product calculation on each first correlation feature to obtain preset importance score data, and then normalizes the preset importance score data through a second normalization function to obtain feature weight data, and performs weighted summation on the feature weight data to obtain a weighted feature vector. Specifically, the preset importance score data in the embodiment of the present invention refers to the importance score corresponding to each feature. For example, the first correlation feature matrix A in the embodiment of the present invention = [x1, x2, ..., x n ], where x i represents the vectorized representation of the i-th feature in the hidden layer, and the weight of the feature can be dynamically calculated through the attention mechanism. i With x i Perform dot product to obtain the importance score of each feature, as shown in the following formula (1):

[0111] score(x i )=x i T x i (1)

[0112] Among them, score(x i ) represents the feature x i The preset importance score data of .

[0113] Next, the second normalization function in the embodiment of the present invention includes a Softmax normalization function, and the score of each feature is normalized by the Softmax normalization function to convert it into a weight α i , as shown in the following formula (2):

[0114]

[0115] Among them, α i Represents the weight of the i-th feature. The weight value is between 0 and 1, and the sum is 1.

[0116] Furthermore, in the attention weight α i After the calculation is completed, the embodiment of the present invention performs weighted summation on the input features to obtain a weighted feature vector z, as shown in the following formula (3):

[0117]

[0118] Accordingly, the weighted feature vector z calculated in the embodiment of the present invention will be passed as input to the subsequent layers of the preset neural network model for final identity recommendation.

[0119] Combination Figure 1 , refer to Figure 8 In some embodiments of the present invention, after obtaining user selection data to determine target identity information from identity recommendation information through the user selection data, the business process management method provided by the embodiment of the present invention further includes but is not limited to the following steps:

[0120] Step S710: Optimize parameters of a preset neural network model through a preset reward mechanism according to the target identity information and the identity recommendation information.

[0121] In this specific embodiment, the embodiment of the present invention optimizes the parameters of the preset neural network model through a preset reward mechanism in combination with the identity recommendation information output by the preset neural network model and the target identity information finally determined. Specifically, during the operation of the business process management system, the embodiment of the present invention continuously detects the behavior data of the target user to collect the target user's feedback on the identity information recommended by the preset neural network model. Exemplarily, when the target user accepts the recommended identity, such as the target user selects the recommended identity information with the highest recommendation index as the target identity information, the embodiment of the present invention takes this as positive feedback. Conversely, when the target user rejects the recommended identity, such as the target user does not select the recommended identity information with the highest recommendation index as the target identity information or selects the new identity information generated by the user as the target identity information, the embodiment of the present invention takes this as negative feedback. Then, the embodiment of the present invention continuously adjusts the model output according to these feedback information, optimizes the model parameters of the preset neural network model through the corresponding preset reward mechanism, and gradually optimizes the recommendation accuracy to adapt to the personalized needs of users and improve the accuracy of recommendations.

[0122] The following describes the solution of the embodiment of the present invention in detail in conjunction with a specific business process management scenario:

[0123] For example, refer to Fig. 9 , Fig. 9It is a flowchart of the dynamic approval of mobile business processes provided by an embodiment of the present invention. Specifically, in the application scenario of business travel process management, in the embodiment of the present invention, there is an identity intersection relationship between the employees of W Company and T Company, and the departmental organizational structure, personnel information, identity position, and rank difference information of the two companies are synchronized to the approval system (business process management system). Among them, the information of the personnel includes mobile phone numbers, etc., and there may be duplicate data. At this time, the embodiment of the present invention synchronizes the hrId (unique identity identifier) ​​and the main, deputy and part-time information of the role identity position to the personnel information, and the system will assign a primary key to it. Correspondingly, an employee Q holds the position of an ordinary employee in the company. According to the company's regulations, the daily travel expense standard for his business trip is 400 yuan. Due to work needs, employee Q often travels on business. For example, on May 30, 2024, employee Q logged into the business process management system in L city and submitted the latest business trip application. Among them, according to the application, employee Q plans to go to P city to perform work tasks on June 30, 2024. Correspondingly, the embodiment of the present invention extracts the relevant information of the user based on the above content and fills it in the corresponding fields, as shown in Table 1 below:

[0124] Table 1

[0125] Field Name Value Field Description department W Company Employee's company Rank Ordinary employees Employee's rank Difference 400 The employee travel allowance is 300 yuan per day Login time 2024-05-30 The current login time is May 30, 2024 Landing city L City The location where the employee logs into the system Business trip time 2024-06-30 The time of the employee's most recent business trip application Business trip city P City The city where the employee applied for a business trip in the most recent day Frequently used cities P City Cities where employees frequently travel Business trip time 2024-06-30 Business trip time Frequency of identity use 5 In the past 3 months, the employee's current identity has been used 5 times Identity Weight 0.7 Indicates that in historical choices, the user tends to choose this identity Average order amount 379.5 Average hotel price for employees on business trips

[0126] Next, the embodiment of the present invention preprocesses the collected data information. For example, the embodiment of the present invention performs feature normalization processing on numerical data such as travel standards, identity usage frequency, travel amount and average order amount. Next, the numerical features such as business trip frequency, difference standard, order expenditure and identity usage frequency are standardized to convert these data into a distribution with a mean of 0 and a variance of 1. Correspondingly, the embodiment of the present invention generates a four-dimensional feature vector through this process. First, the embodiment of the present invention generates the corresponding column normalized data [0.1, 0.15, 0.1, 0.9] from the original data of business trip frequency, difference standard, order expenditure and identity usage frequency [5,400, 379.5, 0.7]. Secondly, the embodiment of the present invention embeds and encodes the category feature information such as department, rank, business trip city, and frequently used city, and converts it into a dense vector of fixed dimension. The embodiment of the present invention uses a hash feature method to embed the department rank, business trip city, and frequently used city to generate 4, 4, 32, and 32-dimensional features, such as embedding "P city" as [0.85, 0.15, 0.2, ..., 0.3]. Correspondingly, the feature vectors output by the above features are [0.25, 0.75, 0.1, 0.9], [0.6, 0.4, 0.7, 0.3], [0.85, 0.15, 0.2, ..., 0.3], [0.95, 0.05, 0.4, ..., 0.6]. At the same time, the embodiment of the present invention encodes time-type data, such as login time, business trip time, and other fields. Among them, the embodiment of the present invention splits the login time and business trip time according to the month and day, and then performs unique hot encoding on the month and day to generate 12-dimensional and 31-dimensional codes. Accordingly, in the embodiment of the present invention, the first 12 bits of the code represent the month, and the last 31 bits represent the date. When the date is May 30, the 5th and 42nd bits of the feature value are 1, and the others are 0. Among them, the two time parameters finally obtain an 86-bit time feature. Finally, the above features are merged to generate a 162-dimensional feature for the input data of the network model.

[0127] Next, the embodiment of the present invention inputs the constructed feature data into the preset neural network model for identity prediction to obtain the corresponding recommended identity. Then, the target user (employee Q) selects the target identity information from the recommended identities and initiates approval. Finally, the system matches the corresponding company configuration approval flow conditions, submits for approval, and completes the approval management process of the business process. In addition, the embodiment of the present invention also iterates the system based on the identity information selected by employee Q to improve the prediction accuracy of the preset neural network model. For example, after employee Q selects the identity of an ordinary employee, the system will automatically generate an approval process for employee Q's business trip application based on the approval flow rules configured by the company in the background. At the same time, the system will perform iterative optimization based on the user's selection. Among them, as shown in Table 2 below, there are two scenarios selected by the user in the embodiment of the present invention, and the corresponding system tuning method:

[0128] Table 2

[0129]

[0130] It is easy to understand that the embodiment of the present invention integrates time features, text features, and numerical features through heterogeneous feature extraction, which effectively improves the accuracy of user identity prediction. At the same time, in processing urban text information, the dimensionality explosion problem caused by unique hot encoding is solved by using hash hashing. At the same time, based on employee behavior data, the embodiment of the present invention uses deep neural networks (DNNs) and attention mechanisms to automatically identify employees' priority identities in different situations, and iteratively trains models based on user selections, effectively improving the accuracy of business process management. In addition, the embodiment of the present invention effectively simplifies approval operations and improves the efficiency of business process management through automatic identity recommendation and dynamic configuration of business approval processes based on determined target identity information. At the same time, the embodiment of the present invention effectively improves the security of business process management by digitalizing and automating the management of business processes.

[0131] See also Fig.10 The embodiment of the present application also provides a business process management system, which can implement the above-mentioned business process management method, and the system includes:

[0132] The first module 810 is used to obtain target user information and business process application information.

[0133] The second module 820 is used to perform data preprocessing on the target user information and the business process application information to obtain business feature information.

[0134] The third module 830 is used to input the business feature information into a preset neural network model to perform identity prediction and obtain identity recommendation information.

[0135] The fourth module 840 is used to obtain user selection data to determine target identity information from the identity recommendation information through the user selection data.

[0136] The fifth module 850 is used to dynamically construct approval process data according to the target identity information.

[0137] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0138] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned business process management method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0139] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0140] See also Fig.11 , Fig.11 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0141] The processor 910 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0142] The memory 920 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 920 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 920, and the processor 910 calls and executes the business process management method of the embodiment of this application;

[0143] Input / output interface 930, used to implement information input and output;

[0144] Communication interface 940, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0145] bus 950 , which transmits information between the various components of the device (e.g., processor 910 , memory 920 , input / output interface 930 , and communication interface 940 );

[0146] The processor 910 , the memory 920 , the input / output interface 930 , and the communication interface 940 are connected to each other in communication within the device via a bus 950 .

[0147] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above-mentioned business process management method is implemented.

[0148] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0149] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0150] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0151] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0152] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0154] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0155] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0156] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0157] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0159] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0160] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A business process management method, characterized in that: The method comprises the following steps: Obtain target user information and business process application information; Performing data preprocessing on the target user information and the business process application information to obtain business feature information; Inputting the service feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information; Acquiring user selection data to determine target identity information from the identity recommendation information through the user selection data; The approval process data is dynamically constructed according to the target identity information.

2. The method according to claim 1, characterized in that: The obtaining of target user information and business process application information includes: Acquire the identity status information of the target object to obtain the target user information; wherein the target user information includes department information, rank information and business standard information; Acquire the business process application information; wherein the business process application information includes time information, location information and historical business data.

3. The method according to claim 1, characterized in that The data preprocessing of the target user information and the business process application information to obtain business feature information includes: Performing data segmentation on the target user information and the business process application information to obtain target feature data; wherein the target feature data includes numerical feature data, category feature data and time feature data; Performing feature normalization processing on the numerical feature data to obtain first feature data; The category feature data is embedded and encoded by a hash feature algorithm to obtain second feature data; Performing one-hot encoding on the time feature data to obtain third feature data; The first feature data, the second feature data, and the third feature data are combined to obtain the service feature information.

4. The method according to claim 1, characterized in that: Before executing the step of inputting the service feature information into a preset neural network model to perform identity prediction to obtain identity recommendation information, the method further includes: Constructing an initial neural network model; wherein the initial neural network model includes an attention mechanism layer, and the attention mechanism layer is used to determine feature weight data; Constructing a preset data set; wherein the preset data set includes input feature data and identity tag information, and the input feature data corresponds to the identity tag information; According to the preset data set, the cross entropy loss function is used as the model optimization target to perform model training on the initial neural network model to obtain the preset neural network model.

5. The method according to claim 1, characterized in that The step of inputting the service feature information into a preset neural network model for identity prediction to obtain identity recommendation information includes: Inputting the service feature information into an input layer of a preset neural network model to perform feature calculation to obtain fourth feature data; Adding the fourth feature data to the embedding layer for averaging to obtain fifth feature data; wherein the embedding layer is connected to the input layer in a fully connected manner; Inputting the fifth feature data into the first hidden layer to extract feature association information through a first activation function to obtain a first association feature; Input the first associated feature into the attention mechanism layer to perform weight distribution calculation to obtain a weighted feature vector; Inputting the weighted feature vector into a second hidden layer to extract feature association information through a second activation function to obtain a second association feature; The second association feature is input into the output layer to output the identity recommendation information through a first normalization function; wherein the identity recommendation information includes recommendation category information and recommendation index information, and the recommendation index information corresponds to the recommendation category information.

6. The method according to claim 5, characterized in that The step of inputting the first associated feature into the attention mechanism layer for weight distribution calculation to obtain a weighted feature vector includes: Performing dot product calculation on each of the first associated features to obtain preset importance score data; Normalizing the preset importance score data by a second normalization function to obtain feature weight data; The feature weight data is weighted and summed to obtain the weighted feature vector.

7. The method according to claim 1, characterized in that After performing the step of acquiring the user selection data to determine the target identity information from the identity recommendation information through the user selection data, the method further includes: Parameters of the preset neural network model are optimized through a preset reward mechanism according to the target identity information and the identity recommendation information.

8. A business process management system, characterized in that: The system comprises: The first module is used to obtain target user information and business process application information; The second module is used to perform data preprocessing on the target user information and the business process application information to obtain business feature information; The third module is used to input the business feature information into a preset neural network model to perform identity prediction and obtain identity recommendation information; A fourth module is used to obtain user selection data, so as to determine target identity information from the identity recommendation information through the user selection data; The fifth module is used to dynamically construct approval process data according to the target identity information.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images