Enterprise management method and system based on big data

Abstract

The invention provides an enterprise management method and system based on big data, and relates to the technical field of enterprise management, and the method comprises the steps: determining a first reimbursement reasonable degree of a current enterprise reimbursement order of a user through employing a time sequence data processing model based on the enterprise reimbursement order time sequence data of the user; processing the current enterprise reimbursement order of the user based on a graph neural network model to determine a second reimbursement reasonable degree of the current enterprise reimbursement order of the user; determining a target reimbursement reasonable degree of the current enterprise reimbursement order of the user based on the first reimbursement reasonable degree of the current enterprise reimbursement order of the user and the second reimbursement reasonable degree of the current enterprise reimbursement order of the user; the approval result of the current enterprise reimbursement order of the user is determined based on the target reimbursement reasonable degree of the current enterprise reimbursement order of the user. The method can quickly and accurately perform approval management on the enterprise reimbursement order.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management, and in particular to an enterprise management method and system based on big data. Background Art

[0002] With the in-depth development of enterprise informatization and digital transformation, the traditional paper-based reimbursement process can no longer meet the efficiency, transparency and compliance requirements of modern enterprise management. Traditional methods usually rely on manual approval and processing, which may consume a lot of time and human resources. Manual review of reimbursement documents is easily affected by subjective judgment, making it difficult to effectively identify potential fraud and non-compliance, increasing the financial risks of enterprises.

[0003] Therefore, how to quickly and accurately approve and manage corporate expense reimbursement forms is an urgent problem that needs to be solved. Summary of the invention

[0004] The main technical problem solved by the present invention is how to quickly and accurately manage the approval of enterprise expense reimbursement forms.

[0005] According to a first aspect, the present invention provides an enterprise management method based on big data, including: obtaining a user's current enterprise reimbursement form and the user's historical enterprise reimbursement form; organizing the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form into user enterprise reimbursement form time series data based on chronological order; determining a first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; processing the user's current enterprise reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise reimbursement form; determining a target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form; and determining the approval result of the user's current enterprise reimbursement form based on the target reimbursement rationality of the user's current enterprise reimbursement form.

[0006] In one possible implementation, the time series data processing model is a gated recurrent unit, the input of the time series data processing model is the time series data of the user's enterprise reimbursement form, and the output of the time series data processing model is the first reimbursement rationality level of the user's current enterprise reimbursement form.

[0007] In one possible implementation, the processing of the user's current enterprise reimbursement form based on the graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form includes: determining multiple similar approved enterprise reimbursement forms based on the user's current enterprise reimbursement form and the similarity between the user's current enterprise reimbursement form and each similar approved enterprise reimbursement form; constructing a graph structure, the graph structure including multiple nodes and edges between the multiple nodes, the multiple nodes including a user reimbursement form node and multiple similar reimbursement form nodes, the one user reimbursement form node being a central node, each similar reimbursement form node establishing an edge with the one user reimbursement form node, the feature of each similar reimbursement form node establishing an edge with the one user reimbursement form node being the similarity between the user's current enterprise reimbursement form and each similar approved enterprise reimbursement form, the node feature of the one user reimbursement form node being the user's current enterprise reimbursement form, and the node feature of each similar reimbursement form node being each similar approved enterprise reimbursement form; processing the graph structure based on the graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form.

[0008] In one possible implementation, determining the target reimbursement reasonableness of the user's current enterprise reimbursement form based on the first reimbursement reasonableness of the user's current enterprise reimbursement form and the second reimbursement reasonableness of the user's current enterprise reimbursement form includes: assigning different weights to the first reimbursement reasonableness of the user's current enterprise reimbursement form and the second reimbursement reasonableness of the user's current enterprise reimbursement form, respectively, and then performing weighted summation to obtain the target reimbursement reasonableness of the user's current enterprise reimbursement form.

[0009] In a possible implementation, determining the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form includes: When the target reimbursement rationality is higher than or equal to the preset threshold, the user's current enterprise reimbursement form is automatically approved; If the target reimbursement rationality is lower than the preset threshold, it will be submitted for manual review.

[0010] According to a second aspect, the present invention provides an enterprise management system based on big data, comprising: The acquisition module is used to obtain the user's current enterprise expense reimbursement form and the user's historical enterprise expense reimbursement form; A sorting module, used to sort the current enterprise expense reimbursement form of the user and the historical enterprise expense reimbursement form of the user into time series data of the enterprise expense reimbursement form of the user based on the time sequence; A time series data processing module, used to determine a first reimbursement rationality of the user's current enterprise reimbursement form based on the user's enterprise reimbursement form time series data using a time series data processing model; A graph neural network module, used to process the user's current enterprise expense reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise expense reimbursement form; A target reimbursement rationality determination module, configured to determine a target reimbursement rationality of the user's current enterprise reimbursement form based on a first reimbursement rationality of the user's current enterprise reimbursement form and a second reimbursement rationality of the user's current enterprise reimbursement form; The approval module is used to determine the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form.

[0011] In one possible implementation, the time series data processing model is a gated recurrent unit, the input of the time series data processing model is the time series data of the user's enterprise reimbursement form, and the output of the time series data processing model is the first reimbursement rationality level of the user's current enterprise reimbursement form.

[0012] In a possible implementation, the graph neural network module is further used to: Determining, based on the user's current enterprise expense reimbursement form, a plurality of similar approved enterprise expense reimbursement forms and a similarity between the user's current enterprise expense reimbursement form and each of the similar approved enterprise expense reimbursement forms; Constructing a graph structure, the graph structure comprising a plurality of nodes and edges between the plurality of nodes, the plurality of nodes comprising a user expense report node and a plurality of similar expense report nodes, the user expense report node being a central node, each similar expense report node establishing an edge with the user expense report node, the feature of each similar expense report node establishing an edge with the user expense report node being the similarity between the user's current enterprise expense report and each similar approved enterprise expense report, the node feature of the user expense report node being the user's current enterprise expense report, and the node feature of each similar expense report node being each similar approved enterprise expense report; The graph structure is processed based on a graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form.

[0013] According to a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as described above, the method comprising: obtaining a user's current corporate reimbursement form and a user's historical corporate reimbursement form; arranging the user's current corporate reimbursement form and the user's historical corporate reimbursement form into user corporate reimbursement form time series data based on a chronological order; determining a first reimbursement rationality of the user's current corporate reimbursement form using a time series data processing model based on the user's corporate reimbursement form time series data; processing the user's current corporate reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current corporate reimbursement form; determining a target reimbursement rationality of the user's current corporate reimbursement form based on the first reimbursement rationality of the user's current corporate reimbursement form and the second reimbursement rationality of the user's current corporate reimbursement form; determining the approval result of the user's current corporate reimbursement form based on the target reimbursement rationality of the user's current corporate reimbursement form.

[0014] According to the fourth aspect, the present embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned enterprise management method based on big data, the method comprising: obtaining a user's current enterprise reimbursement form and a user's historical enterprise reimbursement form; arranging the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form into user enterprise reimbursement form time series data based on chronological order; determining a first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; processing the user's current enterprise reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise reimbursement form; determining a target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form; and determining the approval result of the user's current enterprise reimbursement form based on the target reimbursement rationality of the user's current enterprise reimbursement form.

[0015] The present invention provides a big data-based enterprise management method and system, the method comprising obtaining a user's current enterprise reimbursement form and the user's historical enterprise reimbursement form; arranging the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form into user enterprise reimbursement form time series data based on chronological order; determining a first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; processing the user's current enterprise reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise reimbursement form; determining a target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form; and determining an approval result of the user's current enterprise reimbursement form based on the target reimbursement rationality of the user's current enterprise reimbursement form. The method can quickly and accurately approve and manage enterprise reimbursement forms. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of an enterprise management method based on big data provided by an embodiment of the present invention; Figure 2 A schematic diagram of a process for determining the rationality of a second reimbursement of a user's current enterprise expense reimbursement form based on a graph neural network model provided by an embodiment of the present invention; Figure 3 A schematic diagram of an enterprise management system based on big data provided by an embodiment of the present invention; Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention; Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present invention better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present invention are not shown or described in the specification, this is to avoid the core part of the present invention being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0018] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0019] The serial numbers assigned to the components herein, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in the present invention include direct and indirect connections (couplings) unless otherwise specified.

[0020] In an embodiment of the present invention, there is provided Figure 1 A method for enterprise management based on big data is shown, and the method for enterprise management based on big data includes steps S1 to S6: Step S1, obtaining the user's current enterprise expense reimbursement form and the user's historical enterprise expense reimbursement form.

[0021] The user's current corporate expense reimbursement form refers to the most recent expense reimbursement application submitted by the user, including the specific content, amount, date, type, etc. of the reimbursement. For example, a business trip expense reimbursement form submitted by Zhang San on March 15, 2023 includes details such as accommodation, transportation, and meal expenses.

[0022] The user's historical corporate expense reports refer to all expense reports submitted by the user in the past, covering all expense reports from joining the company to the present. For example, all expense reports submitted by Zhang San since joining the company in 2020, including various types of expense reports such as daily office supplies purchases, client entertainment, and training fees.

[0023] The user's historical enterprise reimbursement forms include the user's historical enterprise reimbursement forms and approval results, and the approval results include whether the reimbursement is approved or not approved.

[0024] Step S2, sorting the user's current enterprise expense reimbursement form and the user's historical enterprise expense reimbursement form into user enterprise expense reimbursement form time series data based on time sequence.

[0025] User enterprise expense report time series data is a data set arranged in chronological order, and each data point corresponds to a specific timestamp. User enterprise expense report time series data refers to expense report records sorted by submission date.

[0026] In some embodiments, the data set may be sorted according to the submission date of the reimbursement form or the occurrence date of the reimbursement event, which can usually be achieved through a data sorting function in a programming language.

[0027] Step S3, based on the time series data of the user's enterprise expense reimbursement form, use a time series data processing model to determine a first reimbursement rationality of the user's current enterprise expense reimbursement form.

[0028] The time series data processing model is a gated recurrent unit, the input of which is the time series data of the user's enterprise expense reimbursement form, and the output of which is the first reimbursement rationality of the user's current enterprise expense reimbursement form. The gated recurrent unit is a recurrent neural network that controls the flow of information by introducing a gating mechanism to solve the problem of long-term dependency. The gated recurrent unit includes an update gate and a reset gate, which are used to determine which information needs to be retained or updated, so as to better handle long-term dependencies in sequence data.

[0029] The main purpose of executing step S3 is to automatically evaluate the rationality of the user's current reimbursement form by analyzing the user's historical reimbursement behavior using a time series data processing model (such as a gated recurrent unit). This step predicts and identifies abnormal reimbursement behavior based on historical data, which helps to automate the reimbursement review process, reduce human intervention, and improve review efficiency and accuracy.

[0030] By using time series models such as gated recurrent units, we can capture the complex patterns of reimbursement behavior over time, such as seasonal fluctuations in reimbursement amounts and periodic changes in reimbursement frequency, which is crucial for identifying abnormal reimbursement behavior. This automated analysis can not only speed up reimbursement review, but also reduce erroneous approvals caused by human judgment errors, ensuring the fairness and compliance of the reimbursement review process.

[0031] The time series data of user enterprise expense reports contains rich historical expense information, reflecting the user's past behavior patterns in expense reports. By analyzing this data, we can identify the common types, frequencies, and amount ranges of user expense reports, thereby establishing a "normal" expense report behavior benchmark. When a new expense report appears, it can be compared with the historical pattern to assess whether it deviates from the normal range, and then determine the rationality of the expense report. Based on historical expense report data, abnormal expense report behavior can be detected through gated recurrent units. For example, if a user usually reports small office supply expenses, and suddenly a large travel expense report appears, this may be considered an anomaly and requires further review. Time series data analysis can help identify such anomalies.

[0032] Step S4: Process the user's current enterprise expense reimbursement form based on the graph neural network model to determine the second reimbursement rationality of the user's current enterprise expense reimbursement form.

[0033] The graph neural network model is specially designed to process graph structured data and can capture and learn the complex relationships between nodes in the graph. In the reimbursement rationality assessment, the graph neural network model can analyze the relationship between the user's current reimbursement form and historical reimbursement forms to evaluate the rationality of the second reimbursement.

[0034] In some embodiments, Figure 2 A flow chart of determining the reasonableness of the second reimbursement of the user's current enterprise expense reimbursement form based on a graph neural network model provided in an embodiment of the present invention, wherein determining the reasonableness of the second reimbursement of the user's current enterprise expense reimbursement form based on a graph neural network model includes steps S21 to S23: Step S21, based on the current enterprise expense reimbursement form of the user, determining a plurality of similar approved enterprise expense reimbursement forms and a similarity between the current enterprise expense reimbursement form of the user and each of the similar approved enterprise expense reimbursement forms.

[0035] In some embodiments, a similarity calculation algorithm may be used to calculate the similarity between the user's current expense report and a historical similar expense report. The similarity calculation algorithm includes cosine similarity, Jaccard similarity, and the like.

[0036] Step S22, construct a graph structure, the graph structure includes multiple nodes and edges between the multiple nodes, the multiple nodes include a user reimbursement form node and multiple similar reimbursement form nodes, the one user reimbursement form node is a central node, each similar reimbursement form node establishes an edge with the one user reimbursement form node, the characteristics of each similar reimbursement form node establishing an edge with the one user reimbursement form node are the similarity between the user's current enterprise reimbursement form and each similar approved enterprise reimbursement form, the node characteristic of the one user reimbursement form node is the user's current enterprise reimbursement form, and the node characteristics of each similar reimbursement form node are each similar approved enterprise reimbursement form.

[0037] By building a graph structure, the information of the user's current expense report can be fused with the information of similar historical expense reports. In a graph neural network, nodes can transmit and aggregate information through edges, which means that the user's current expense report can learn clues about the rationality of the expense report from similar expense reports. This information fusion helps to more comprehensively evaluate the rationality of the expense report.

[0038] Graph neural networks can learn local features of nodes and edges, which means that even in a large-scale graph structure, the model can focus on the nodes directly related to the current expense report and extract their features for analysis. This local feature learning capability allows the model to process complex expense report data sets without being disturbed by unnecessary information.

[0039] The relationship between expense reports may be very complex, not just a simple similarity score, but may involve multiple types of relationships, such as expense type, expense frequency, and the trend of reimbursement amount. The graph structure can capture these complex relationships, and the graph neural network model can learn deep patterns and rules from these complex relationships.

[0040] Step S23: Process the graph structure based on the graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form.

[0041] The input of the graph neural network model is the graph structure, and the output of the graph neural network model is the second reimbursement rationality of the user's current enterprise reimbursement form.

[0042] The second reimbursement rationality is evaluated by the graph neural network model based on the graph structure of the user's current reimbursement form and similar approved reimbursement forms, reflecting the similarity between the user's current reimbursement form and similar historical reimbursement forms.

[0043] Step S5, determining a target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form.

[0044] In some embodiments, different weights may be assigned to the first reimbursement rationality of the user's current corporate reimbursement form and the second reimbursement rationality of the user's current corporate reimbursement form, respectively, and then the target reimbursement rationality of the user's current corporate reimbursement form may be obtained by weighted summation.

[0045] The purpose of executing step S5 is to combine the reimbursement rationality evaluations from two different angles to provide a more comprehensive and accurate judgment of the authenticity of the reimbursement form. The first reimbursement rationality focuses on the pattern of the time series, focusing on the consistency and periodicity of the reimbursement behavior. The second reimbursement rationality focuses on the similarity between reimbursement forms, focusing on the context and consistency of the reimbursement forms. Through the weighted summation method, these two independent evaluation results can be integrated to form a unified indicator - the target reimbursement rationality, which is used for the final reimbursement approval decision.

[0046] The target reimbursement rationality is a comprehensive score obtained by weighted summing the first reimbursement rationality and the second reimbursement rationality, and is used to finally determine whether the user's current reimbursement form should be approved. For example, if the first reimbursement rationality is 0.8 and the second reimbursement rationality is 0.9, and the weights are 0.6 and 0.4 respectively, then the target reimbursement rationality is 0.84.

[0047] Step S6, determining the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form.

[0048] When the target reimbursement rationality is higher than or equal to the preset threshold, the user's current enterprise reimbursement form is automatically approved; If the target reimbursement rationality is lower than the preset threshold, it will be submitted for manual review.

[0049] The purpose of executing step S6 is to automatically or semi-automatically complete the reimbursement approval process based on the target reimbursement rationality, improve approval efficiency, reduce human errors, and ensure strict implementation of reimbursement policies. By setting thresholds, obviously reasonable reimbursement forms can be automatically approved, and suspicious or complex reimbursement forms can be transferred to manual review, which can not only speed up the approval process, but also ensure the accuracy and fairness of the approval process, avoid misapproval or missed approval, and maintain corporate financial discipline.

[0050] Based on the same inventive concept, Figure 3 A schematic diagram of an enterprise management system based on big data provided by an embodiment of the present invention, wherein the enterprise management system based on big data includes: An acquisition module 31 is used to acquire the user's current enterprise expense reimbursement form and the user's historical enterprise expense reimbursement form; A sorting module 32, for sorting the current enterprise expense reimbursement form of the user and the historical enterprise expense reimbursement form of the user into user enterprise expense reimbursement form time series data based on time sequence; A time series data processing module 33, configured to determine a first reimbursement rationality of the user's current enterprise reimbursement form based on the user's enterprise reimbursement form time series data using a time series data processing model; A graph neural network module 34, configured to process the current enterprise expense reimbursement form of the user based on a graph neural network model to determine a second reimbursement rationality of the current enterprise expense reimbursement form of the user; A target reimbursement rationality determination module 35, configured to determine a target reimbursement rationality of the user's current enterprise reimbursement form based on a first reimbursement rationality of the user's current enterprise reimbursement form and a second reimbursement rationality of the user's current enterprise reimbursement form; The approval module 36 is used to determine the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form.

[0051] Based on the same inventive concept, an embodiment of the present invention provides an electronic device, such as Figure 4 As shown, including: It includes: a processor 41; a memory 42; and a computer program; wherein the computer program is stored in the memory 42 and is configured to be executed by the processor 41 to implement the enterprise management method based on big data as provided above, the method including: obtaining the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form; arranging the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form into user enterprise reimbursement form time series data based on chronological order; determining the first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; processing the user's current enterprise reimbursement form based on a graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form; determining the target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form; determining the approval result of the user's current enterprise reimbursement form based on the target reimbursement rationality of the user's current enterprise reimbursement form.

[0052] Based on the same inventive concept, this embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by the processor 41, implements the aforementioned enterprise management method based on big data, the method comprising: obtaining the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form; arranging the user's current enterprise reimbursement form and the user's historical enterprise reimbursement form into user enterprise reimbursement form time series data based on chronological order; determining a first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; processing the user's current enterprise reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise reimbursement form; determining a target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form; and determining the approval result of the user's current enterprise reimbursement form based on the target reimbursement rationality of the user's current enterprise reimbursement form.

[0053] The enterprise management method based on big data provided in the embodiment of the present application can be applied to terminal devices (such as mobile phones), tablet computers, laptops, ultra-mobile personal computers (UMPCs), handheld computers, netbooks, personal digital assistants (PDAs), wearable devices (such as smart watches, smart glasses or smart helmets, etc.), augmented reality (AR) and virtual reality (VR) devices, smart home devices, car computers and other electronic devices, and the embodiment of the present application does not impose any restrictions on this.

[0054] Taking the mobile phone 100 as an example of the electronic device, Figure 5 A schematic structural diagram of the mobile phone 100 is shown.

[0055] like Figure 5 As shown, the mobile phone 100 may include a processing module 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc.

[0056] The processing module 110 can be used to: obtain the user's current corporate reimbursement form and the user's historical corporate reimbursement form; organize the user's current corporate reimbursement form and the user's historical corporate reimbursement form into user corporate reimbursement form time series data based on chronological order; determine the first reimbursement rationality of the user's current corporate reimbursement form using a time series data processing model based on the user's corporate reimbursement form time series data; process the user's current corporate reimbursement form based on a graph neural network model to determine the second reimbursement rationality of the user's current corporate reimbursement form; determine the target reimbursement rationality of the user's current corporate reimbursement form based on the first reimbursement rationality of the user's current corporate reimbursement form and the second reimbursement rationality of the user's current corporate reimbursement form; determine the approval result of the user's current corporate reimbursement form based on the target reimbursement rationality of the user's current corporate reimbursement form.

[0057] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A business management method based on big data, characterized in that: include: Obtain the user's current enterprise expense report and the user's historical enterprise expense reports; Arrange the current enterprise expense report of the user and the historical enterprise expense report of the user into time series data of the enterprise expense report of the user based on the time sequence; Determine a first reimbursement rationality of the user's current enterprise reimbursement form using a time series data processing model based on the user's enterprise reimbursement form time series data; Processing the user's current enterprise expense reimbursement form based on the graph neural network model to determine a second reimbursement rationality of the user's current enterprise expense reimbursement form; Determining a target reimbursement rationality of the user's current enterprise reimbursement form based on a first reimbursement rationality of the user's current enterprise reimbursement form and a second reimbursement rationality of the user's current enterprise reimbursement form; The approval result of the user's current enterprise reimbursement form is determined based on the target reimbursement rationality of the user's current enterprise reimbursement form.

2. The enterprise management method based on big data according to claim 1, characterized in that: The time series data processing model is a gated recurrent unit, the input of the time series data processing model is the time series data of the user's enterprise reimbursement form, and the output of the time series data processing model is the first reimbursement rationality of the user's current enterprise reimbursement form.

3. The enterprise management method based on big data according to claim 1, characterized in that: The processing of the current enterprise expense reimbursement form of the user based on the graph neural network model to determine the second reimbursement rationality of the current enterprise expense reimbursement form of the user includes: Determining, based on the user's current enterprise expense reimbursement form, a plurality of similar approved enterprise expense reimbursement forms and a similarity between the user's current enterprise expense reimbursement form and each of the similar approved enterprise expense reimbursement forms; Constructing a graph structure, the graph structure comprising a plurality of nodes and edges between the plurality of nodes, the plurality of nodes comprising a user expense report node and a plurality of similar expense report nodes, the user expense report node being a central node, each similar expense report node establishing an edge with the user expense report node, the feature of each similar expense report node establishing an edge with the user expense report node being the similarity between the user's current enterprise expense report and each similar approved enterprise expense report, the node feature of the user expense report node being the user's current enterprise expense report, and the node feature of each similar expense report node being each similar approved enterprise expense report; The graph structure is processed based on a graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form.

4. The enterprise management method based on big data according to claim 1, characterized in that: The method of determining the target reimbursement rationality of the user's current enterprise reimbursement form based on the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form includes: assigning different weights to the first reimbursement rationality of the user's current enterprise reimbursement form and the second reimbursement rationality of the user's current enterprise reimbursement form, respectively, and then performing weighted summation to obtain the target reimbursement rationality of the user's current enterprise reimbursement form.

5. The enterprise management method based on big data according to claim 1, characterized in that: The determining of the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form includes: When the target reimbursement rationality is higher than or equal to the preset threshold, the user's current enterprise reimbursement form is automatically approved; If the target reimbursement rationality is lower than the preset threshold, it will be submitted for manual review.

6. An enterprise management system based on big data, characterized in that: include: The acquisition module is used to obtain the user's current enterprise expense reimbursement form and the user's historical enterprise expense reimbursement form; A sorting module, used to sort the current enterprise expense reimbursement form of the user and the historical enterprise expense reimbursement form of the user into time series data of the enterprise expense reimbursement form of the user based on the time sequence; A time series data processing module, used to determine a first reimbursement rationality of the user's current enterprise reimbursement form based on the user's enterprise reimbursement form time series data using a time series data processing model; A graph neural network module, used to process the user's current enterprise expense reimbursement form based on a graph neural network model to determine a second reimbursement rationality of the user's current enterprise expense reimbursement form; A target reimbursement rationality determination module, configured to determine a target reimbursement rationality of the user's current enterprise reimbursement form based on a first reimbursement rationality of the user's current enterprise reimbursement form and a second reimbursement rationality of the user's current enterprise reimbursement form; The approval module is used to determine the approval result of the user's current enterprise expense reimbursement form based on the target reimbursement rationality of the user's current enterprise expense reimbursement form.

7. The enterprise management system based on big data according to claim 6, characterized in that: The time series data processing model is a gated recurrent unit, the input of the time series data processing model is the time series data of the user's enterprise reimbursement form, and the output of the time series data processing model is the first reimbursement rationality of the user's current enterprise reimbursement form.

8. The enterprise management system based on big data according to claim 6, characterized in that: The graph neural network module is also used to: Determining, based on the user's current enterprise expense reimbursement form, a plurality of similar approved enterprise expense reimbursement forms and a similarity between the user's current enterprise expense reimbursement form and each of the similar approved enterprise expense reimbursement forms; Constructing a graph structure, the graph structure comprising a plurality of nodes and edges between the plurality of nodes, the plurality of nodes comprising a user expense report node and a plurality of similar expense report nodes, the user expense report node being a central node, each similar expense report node establishing an edge with the user expense report node, the feature of each similar expense report node establishing an edge with the user expense report node being the similarity between the user's current enterprise expense report and each similar approved enterprise expense report, the node feature of the user expense report node being the user's current enterprise expense report, and the node feature of each similar expense report node being each similar approved enterprise expense report; The graph structure is processed based on a graph neural network model to determine the second reimbursement rationality of the user's current enterprise reimbursement form.

9. An electronic device, characterized in that: include: processor; Memory; And a computer program; wherein the computer program is stored in the memory and is configured to be executed by the processor to implement the big data-based enterprise management method as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the enterprise management method based on big data as described in any one of claims 1 to 5 is implemented.

Images