Intelligent approver recommendation method and device, storage medium and computer equipment

Abstract

The invention discloses an intelligent approver recommendation method and device, a storage medium and computer equipment, relates to the technical field of intelligent pushing, and mainly aims at solving the problems that an approver recommendation mode in a patent approval scene is high in subjectivity, low in efficiency and difficult to adapt to complex and changeable approval requirements. Constructing an approver feature database based on the historical approval data; receiving an approval application of a handler on the flow node, and obtaining handler basic information and handler flow processing information of the handler; extracting features of the handler based on the basic information of the handler and the process processing information of the handler to obtain feature data of the handler; performing data matching processing based on the approver feature database to obtain a target approver having the highest matching degree with the handler feature data; and pushing the approval application to the target approver.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent push, and particularly to a method and device for intelligent recommendation of approvers, a storage medium, and a computer device. Background Art

[0002] In the scenario of patent examination and approval, the assignment of approvers often relies on manual judgment or fixed rules. This method has problems such as strong subjectivity, low efficiency, and difficulty in adapting to complex and changeable examination and approval requirements.

[0003] Currently, in existing workflow approval nodes, there have been related applications for intelligent recommendation of approvers. However, for the scenario of patent examination and approval, the existing methods for intelligent recommendation of approvers in workflows generally do not involve professional fields. For the examination and approval scenario of patents, which has relatively high requirements for professional fields, the existing methods for intelligent recommendation of approvers in workflows are no longer applicable. Therefore, there is an urgent need for a method for intelligent recommendation of approvers suitable for the patent examination and approval scenario. Summary of the Invention

[0004] In view of this, the present invention provides a method and device for intelligent recommendation of approvers, a storage medium, and a computer device, mainly aiming to solve the problems of strong subjectivity, low efficiency, and difficulty in adapting to complex and changeable examination and approval requirements in the approver recommendation method in the patent examination and approval scenario.

[0005] According to one aspect of the present invention, a method for intelligent recommendation of approvers is provided, including:

[0006] Obtain historical approval data, and construct an approver feature database based on the historical approval data;

[0007] Receive an approval application from a handler at a process node, and obtain the basic information of the handler and the process handling information of the handler;

[0008] Extract handler feature data based on the basic information of the handler and the process handling information of the handler;

[0009] Perform data matching processing based on the approver feature database to obtain a target approver with the highest matching degree with the handler feature data; and push the approval application to the target approver.

[0010] Further, the historical approval data includes the basic information of the approver and the process association information of the approver;

[0011] The constructing of the approver feature database based on the historical approval data includes:

[0012] Divide the basic information of the approver and the process association information of the approver into multiple association information categories;

[0013] Perform data mining and analysis on the historical approval behaviors of each approver based on the associated information categories to obtain a mining and analysis result corresponding to the approver;

[0014] Perform approver feature extraction processing based on the mining and analysis result to obtain approver feature data corresponding to each approver; and save the approver feature data to the approver feature database.

[0015] Further, the associated information categories include business scope categories and professional knowledge categories;

[0016] The performing data mining and analysis on the historical approval behaviors of each approver based on the associated information categories to obtain a mining and analysis result corresponding to the approver includes:

[0017] Analyze the business scope distribution of the approver's processed business based on the relevant information corresponding to the business scope category;

[0018] Analyze the professional expertise of the approver based on the relevant information corresponding to the professional knowledge category;

[0019] Analyze the sorting of approval time corresponding to each business scope type based on the relevant information corresponding to the business scope category.

[0020] Further, the performing approver feature extraction processing based on the mining and analysis result to obtain approver feature data corresponding to each approver includes:

[0021] Extract a business scope tendency vector corresponding to each approver based on the business scope distribution of the approver's processed business;

[0022] Extract a professional expertise vector corresponding to each approver based on the professional expertise of the approver;

[0023] Extract an approval passing efficiency vector corresponding to each approver based on the approval time sorting.

[0024] Further, the performing data matching processing based on the approver feature database to obtain the target approver with the highest matching degree to the handler feature data includes:

[0025] Obtain the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector corresponding to each approver in the approver feature database;

[0026] After normalizing the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector, a feature matrix corresponding to each approver is formed;

[0027] Calculate the similarity between the handler feature data and the feature matrix corresponding to each approver; and determine the approver with the highest similarity as the target approver.

[0028] Further, before performing data matching processing based on the approver feature database, the method further includes:

[0029] Obtain the node features of each process node;

[0030] Classify the process nodes based on the node features to obtain multiple process node types;

[0031] Mark each process node based on various process node types to obtain a marked process node type feature library.

[0032] Further, the data matching processing based on the approver feature database further includes:

[0033] Determine the association degree between each approver and each process node type based on the historical approval data;

[0034] Obtain the node features to be analyzed of the current process node; and determine the target process node type of the current process node based on the node features to be analyzed;

[0035] Determine the approvers to be analyzed associated with the target process node type from the approver feature database, so as to determine the target approver with the highest matching degree with the handler feature data from the approvers to be analyzed.

[0036] According to another aspect of the present invention, an intelligent approver recommendation device is provided, including:

[0037] A construction module, configured to obtain historical approval data and construct an approver feature database based on the historical approval data;

[0038] A receiving and obtaining module, configured to receive the approval application of the handler on the process node and obtain the basic information of the handler and the process handling information of the handler;

[0039] A feature extraction module, configured to perform handler feature extraction based on the basic information of the handler and the process handling information of the handler to obtain handler feature data;

[0040] A matching and pushing module, which is used to perform data matching processing based on the approver feature database to obtain a target approver with the highest matching degree with the applicant feature data; and push the approval application to the target approver.

[0041] Further, the historical approval data includes approver basic information and approver process association information; the construction module includes:

[0042] A division unit, which is used to divide the approver basic information and the approver process association information into multiple association information categories;

[0043] A data mining unit, which is used to perform data mining analysis on the historical approval behaviors of each approver based on the association information categories to obtain a mining analysis result corresponding to the approver;

[0044] A feature extraction unit, which is used to perform approver feature extraction processing based on the mining analysis result to obtain approver feature data corresponding to each approver; and save the approver feature data to the approver feature database.

[0045] Further, the association information categories include business scope categories and professional knowledge categories; the data mining unit is also used for:

[0046] Analyzing the business scope distribution of the approvers based on the relevant information corresponding to the business scope categories;

[0047] Analyzing the professional expertise of the approvers based on the relevant information corresponding to the professional knowledge categories;

[0048] Analyzing the sorting of approval time corresponding to each business scope type based on the relevant information corresponding to the business scope categories.

[0049] Further, the feature extraction unit is also used for:

[0050] Extracting a business scope tendency vector corresponding to each approver based on the business scope distribution of the approvers;

[0051] Extracting a professional expertise vector corresponding to each approver based on the professional expertise of the approvers;

[0052] Extracting an approval passing efficiency vector corresponding to each approver based on the sorting of approval time.

[0053] Further, the matching and pushing module is also used for:

[0054] Obtain the business scope preference vector, the professional expertise vector, and the approval passing efficiency vector corresponding to each approver in the approver feature database;

[0055] After normalizing the business scope preference vector, the professional expertise vector, and the approval passing efficiency vector, form a feature matrix corresponding to each approver;

[0056] Calculate the similarity between the processor feature data and the feature matrix corresponding to each approver; and determine the approver with the highest similarity as the target approver.

[0057] Further, the device further includes a node type analysis module, and the node type analysis module is used for:

[0058] Obtain the node features of each process node;

[0059] Based on the node features, classify the process nodes to obtain multiple process node types;

[0060] Based on various process node types, perform marking processing on each process node to obtain a marked process node type feature library.

[0061] Further, the matching and pushing module is further used for:

[0062] Based on the historical approval data, determine the association degree between each approver and each process node type;

[0063] Obtain the node features to be analyzed of the current process node; and based on the node features to be analyzed, determine the target process node type of the current process node;

[0064] Determine the approvers to be analyzed associated with the target process node type from the approver feature database, so as to determine the target approver with the highest matching degree with the processor feature data from the approvers to be analyzed.

[0065] According to another aspect of the present invention, there is provided a storage medium in which at least one executable instruction is stored, and the executable instruction causes a processor to perform operations corresponding to the above-mentioned approver intelligent recommendation method.

[0066] According to another aspect of the present invention, there is provided a computer device, including a processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0067] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the above-mentioned intelligent reviewer recommendation method.

[0068] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0069] The present invention provides an intelligent reviewer recommendation method, device, storage medium, and computer device. Compared with the prior art, by obtaining historical approval data and constructing a reviewer feature database based on the historical approval data, the present invention realizes feature extraction of reviewers, can effectively consider the professional fields of reviewers, and is the basis for ensuring higher professionalism in intelligent reviewer recommendation. The present invention extracts the characteristics of the handler through the basic information of the handler and the process processing information of the handler to obtain the handler feature data; then, based on the reviewer feature database, data matching processing is performed to obtain the target reviewer with the highest matching degree with the handler feature data, realizing the technical effect of intelligently recommending the optimal reviewer, improving the approval efficiency, reducing the manual judgment time, and accelerating the approval process. The present invention can also provide detailed recommendation reasons and data analysis, enhancing the scientificity and transparency of the approval decision.

[0070] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present invention more obvious and understandable, the following specifically describes the embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0072] Figure 1 A flowchart showing an intelligent reviewer recommendation method provided by an embodiment of the present invention is shown;

[0073] Figure 2 A flowchart showing another intelligent reviewer recommendation method provided by an embodiment of the present invention is shown;

[0074] Figure 3 A flowchart showing still another intelligent reviewer recommendation method provided by an embodiment of the present invention is shown;

[0075] Figure 4 A flowchart showing yet another intelligent reviewer recommendation method provided by an embodiment of the present invention is shown;

[0076] Figure 5 Shows a schematic structural diagram of an approval person intelligent recommendation device provided by an embodiment of the present invention;

[0077] Figure 6 Shows a schematic structural diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0078] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0079] An embodiment of the present invention provides an approval person intelligent recommendation method, as Figure 1 shown, the method includes:

[0080] 101. Obtain historical approval data and construct an approval person feature database based on the historical approval data;

[0081] In an embodiment of the present invention, the current execution end can collect and obtain historical approval data from the process database of the existing business process platform or the process database self-constructed by the business system. Among them, the historical approval data may include data around process approval such as approval person basic information, approval person process association information, process node information, etc., which are not specifically limited in the embodiment of the present invention. After the current execution end obtains the historical approval data, it also constructs an approval person feature database based on the obtained historical approval data. Among them, the approval person feature database stores the feature information of all approval persons in the system, such as the professional expertise information of the approval person, the business tendency information of the approval person, the approval passing efficiency information of the approval person, etc., which are not specifically limited in the embodiment of the present invention.

[0082] 102. Receive the approval application of the handler on the process node and obtain the handler's basic information and handler process processing information;

[0083] In an embodiment of the present invention, the current execution end receives the approval application of the handler on the process node in the business process platform. Among them, the approval application carries the unique identity code of the handler, and the current execution end can obtain the handler's basic information and handler process processing information based on the unique identity code of the handler. Among them, the handler's basic information may include the handler's name, gender, major, position, etc., which are not specifically limited in the embodiment of the present invention. Among them, the handler process processing information may include the process nodes processed by the handler, the business scope processed by the handler, etc., which are not specifically limited in the embodiment of the present invention.

[0084] 103. Extract the characteristics of the applicant based on the basic information of the applicant and the applicant process processing information to obtain applicant characteristic data;

[0085] In the embodiment of the present invention, the current execution end performs applicant characteristic extraction processing based on the applicant basic information and applicant process processing information obtained in step 102 to obtain applicant characteristic data. Among them, when extracting the characteristics of the applicant, the professional expertise, business scope tendency, etc. of the applicant can be extracted, and the embodiment of the present invention does not make specific limitations.

[0086] 104. Perform data matching processing based on the reviewer characteristic database to obtain the target reviewer with the highest matching degree with the applicant characteristic data; and push the approval application to the target reviewer.

[0087] In the embodiment of the present invention, the current execution end performs data matching processing based on the reviewer characteristic database constructed in step 101, and determines the target reviewer with the highest matching degree with the applicant characteristic data from the reviewer characteristic database. Among them, the matching processing can be processed by using a similarity calculation method, such as using a cosine similarity calculation method, an Euclidean distance method, a Pearson correlation coefficient method, etc., and the embodiment of the present invention does not make specific limitations. After the current execution end obtains the target reviewer with the highest matching degree, it pushes the approval application submitted by the applicant at the process node in step 102 to the target reviewer for approval processing. In the patent approval scenario, fully analyze the characteristics of the applicant and the reviewer, which can improve the automation of the approval process while also improving the approval efficiency and approval passing rate.

[0088] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to fully explore various characteristics of the reviewer in the patent approval scenario, enable the reviewer to give full play to their own advantages, and better serve the patent approval work, another reviewer intelligent recommendation method is provided, such as Figure 2 shown, the steps of constructing a reviewer characteristic database based on the historical approval data include:

[0089] 201. Divide the reviewer basic information and the reviewer process association information into multiple association information categories;

[0090] In an embodiment of the present invention, the current execution end extracts the basic information of the approver and the process association information of the approver from the historical approval data collected from the process database of the existing business process platform or the process database self-constructed by the business system based on the unique identity code of the approver. Among them, the basic information of the approver may include the name, gender, major, position, etc. of the approver, which is not specifically limited in the embodiments of the present invention. Among them, the process association information of the approver may include the process nodes processed by the approver, the business scope processed by the approver, the majors involved in the processes processed by the approver, the process processing timeliness of the approver, etc., which is not specifically limited in the embodiments of the present invention. After obtaining the basic information of the approver and the process association information of the approver, the current execution end divides the obtained information into multiple associated information categories. Among them, the associated information categories may include business scope categories, professional knowledge categories, etc., which are not specifically limited in the embodiments of the present invention. Among them, the professional knowledge category may include data such as the major of the approver and the educational background information of the approver; the business scope category may include the process nodes processed by the approver, the business scope processed by the approver, the majors involved in the processes processed by the approver, the process processing timeliness of the approver, etc., which is not specifically limited in the embodiments of the present invention.

[0091] 202. Perform data mining and analysis on the historical approval behaviors of each approver based on the associated information categories to obtain a mining and analysis result corresponding to the approver;

[0092] In an embodiment of the present invention, the current execution end performs data mining and analysis on the historical approval behaviors of each approver based on the associated information categories divided in step 201. Specifically, the following data mining and analysis methods may be adopted:

[0093] Method 1: Analyze the distribution of the business scope processed by the approver based on the relevant information corresponding to the business scope category;

[0094] Method 2: Analyze the professional expertise of the approver based on the relevant information corresponding to the professional knowledge category;

[0095] Method 3: Analyze the sorting of the approval time corresponding to each business scope type based on the relevant information corresponding to the business scope category.

[0096] It should be noted that in addition to the above three data mining and analysis methods that can be exemplified in the embodiments of the present invention, other data mining and analysis methods may also be set according to needs to analyze the relevant characteristics of the approver, which is not specifically limited in the embodiments of the present invention.

[0097] 203. Perform approver feature extraction processing based on the mining and analysis result to obtain approver feature data corresponding to each approver; and save the approver feature data to the approver feature database.

[0098] In the embodiment of the present invention, the current execution end performs approval person feature extraction processing based on the mining and analysis results obtained in the above step 202. Specifically, the following feature extraction methods can be adopted:

[0099] Method 1: Based on the distribution of the business scope processed by the approval person, extract the business scope tendency vector corresponding to each approval person, denoted as vector V1;

[0100] Method 2: Based on the professional expertise of the approval person, extract the professional expertise vector corresponding to each approval person, denoted as vector V2;

[0101] Method 3: Based on the sorting of the approval time used, extract the approval passing efficiency vector corresponding to each approval person, denoted as vector V3.

[0102] It should be noted that in the embodiment of the present invention, in addition to the above three feature extraction methods that can be exemplified, other features of the approval person can also be extracted as needed, and the embodiment of the present invention does not make specific limitations.

[0103] In the embodiment of the present invention, the current execution end stores the approval person feature data corresponding to the approval person in correspondence with the unique identity code of the approval person in the approval person feature database, so as to query the approval person feature data corresponding to each approval person based on the unique identity code of the approval person, improving the query efficiency of the approval person feature data.

[0104] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiment, in order to improve the matching efficiency and thus improve the processing speed of the approval process, another approval person intelligent recommendation method is provided. As Figure 3 shown, the steps perform data matching processing based on the approval person feature database to obtain the target approval person with the highest matching degree with the handler feature data, including:

[0105] 301. Obtain the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector corresponding to each approval person in the approval person feature database;

[0106] 302. After normalizing the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector, form a feature matrix corresponding to each approval person;

[0107] 303. Calculate the similarity between the handler feature data and the feature matrix corresponding to each approval person; and determine the approval person with the highest similarity as the target approval person.

[0108] In an embodiment of the present invention, the current execution end obtains the business scope preference vector V1, professional expertise vector V2, and approval passing efficiency vector V3 corresponding to each approver in the approver feature database constructed in steps 201 to 203. Then, the obtained business scope preference vector V1, professional expertise vector V2, and approval passing efficiency vector V3 are respectively normalized, that is, the range of the element values in the above vectors V1, V2, and V3 is converted to be between 0 and 1. The current execution end then forms a feature matrix corresponding to each approver with the normalized vectors; in specific matching processing, the current execution end calculates the similarity between the handler feature data and the feature matrix corresponding to each approver, and determines the approver with the highest similarity as the target approver.

[0109] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to further screen approvers from the perspective of the work process, another intelligent approver recommendation method is provided. As Figure 4 shown, before the method performs data matching processing based on the approver feature database, the method further includes:

[0110] 401. Obtain the node features of each process node;

[0111] 402. Classify the process nodes based on the node features to obtain multiple process node types;

[0112] 403. Mark each of the process nodes based on various types of the process node types to obtain a marked process node type feature library.

[0113] In an embodiment of the present invention, the current execution end collects and obtains the node features of each process node from the process database of the existing business process platform or the process database self-constructed by the business system; where the node features include the technical field, complexity, urgency, etc., and the embodiments of the present invention do not make specific limitations. Then, the current execution end classifies the process nodes based on the node features obtained in step 401 to obtain multiple process node types. Among them, the classification processing can adopt a multi-category classification model, including a mean clustering model, a neural network model, etc., and the embodiments of the present invention do not make specific limitations. Then, the current execution end marks each of the process nodes based on various types of the process node types to obtain a marked process node type feature library. The marked process node type feature library stores the marked node features of each historical process node, which can be used as the classification basis for newly generated process nodes in the system, and the embodiments of the present invention do not make specific limitations.

[0114] The step of performing data matching processing based on the approver feature database further includes:

[0115] 404. Determine the correlation degree between each approver and each of the process node types based on the historical approval data;

[0116] In an embodiment of the present invention, the current execution end obtains the process nodes processed by each approver in the historical approval data, and counts the process node types to which the process nodes processed by each approver belong, obtaining a statistical result of the process node types processed corresponding to each approver, and then determines the correlation degree between each approver and each process node type based on the statistical result of the process node types.

[0117] 405. Obtain the node features to be analyzed of the current process node; and determine the target process node type of the current process node based on the node features to be analyzed;

[0118] In an embodiment of the present invention, the current execution end obtains the node features to be analyzed of the current process node, including the technical field, complexity, urgency, etc. of the current process node, which are not specifically limited in the embodiments of the present invention. And determine the target process node type of the current process node based on the obtained node features to be analyzed. Specifically, the multi-category classification model in steps 401 to 403 can be used to perform data analysis on the node features to be analyzed of the current process node and the process node features in the process node type feature library, obtaining the type label of the most similar process node type, so as to determine the target process node type of the current process node, which is not specifically limited in the embodiments of the present invention.

[0119] 406. Determine the approvers to be analyzed associated with the target process node type from the approver feature database, so as to determine the target approver with the highest matching degree with the handler feature data from the approvers to be analyzed.

[0120] In an embodiment of the present invention, the current execution end determines the approvers to be analyzed associated with the target process node type from the approver feature database based on the correlation degree between each approver and each process node type determined in step 404, completing the first screening of the approvers. Then, the method in steps 301 to 303 is used to determine the target approver with the highest matching degree with the handler feature data from the approvers to be analyzed, completing the second screening of the approvers.

[0121] An embodiment of the present invention provides a method for intelligent recommendation of approvers. Compared with the prior art, by obtaining historical approval data and constructing an approver feature database based on the historical approval data, the present invention realizes feature extraction of approvers, can effectively consider the professional fields of approvers, and is the basis for ensuring higher professionalism in the intelligent recommendation of approvers. The present invention extracts the characteristics of the handler through the basic information of the handler and the process processing information of the handler to obtain the handler characteristic data; then, based on the approver feature database, data matching processing is performed to obtain the target approver with the highest matching degree with the handler characteristic data, realizing the technical effect of intelligently recommending the optimal approver, improving the approval efficiency, reducing the manual judgment time, and accelerating the approval process. The present invention can also provide detailed recommendation reasons and data analysis, enhancing the scientificity and transparency of the approval decision-making.

[0122] As an implementation of the method described above Figure 1 An embodiment of the present invention provides an approver intelligent recommendation device, as Figure 5 shown, the device includes:

[0123] A construction module 51, configured to obtain historical approval data and construct an approver feature database based on the historical approval data;

[0124] A receiving and obtaining module 52, configured to receive the approval application of the handler at the process node and obtain the basic information of the handler and the process processing information of the handler;

[0125] A feature extraction module 53, configured to perform handler feature extraction based on the basic information of the handler and the process processing information of the handler to obtain handler characteristic data;

[0126] A matching and pushing module 54, configured to perform data matching processing based on the approver feature database to obtain the target approver with the highest matching degree with the handler characteristic data; and push the approval application to the target approver.

[0127] Further, the historical approval data includes the basic information of the approver and the process association information of the approver; the construction module 51 includes:

[0128] A division unit, configured to divide the basic information of the approver and the process association information of the approver into multiple association information categories;

[0129] A data mining unit, configured to perform data mining analysis on the historical approval behaviors of each approver based on the association information categories to obtain the mining analysis results corresponding to the approver;

[0130] A feature extraction unit, configured to perform approver feature extraction processing based on the mining and analysis results, so as to obtain approver feature data corresponding to each approver; and save the approver feature data to the approver feature database.

[0131] Further, the associated information categories include business scope categories and professional knowledge categories; the data mining unit is further configured to:

[0132] Analyze the business scope distribution of the approvers' processed based on the relevant information corresponding to the business scope categories;

[0133] Analyze the professional expertise of the approvers based on the relevant information corresponding to the professional knowledge categories;

[0134] Analyze the sorting of approval time corresponding to each business scope type based on the relevant information corresponding to the business scope categories.

[0135] Further, the feature extraction unit is further configured to:

[0136] Extract a business scope tendency vector corresponding to each approver based on the business scope distribution of the approvers' processed;

[0137] Extract a professional expertise vector corresponding to each approver based on the professional expertise of the approvers;

[0138] Extract an approval passing efficiency vector corresponding to each approver based on the approval time sorting situation.

[0139] Further, the matching and pushing module 54 is further configured to:

[0140] Obtain the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector corresponding to each approver in the approver feature database;

[0141] After performing normalization processing on the business scope tendency vector, the professional expertise vector, and the approval passing efficiency vector, form a feature matrix corresponding to each approver;

[0142] Calculate the similarity between the applicant feature data and the feature matrix corresponding to each approver; and determine the approver with the highest similarity as the target approver.

[0143] Further, the device further includes a node type analysis module, and the node type analysis module is configured to:

[0144] Obtain the node features of each process node;

[0145] Classify the process nodes based on the node features to obtain multiple types of process nodes;

[0146] Mark each of the process nodes based on the various types of process nodes to obtain a marked process node type feature library.

[0147] Further, the matching and pushing module 54 is further configured to:

[0148] Determine the association degree between each approver and each type of process node based on the historical approval data;

[0149] Obtain the node features to be analyzed of the current process node; and determine the target process node type of the current process node based on the node features to be analyzed;

[0150] Determine the approvers to be analyzed associated with the target process node type from the approver feature database, so as to determine the target approver with the highest matching degree with the handler feature data from the approvers to be analyzed.

[0151] An embodiment of the present invention provides an intelligent approver recommendation device. Compared with the prior art, the present invention realizes the feature extraction of approvers by obtaining historical approval data and constructing an approver feature database based on the historical approval data, which can effectively consider the professional fields of approvers and is the basis for ensuring higher professionalism in intelligent approver recommendation. The present invention extracts handler feature data through the basic information of the handler and the handler process processing information; and then performs data matching processing based on the approver feature database to obtain the target approver with the highest matching degree with the handler feature data, realizing the technical effect of intelligently recommending the optimal approver, and can also improve the approval efficiency, reduce the manual judgment time, and speed up the approval process. The present invention can also provide detailed recommendation reasons and data analysis to enhance the scientificity and transparency of approval decisions.

[0152] According to an embodiment of the present invention, there is provided a storage medium storing at least one executable instruction, and the computer executable instruction can execute the approver intelligent recommendation method in any of the above method embodiments.

[0153] Figure 6 The structure diagram of a computer device provided according to an embodiment of the present invention is shown. The specific implementation of the computer device is not limited in the specific embodiments of the present invention.

[0154] As Figure 6As shown in the figure, the computer device may include: a processor 602, a communications interface 604, a memory 606, and a communication bus 608.

[0155] Among them: The processor 602, the communications interface 604, and the memory 606 communicate with each other through the communication bus 608.

[0156] The communications interface 604 is used to communicate with network elements of other devices such as clients or other servers.

[0157] The processor 602 is used to execute the program 610, and specifically can execute the relevant steps of the above-mentioned intelligent recommendation method for approvers.

[0158] Specifically, the program 610 may include program code, and the program code includes computer operation instructions.

[0159] The processor 602 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the computer device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0160] The memory 606 is used to store the program 610. The memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0161] The program 610 is specifically used to cause the processor 602 to perform the following operations:

[0162] Obtain historical approval data, and construct an approver feature database based on the historical approval data;

[0163] Receive the approval application of the handler at the process node, and obtain the basic information of the handler and the process handling information of the handler;

[0164] Extract handler features based on the basic information of the handler and the process handling information of the handler to obtain handler feature data;

[0165] Perform data matching processing based on the approver feature database to obtain the target approver with the highest matching degree with the handler feature data; and push the approval application to the target approver.

[0166] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a sequence different from that here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0167] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for intelligently recommending approvers, characterized in that: include: Acquire historical approval data, and build an approver feature database based on the historical approval data; Receive the approval application from the handler at the process node, and obtain the basic information of the handler and the handler's process processing information; Extracting the characteristics of the handler based on the basic information of the handler and the process processing information of the handler to obtain the characteristic data of the handler; Data matching is performed based on the approver feature database to obtain a target approver with the highest matching degree with the handler feature data; and the approval application is pushed to the target approver.

2. The method according to claim 1, characterized in that The historical approval data includes basic information of the approver and process-related information of the approver; The constructing of the approver feature database based on the historical approval data includes: Dividing the approver's basic information and the approver's process-related information into multiple related information categories; Performing data mining analysis on the historical approval behaviors of each of the approvers based on the associated information categories to obtain mining analysis results corresponding to the approvers; Based on the mining analysis results, the approver feature extraction process is performed to obtain the approver feature data corresponding to each approver; and the approver feature data is saved in the approver feature database.

3. The method according to claim 2, characterized in that The related information categories include business scope categories and professional knowledge categories; The data mining and analysis of the historical approval behaviors of each of the approvers based on the associated information categories to obtain the mining and analysis results corresponding to the approvers includes: Analyze the distribution of business scopes handled by approvers based on relevant information corresponding to the business scope categories; Analyze the professional expertise of the approver based on the relevant information corresponding to the professional knowledge category; Based on the relevant information corresponding to the business scope category, the approval time ranking corresponding to each business scope type is analyzed.

4. The method according to claim 3, characterized in that The extracting process of the approver features based on the mining analysis results to obtain the approver feature data corresponding to each approver includes: Based on the distribution of the business scopes handled by the approvers, extracting the business scope tendency vectors corresponding to the respective approvers; Based on the professional expertise of the approver, extract the professional expertise vector corresponding to each approver; Based on the approval time ranking, an approval efficiency vector corresponding to each approver is extracted.

5. The method according to claim 4, characterized in that The data matching process based on the approver feature database is performed to obtain the target approver with the highest matching degree with the handler feature data, including: Obtain the business scope tendency vector, the professional expertise vector and the approval efficiency vector corresponding to each approver in the approver feature database; After normalizing the business scope tendency vector, the professional expertise vector and the approval efficiency vector, a feature matrix corresponding to each approver is formed; The similarity between the characteristic data of the handler and the characteristic matrix corresponding to each approver is calculated; and the approver with the highest similarity is determined as the target approver.

6. The method according to any one of claims 1 to 5, characterized in that: Before performing data matching processing based on the approver feature database, the method further includes: Get the node characteristics of each process node; Classify the process nodes based on the node characteristics to obtain multiple process node types; Each of the process nodes is marked based on the various process node types to obtain a marked process node type feature library.

7. The method according to claim 6, characterized in that The data matching process based on the approver feature database further includes: Determine the association between each approver and each process node type based on the historical approval data; Acquire the node characteristics to be analyzed of the current process node; and determine the target process node type of the current process node based on the node characteristics to be analyzed; The approvers to be analyzed that are associated with the target process node type are determined from the approver feature database, so that the target approver with the highest matching degree with the handler feature data is determined from the approvers to be analyzed.

8. An intelligent recommendation device for approvers, characterized in that: include: A construction module is used to obtain historical approval data and construct an approver feature database based on the historical approval data; The receiving and obtaining module is used to receive the approval application of the handler at the process node and obtain the basic information of the handler and the handler's process processing information; A feature extraction module, used to extract the features of the handler based on the basic information of the handler and the process processing information of the handler, and obtain the feature data of the handler; The matching and pushing module is used to perform data matching processing based on the approver feature database to obtain the target approver with the highest matching degree with the handler feature data; and push the approval application to the target approver.

9. A storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction executes an operation corresponding to the intelligent recommendation method for approvers as described in any one of claims 1 to 7.

10. A computer device, comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the approver intelligent recommendation method according to any one of claims 1 to 7.

Images