Personnel behavior-based resource distribution determination method and system

Abstract

The invention discloses a resource distribution judgment method and system based on personnel behaviors. Through the technical scheme provided by the embodiment of the invention, the personnel video, the personnel information, the post information, the post resource issuing information and the work information of the target work when the target personnel participate in the target work on the target date are acquired. And performing action recognition and voice recognition on the personnel video, and determining action information and voice information of the target personnel during working. And determining action matching information and voice matching information of the target person based on the action information, the voice information and the work content information of the target person during work. Based on the post information, the personnel information, the post resource issuing information, the work achievement information, the action matching information and the voice matching information, the target resource quantity issued to the target personnel at the target date is determined, information of multiple dimensions is combined in the determination process, and the accuracy of the determined target resource quantity is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and system for determining resource distribution based on personnel behavior. Background Art

[0002] In the management and operation of modern enterprises and organizations, the reasonable allocation of resources is of great significance for motivating employees and ensuring work efficiency. Traditional resource allocation decisions are often based on relatively single factors, such as only based on work results or fixed job standards.

[0003] In the related art, resource allocation is usually calculated manually by financial personnel of enterprises and organizations. However, this method is not only inefficient, but also has a high probability of error.

[0004] Therefore, how to improve the efficiency and accuracy of resource distribution is a hot topic of research. Summary of the invention

[0005] The embodiment of the present application provides a method and system for determining resource distribution based on personnel behavior, which can improve the efficiency and accuracy of resource distribution. The technical solution is as follows:

[0006] On the one hand, a method for determining resource release based on personnel behavior is provided, the method comprising:

[0007] Obtaining personnel video, personnel information, position information, position resource distribution information, and work information of the target job when the target personnel participates in the target job on the target date, wherein the position information is used to describe the job responsibilities of the position of the target personnel, the position resource distribution information is used to describe the resource distribution standards of the target personnel, and the work information includes work content information and work results information;

[0008] Performing action recognition and voice recognition on the personnel video to determine the action information and voice information of the target personnel at work;

[0009] Based on the action information, voice information and work content information of the target person at work, determine the action matching information and voice matching information of the target person, wherein the action matching information is used to indicate the matching degree between the action information and the work content information, and the voice matching information is used to indicate the matching degree between the voice information and the work content information;

[0010] Based on the position information, the personnel information, the position resource issuance information, the work achievement information, the action matching information and the voice matching information, a target resource amount to be issued to the target personnel on the target date is determined.

[0011] On the one hand, a resource distribution determination system based on personnel behavior is provided, the system comprising:

[0012] An acquisition module is used to acquire a personnel video, personnel information, position information, position resource distribution information, and work information of a target person when he / she participates in a target job on a target date. The position information is used to describe the job responsibilities of the position where the target person is located. The position resource distribution information is used to describe the resource distribution standard of the target person. The work information includes work content information and work achievement information.

[0013] A recognition module, used to perform action recognition and voice recognition on the personnel video to determine the action information and voice information of the target personnel at work;

[0014] A matching information determination module, used to determine the action matching information and voice matching information of the target person based on the action information, voice information and work content information of the target person at work, wherein the action matching information is used to indicate the matching degree between the action information and the work content information, and the voice matching information is used to indicate the matching degree between the voice information and the work content information;

[0015] A resource quantity determination module is used to determine the target resource quantity to be distributed to the target personnel on the target date based on the position information, the personnel information, the position resource distribution information, the work result information, the action matching information and the voice matching information.

[0016] In a possible implementation, the action information includes multiple actions of the target person, the voice information includes multiple voice contents, and the work content information includes work type, work project and work goal. The matching information determination module is used to determine the first action matching information and the second action matching information of the target person based on the multiple actions, the work type and the work goal, the first action matching information is used to indicate the matching degree between the multiple actions and the work type, and the second action matching information is used to indicate the matching degree between the multiple actions and the work goal; based on the first action matching information and the second action matching information, determine the action matching information of the target person; based on the multiple voice contents, the work type and the work project, determine the first voice matching information and the second voice matching information of the target person, the first voice matching information is used to indicate the matching degree between the multiple voice contents and the work type, and the second voice matching information is used to indicate the matching degree between the multiple voice contents and the work project; based on the first voice matching information and the second voice matching information, determine the voice matching information of the target person.

[0017] In a possible implementation, the matching information determination module is used to cluster temporally adjacent actions among the multiple actions to obtain multiple action sets; perform feature extraction on the multiple action sets to obtain action set features of each of the action sets; determine first action matching information of the target person based on the action set features of each of the action sets and the work type; determine second action matching information of the target person based on the action set features of each of the action sets and the work goal;

[0018] The matching information determination module is used to cluster the temporally adjacent voice contents among the multiple voice contents to obtain multiple voice content segments; perform feature extraction on the multiple voice content segments to obtain voice content segment features of each of the voice content segments; determine the first voice content matching information of the target person based on the voice content segment features of each of the voice content segments and the work type; and determine the second voice content matching information of the target person based on the voice content segment features of each of the voice content segments and the work project.

[0019] In a possible implementation, the matching information determination module is used to encode the first action matching information and the second action matching information respectively based on an attention mechanism to obtain a first action matching feature of the first action matching information and a second action matching feature of the second action matching information; fuse the first action matching feature and the second action matching feature to obtain a first fused matching feature; perform multiple rounds of iterative decoding on the first fused matching feature based on the attention mechanism to obtain the action matching information of the target person;

[0020] The matching information determination module is used to encode the first voice matching information and the second voice matching information respectively based on the attention mechanism to obtain a first voice matching feature of the first voice matching information and a second voice matching feature of the second voice matching information; fuse the first voice matching feature and the second voice matching feature to obtain a second fused matching feature; perform multiple rounds of iterative decoding on the second fused matching feature based on the attention mechanism to obtain the voice matching information of the target person;

[0021] In one possible implementation, the resource quantity determination module is used to determine the initial resource quantity based on the position information, the position resource issuance information and the work results information; determine the first resource quantity correction coefficient based on the position resource issuance information, the action matching information and the personnel information; determine the second resource quantity correction coefficient based on the position resource issuance information, the voice matching information and the personnel information; determine the third resource quantity correction coefficient based on the action matching information and the voice matching information; and use the first resource quantity correction coefficient, the second resource quantity correction coefficient and the third resource quantity correction coefficient to correct the initial resource quantity to obtain the target resource quantity to be issued to the target personnel on the target date.

[0022] In one possible implementation, the resource quantity determination module is used to determine the result resource description information of the position based on the position information and the position resource issuance information, wherein the result resource description information is used to indicate the resource issuance standards corresponding to different results achieved by the position in completing the target work; and determine the initial resource quantity based on the result resource description information and the work result information.

[0023] In a possible implementation, the personnel information includes physiological parameters and personnel attributes, and the resource quantity determination module is used to determine a first reference correction coefficient based on the job resource issuance information and the action matching information; determine a second reference correction coefficient based on the action matching information and the personnel information; and merge the first reference correction coefficient and the second reference correction coefficient to obtain the first resource quantity correction coefficient.

[0024] In one possible implementation, the resource quantity determination module is used to determine a third reference correction coefficient based on the job resource issuance information and the voice matching information; determine a fourth reference correction coefficient based on the voice matching information and the personnel information; and merge the third reference correction coefficient and the fourth reference correction coefficient to obtain the second resource quantity correction coefficient.

[0025] In one possible implementation, the resource quantity determination module is used to perform feature extraction on the action matching information and the voice matching information to obtain action matching features of the action matching information and voice matching features of the voice matching information; fuse the action matching features and the voice matching features to obtain target matching features; and fully connect and normalize the target matching features to obtain the third resource quantity correction coefficient.

[0026] On the one hand, a computer device is provided, which includes one or more processors and one or more memories, wherein at least one computer program is stored in the one or more memories, and the computer program is loaded and executed by the one or more processors to implement the resource issuance determination method based on personnel behavior.

[0027] On the one hand, a computer-readable storage medium is provided, wherein at least one computer program is stored in the computer-readable storage medium, and the computer program is loaded and executed by a processor to implement the resource issuance determination method based on personnel behavior.

[0028] On the one hand, a computer program product or a computer program is provided, which includes a program code, and the program code is stored in a computer-readable storage medium. A processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned method for determining resource issuance based on personnel behavior.

[0029] Through the technical solution provided in the embodiment of the present application, the personnel video, personnel information, position information, position resource issuance information and work information of the target work of the target personnel when participating in the target work on the target date are obtained. Action recognition and voice recognition are performed on the personnel video to determine the action information and voice information of the target personnel at work. Based on the action information, voice information and the work content information of the target personnel at work, the action matching information and voice matching information of the target personnel are determined. Based on the position information, the personnel information, the position resource issuance information, the work results information, the action matching information and the voice matching information, the target resource amount to be issued to the target personnel on the target date is determined, and the target resource amount is automatically determined, and the information of multiple dimensions is combined in the determination process, thereby improving the accuracy of the determined target resource amount. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0031] Figure 1 It is a schematic diagram of an implementation environment of a method for determining resource distribution based on personnel behavior provided in an embodiment of the present application;

[0032] Figure 2 This is a flow chart of a method for determining resource distribution based on personnel behavior provided in an embodiment of the present application;

[0033] Figure 3 is a flow chart of another method for determining resource distribution based on personnel behavior provided in an embodiment of the present application;

[0034] Figure 4 It is a structural diagram of a resource distribution determination system based on personnel behavior provided in an embodiment of the present application;

[0035] Figure 5 It is a structural diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0037] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with basically the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on quantity and execution order.

[0038] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain better results.

[0039] Machine Learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, and other disciplines. It specializes in studying how computers simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge sub-models to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are in all areas of artificial intelligence.

[0040] Action recognition: Action recognition is an important branch of computer vision, which mainly studies how to classify human actions in videos or image sequences into specific action categories. This technology has a wide range of applications in video surveillance, human-computer interaction, sports analysis, medical rehabilitation and other fields.

[0041] Semantic recognition: Speech recognition, also known as Automatic Speech Recognition (ASR), aims to convert the vocabulary content in human speech into computer-readable input, such as keystrokes, binary codes or character sequences. This technology is widely used in smart homes, robots, consumer electronics, medical care and other fields.

[0042] Normalization: Mapping sequences with different value ranges to the interval (0, 1) facilitates data processing. In some cases, the normalized values ​​can be directly implemented as probabilities.

[0043] Embedded Coding: Embedded coding mathematically represents a correspondence, that is, mapping data in X space to Y space through a function F, where the function F is an injective function, and the mapping result is structural preservation. The injective function indicates that the data after mapping is uniquely corresponding to the data before mapping, and structural preservation indicates that the size relationship of the data before mapping is the same as the size relationship of the data after mapping. For example, there are data X1 and X2 before mapping, and after mapping, Y1 corresponding to X1 and Y2 corresponding to X2 are obtained. If the data X1 before mapping is greater than X2, then correspondingly, the data Y1 after mapping is greater than Y2. For words, it means mapping the words to another space to facilitate subsequent machine learning and processing.

[0044] Attention weight: It can indicate the importance of a certain data in the training or prediction process. Importance indicates the influence of input data on output data. Data with high importance has a higher corresponding attention weight value, and data with low importance has a lower corresponding attention weight value. In different scenarios, the importance of data is different. The process of training the attention weight of the model is also the process of determining the importance of data.

[0045] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0046] In the related art, the amount of resources to be issued to personnel is usually calculated manually by the financial personnel of the enterprise. The efficiency of manual calculation is low, and errors are inevitable in manual calculation, resulting in the probability of errors in the amount of resources issued to personnel. The technical solution provided in the embodiment of the present application can improve the efficiency of resource determination while improving the determination of resource amount.

[0047] Figure 1This is a schematic diagram of an implementation environment of a method for determining resource distribution based on personnel behavior provided in an embodiment of the present application, see Figure 1 , the implementation environment may include a terminal 110 and a server 140.

[0048] The terminal 110 is connected to the server 140 via a wireless network or a wired network. Optionally, the terminal 110 is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal 110 is installed and runs an application that supports resource issuance determination based on personnel behavior. The terminal 110 can obtain relevant information for collecting target resource quantities and send the collected relevant information to the server 140.

[0049] The server 140 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, distribution networks (Content Delivery Network, CDN) and big data and artificial intelligence platforms. The server 140 can provide background services for applications running on the terminal 110.

[0050] The following describes the application scenarios of the technical solution provided in the embodiment of the present application. The technical solution provided in the embodiment of the present application can be applied in the scenario of determining employee wages, and can also be applied in the scenario of determining the amount of virtual resources to be issued to employees, which is not limited in the embodiment of the present application.

[0051] The following is a description of a method for determining resource issuance based on personnel behavior provided in an embodiment of the present application. Figure 2 is a flowchart of a method for determining resource distribution based on personnel behavior provided in an embodiment of the present application, see Figure 2 Taking the execution subject as a server as an example, the method includes the following steps.

[0052] 201. The server obtains the personnel video, personnel information, position information, position resource distribution information and work information of the target personnel when they participate in the target work on the target date. The position information is used to describe the job responsibilities of the position of the target personnel, the position resource distribution information is used to describe the resource distribution standards of the target personnel, and the work information includes work content information and work results information.

[0053] Among them, the target personnel refers to the individual employees selected as the object of resource quantity determination within the enterprise or organization. The target date is a specific date predetermined and related to the work performance evaluation. The target work refers to a specific work task or project, which is the basis for the work performance evaluation and resource allocation of the target personnel. The personnel video is a video recorded when the target personnel participates in the target work on the target date. Personnel information is used to describe the basic situation of the target personnel. Position information is used to describe the job responsibilities. The job responsibilities cover the detailed description of the work tasks, the provisions of the work process, the required skills and qualification requirements, etc. Taking the software development position as an example, the job responsibilities may include using a specific programming language (such as Java, Python, etc.) for code writing, following a specific software development process (such as agile development process), and having database management skills. Position resource issuance information is used for resource issuance standards. When the issuance of resources refers to the issuance of salaries, the position resource issuance information includes salary structure (basic salary, performance salary, etc.), bonus calculation method (based on project completion, sales performance, etc.), welfare benefits (insurance, vacation, training opportunities, etc.), etc. Work content information includes work type (such as R&D, sales, customer service, etc.), work projects (such as R&D projects for specific products, sales projects for specific regions, etc.) and work goals (such as functional realization goals for R&D projects, sales goals for sales projects, etc.). Work results information includes actual results records of target personnel in target work, such as completed code function modules, sales achieved, etc.

[0054] 202. The server performs motion recognition and voice recognition on the person video to determine the motion information and voice information of the target person at work.

[0055] The action information includes multiple actions, and the voice information includes multiple voice contents.

[0056] 203. The server determines the action matching information and voice matching information of the target person based on the action information, voice information and work content information of the target person at work. The action matching information is used to indicate the degree of matching between the action information and the work content information, and the voice matching information is used to indicate the degree of matching between the voice information and the work content information.

[0057] The action matching information is used to indicate the matching degree between the action information and the work content information, and can also indicate the matching degree between the action performed by the target person when participating in the target work and the work content of the target work. The voice matching information is used to indicate the matching degree between the voice information and the work content information, and can also indicate the matching degree between the voice of the target person when participating in the target work and the work content of the target work.

[0058] 204. The server determines the target amount of resources to be distributed to the target person on the target date based on the position information, the person information, the position resource distribution information, the work result information, the action matching information and the voice matching information.

[0059] Among them, the target resource amount is the resource for issuing resources to the target personnel, and the resource is salary or virtual resource, which is not limited in the embodiment of the present application.

[0060] Through the technical solution provided in the embodiment of the present application, the personnel video, personnel information, position information, position resource issuance information and work information of the target work of the target personnel when participating in the target work on the target date are obtained. Action recognition and voice recognition are performed on the personnel video to determine the action information and voice information of the target personnel at work. Based on the action information, voice information and the work content information of the target personnel at work, the action matching information and voice matching information of the target personnel are determined. Based on the position information, the personnel information, the position resource issuance information, the work results information, the action matching information and the voice matching information, the target resource amount to be issued to the target personnel on the target date is determined, and the target resource amount is automatically determined, and the information of multiple dimensions is combined in the determination process, thereby improving the accuracy of the determined target resource amount.

[0061] The above steps 201-204 are a brief introduction to a method for determining resource issuance based on personnel behavior provided in an embodiment of the present application. The following will combine some examples to more clearly explain a method for determining resource issuance based on personnel behavior provided in an embodiment of the present application. Figure 3 Taking the execution subject as a server as an example, the method includes the following steps.

[0062] 301. The server obtains the personnel video, personnel information, position information, position resource distribution information and work information of the target personnel when they participate in the target work on the target date. The position information is used to describe the job responsibilities of the position of the target personnel, the position resource distribution information is used to describe the resource distribution standards of the target personnel, and the work information includes work content information and work results information.

[0063] Among them, the target personnel refers to the individual employees selected as the object of resource quantity determination within the enterprise or organization. The target date is a specific date predetermined and related to the work performance evaluation. The target work refers to a specific work task or project, which is the basis for the work performance evaluation and resource allocation of the target personnel. The personnel video is a video recorded when the target personnel participates in the target work on the target date. Personnel information is used to describe the basic situation of the target personnel. Position information is used to describe the job responsibilities. The job responsibilities cover the detailed description of the work tasks, the provisions of the work process, the required skills and qualification requirements, etc. Taking the software development position as an example, the job responsibilities may include using a specific programming language (such as Java, Python, etc.) for code writing, following a specific software development process (such as agile development process), and having database management skills. Position resource issuance information is used for resource issuance standards. When issuing resources refers to issuing salaries, that is, when the target resource quantity refers to the salary value, the position resource issuance information includes salary structure (basic salary, performance salary, etc.), bonus calculation method (based on project completion, sales performance, etc.), welfare benefits (insurance, vacation, training opportunities, etc.), etc. Work content information includes work type (such as R&D, sales, customer service, etc.), work projects (such as R&D projects for specific products, sales projects for specific regions, etc.) and work goals (such as functional realization goals for R&D projects, sales goals for sales projects, etc.). Work results information includes actual results records of target personnel in target work, such as completed code function modules, sales achieved, etc.

[0064] In a possible implementation, the server uses the personnel ID of the target person and the target date to query in a video database to obtain the personnel video of the target person, and the video database stores personnel videos of multiple objects when they participate in work on different dates. The server uses the personnel ID of the target person to query in the human resource management system of the enterprise to obtain the personnel information and position information of the target person. The server uses the personnel ID of the target person to query in the resource management system of the enterprise to obtain the position resource issuance information of the target person. The server uses the work ID of the target job and the personnel ID of the target person to query in the project management system to obtain the work information of the target job.

[0065] 302. The server performs action recognition on the person video to determine the action information of the target person at work.

[0066] The action information includes multiple actions of the target person.

[0067] In a possible implementation, the server inputs multiple video frames of the person's video into an action recognition model, and extracts features from the multiple video frames of the person's video through the action recognition model to obtain video frame features of each video frame. The server determines multiple actions of the target person based on the video frame features of the multiple video frames through the action recognition model to obtain the action information.

[0068] The action recognition model is trained by using multiple sample person videos and action annotations corresponding to each sample person video, and has the ability to recognize actions. In some embodiments, the action recognition model is ResNet-50 or Inception-v3.

[0069] In this implementation mode, the action recognition model can be used to realize the action recognition of the person video, and the efficiency and accuracy of the action recognition are high.

[0070] For example, the server inputs multiple video frames of the person's video into the action recognition model, and extracts features of each video frame of the person's video through the action recognition model to obtain video frame features of each video frame. The server fully connects and normalizes the video frame features of each video frame through the action recognition model to obtain the action of each video frame to obtain the action information.

[0071] For example, the server extracts frames of the person's video at preset time intervals to obtain multiple video frames of the person's video. The server inputs the multiple video frames of the person's video into the action recognition model, and performs multiple rounds of convolution on each video frame of the person's video through the action recognition model to obtain video frame features of each video frame. The server fully connects and normalizes the video frame features of each video frame through the action recognition model to obtain the action of each video frame to obtain the action information.

[0072] Among them, the preset duration is set by the technician according to the actual situation, such as being set to 1s or 1.5s, etc., and the embodiment of the present application does not limit this.

[0073] 303. The server performs voice recognition on the person's video to determine the voice information of the target person at work.

[0074] The voice information includes multiple voice contents.

[0075] In a possible implementation, the server extracts the audio of the person in the video of the person, inputs the audio of the person into a speech recognition model, performs speech recognition on the audio of the person through the speech recognition model, obtains multiple speech contents of the target person at work, and obtains the speech information.

[0076] The speech content is in text form. The speech recognition model may adopt a speech recognition model in the related art, which is not limited in the present embodiment.

[0077] 304. The server determines the action matching information of the target person based on the action information of the target person at work and the work content information, where the action matching information is used to indicate the degree of matching between the action information and the work content information.

[0078] The action matching information is used to indicate the matching degree between the action information and the work content information, and can also indicate the matching degree between the action performed by the target person when participating in the target work and the work content of the target work.

[0079] In a possible implementation, the action information includes multiple actions of the target person, the work content information includes a work type and a work target, and the server determines first action matching information and second action matching information of the target person based on the multiple actions, the work type and the work target, the first action matching information is used to indicate the degree of matching between the multiple actions and the work type, and the second action matching information is used to indicate the degree of matching between the multiple actions and the work target. The server determines the action matching information of the target person based on the first action matching information and the second action matching information.

[0080] In this implementation mode, work content information and multiple actions are processed to obtain action matching information of the target person, and the accuracy of the action matching information is relatively high.

[0081] In order to explain the above implementation more clearly, the above implementation is explained in several parts below.

[0082] In the first part, the server determines the first action matching information and the second action matching information of the target person based on the multiple actions, the work type and the work goal.

[0083] In a possible implementation, the server clusters temporally adjacent actions among the multiple actions to obtain multiple action sets. The server extracts features from the multiple action sets to obtain action set features of each action set. The server determines first action matching information of the target person based on the action set features of each action set and the work type. The server determines second action matching information of the target person based on the action set features of each action set and the work goal.

[0084] For example, the server divides the temporally adjacent and similar actions among the multiple actions into the same action set to obtain the multiple action sets. The server uses an action feature extraction model to extract features from the multiple action sets to obtain action set features of each action set. The server embeds and encodes the work type and the work target to obtain the work type feature of the work type and the work target feature of the work target. The server determines the feature similarity between the action set feature of each action set and the work type feature to obtain multiple first feature similarities, and one first feature similarity corresponds to one action set. The server weightedly fuses the multiple first feature similarities to obtain the first action matching information of the target person. The server determines the feature similarity between the action set feature of each action set and the work target feature to obtain multiple second feature similarities, and one second feature similarity corresponds to one action set. The server weightedly fuses the multiple second feature similarities to obtain the second action matching information of the target person.

[0085] The action feature extraction model is a model trained by a deep neural network (such as an autoencoder). The weight of the weighted fusion is associated with the action type corresponding to the action set.

[0086] The second part: the server determines the action matching information of the target person based on the first action matching information and the second action matching information.

[0087] In a possible implementation, the server encodes the first action matching information and the second action matching information respectively based on the attention mechanism to obtain a first action matching feature of the first action matching information and a second action matching feature of the second action matching information. The server fuses the first action matching feature and the second action matching feature to obtain a first fused matching feature. The server performs multiple rounds of iterative decoding on the first fused matching feature based on the attention mechanism to obtain the action matching information of the target person.

[0088] For example, a multi-head attention mechanism is used, and the number of heads is set to 8 according to the complexity and performance requirements of the model. The server takes the first action matching information and the second action matching information as inputs respectively, and obtains the first action matching feature of the first action matching information and the second action matching feature of the second action matching information after encoding by the multi-head attention mechanism. The server fuses the first action matching feature and the second action matching feature, for example, by weighted addition, and the weight can be determined based on experimental results or business needs. If it is found through experiments that the first action matching information contributes more to the overall action matching, it can be given a higher weight, such as 0.6, and the weight of the second action matching information is 0.4. After obtaining the first fused matching feature, the first fused matching feature is iteratively decoded for multiple rounds based on the attention mechanism, and finally the action matching information of the target person is obtained.

[0089] 305. The server determines voice matching information based on the voice information of the target person at work and the work content information, where the voice matching information is used to indicate the degree of matching between the voice information and the work content information.

[0090] The voice matching information is used to indicate the matching degree between the voice information and the work content information, and can also indicate the matching degree between the voice of the target person when participating in the target work and the work content of the target work.

[0091] In a possible implementation, the voice information includes multiple voice contents, the work content information includes work types and work items, and the server determines the first voice matching information and the second voice matching information of the target person based on the multiple voice contents, the work type and the work item, the first voice matching information is used to indicate the matching degree between the multiple voice contents and the work type, and the second voice matching information is used to indicate the matching degree between the multiple voice contents and the work item. The server determines the voice matching information of the target person based on the first voice matching information and the second voice matching information.

[0092] In order to explain the above implementation more clearly, the above implementation is explained in several parts below.

[0093] In the first part, the server determines the first voice matching information and the second voice matching information of the target person based on the multiple voice contents, the work type and the work project.

[0094] In a possible implementation, the server clusters the temporally adjacent voice contents among the multiple voice contents to obtain multiple voice content segments. The server extracts features from the multiple voice content segments to obtain voice content segment features of each voice content segment. The server determines first voice content matching information of the target person based on the voice content segment features of each voice content segment and the work type. The server determines second voice content matching information of the target person based on the voice content segment features of each voice content segment and the work project.

[0095] For example, the server divides the temporally adjacent and similar voice contents in the multiple voice contents into the same voice content segment to obtain the multiple voice content segments. The server uses a voice content feature extraction model to extract features from the multiple voice content segments to obtain voice content segment features of each voice content segment. The server embeds the work type and the work project to obtain the work type features of the work type and the work project features of the work project. The server determines the feature similarity between the voice content segment features of each voice content segment and the work type features to obtain multiple third feature similarities, one third feature similarity corresponding to one voice content segment. The server weightedly fuses the multiple third feature similarities to obtain the first voice content matching information of the target person. The server determines the feature similarity between the voice content segment features of each voice content segment and the work project features to obtain multiple fourth feature similarities, one fourth feature similarity corresponding to one voice content segment. The server weightedly fuses the multiple fourth feature similarities to obtain the second voice content matching information of the target person.

[0096] The speech content feature extraction model is a model trained by a deep neural network (such as an autoencoder). The weight of the weighted fusion is associated with the speech content type corresponding to the speech content segment.

[0097] The second part: the server determines the voice matching information of the target person based on the first voice matching information and the second voice matching information.

[0098] In a possible implementation, the server encodes the first voice matching information and the second voice matching information respectively based on the attention mechanism to obtain a first voice matching feature of the first voice matching information and a second voice matching feature of the second voice matching information. The server fuses the first voice matching feature and the second voice matching feature to obtain a second fused matching feature. The server performs multiple rounds of iterative decoding on the second fused matching feature based on the attention mechanism to obtain the voice matching information of the target person.

[0099] For example, the server adopts a multi-head attention mechanism, and the number of heads is set to 4. The first voice matching information and the second voice matching information are respectively used as inputs, and after encoding by the multi-head attention mechanism, the first voice matching feature of the first voice matching information and the second voice matching feature of the second voice matching information are obtained. The server fuses the first voice matching feature and the second voice matching feature, for example, by weighted multiplication, and the weight is determined according to experiments or business needs. If it is found through analysis that the first voice matching information has a relatively small impact on the final voice matching, then it can be given a lower weight, such as 0.3, and the weight of the second voice matching information is 0.7. After obtaining the second fused matching feature, the second fused matching information is iteratively decoded for multiple rounds based on the attention mechanism, and finally the voice matching information of the target person is obtained.

[0100] 306. The server determines the target amount of resources to be distributed to the target person on the target date based on the position information, the person information, the position resource distribution information, the work result information, the action matching information, and the voice matching information.

[0101] Among them, the target resource amount is the resource for issuing resources to the target personnel, and the resource is a salary or a virtual resource, which is not limited in the embodiment of the present application. In the case where the resource is a salary, the target resource amount is the value of the salary, and in the case where the resource is a virtual resource (such as a ticket, a shopping card), the target resource amount is the resource amount of the virtual resource. In the above step 306, the target resource amount determined is the resource amount of the target date, that is, the resource amount determination with a day as the granularity is realized, and the target resource amount corresponding to the longer time range can be obtained by combining it with a longer time range.

[0102] In a possible implementation, the server determines the initial resource quantity based on the position information, the position resource issuance information, and the work result information. The server determines the first resource quantity correction coefficient based on the position resource issuance information, the action matching information, and the personnel information. The server determines the second resource quantity correction coefficient based on the position resource issuance information, the voice matching information, and the personnel information. The server determines the third resource quantity correction coefficient based on the action matching information and the voice matching information. The server uses the first resource quantity correction coefficient, the second resource quantity correction coefficient, and the third resource quantity correction coefficient to correct the initial resource quantity and obtain the target resource quantity issued to the target person on the target date.

[0103] The initial resource quantity is a benchmark resource quantity, and the first correction coefficient, the second correction coefficient and the third correction coefficient are correction coefficients used to correct the initial resource quantity, thereby improving the accuracy of the target resource quantity.

[0104] In order to explain the above implementation more clearly, the above implementation is explained in several parts below.

[0105] In the first part, the server determines the initial resource amount based on the position information, the position resource distribution information and the work result information.

[0106] In a possible implementation, the server determines the result resource description information of the position based on the position information and the resource distribution information of the position, and the result resource description information is used to indicate that the position completes the target work and reaches the resource distribution standard corresponding to different results. The server determines the initial resource amount based on the result resource description information and the work result information.

[0107] For example, the server extracts semantic features from the position information and the position resource release information to obtain a first semantic feature of the position information and a second semantic feature of the position resource release information. The server fuses the first semantic feature with the second semantic feature to obtain a first semantic fusion feature. The server performs multiple rounds of iterative decoding on the first semantic fusion feature based on the attention mechanism to obtain the result resource description information. The server uses the work result information to query the result resource description information to obtain the initial resource amount corresponding to the work result information.

[0108] Part 2: The server determines a first resource quantity correction coefficient based on the job resource issuance information, the action matching information and the personnel information.

[0109] In a possible implementation, the personnel information includes physiological parameters and personnel attributes, and the server determines a first reference correction coefficient based on the post resource release information and the action matching information. The server determines a second reference correction coefficient based on the action matching information and the personnel information. The server fuses the first reference correction coefficient and the second reference correction coefficient to obtain the first resource quantity correction coefficient.

[0110] Among them, physiological parameters include heart rate, blood pressure, body temperature and other parameters when the target person participates in the target work.

[0111] For example, the server extracts semantic features from the post resource release information to obtain the third semantic feature of the post resource release information. The server fuses the third semantic feature with the action matching information and performs full connection and normalization to obtain the first reference correction coefficient. The server extracts features from the physiological parameters and personnel attributes to obtain the physiological parameter features of the physiological parameters and the personnel attribute features of the personnel attributes. The server fuses the physiological parameter features, the personnel attribute features and the action matching information and performs full connection and normalization to obtain the second reference correction coefficient. The server performs weighted fusion of the first reference correction coefficient and the second reference correction coefficient to obtain the first resource quantity correction coefficient.

[0112] Among them, the weights of weighted fusion are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.

[0113] Part three: The server determines a second resource quantity correction coefficient based on the job resource issuance information, the voice matching information and the personnel information.

[0114] In a possible implementation, the server determines a third reference correction coefficient based on the job resource release information and the voice matching information. The server determines a fourth reference correction coefficient based on the voice matching information and the personnel information. The server fuses the third reference correction coefficient and the fourth reference correction coefficient to obtain the second resource quantity correction coefficient.

[0115] For example, the server extracts semantic features from the post resource release information to obtain the third semantic feature of the post resource release information. The server fuses the third semantic feature with the voice matching information and performs full connection and normalization to obtain the third reference correction coefficient. The server extracts features from the physiological parameters and personnel attributes to obtain the physiological parameter features of the physiological parameters and the personnel attribute features of the personnel attributes. The server fuses the physiological parameter features, the personnel attribute features and the voice matching information and performs full connection and normalization to obtain the fourth reference correction coefficient. The server performs weighted fusion of the third reference correction coefficient and the fourth reference correction coefficient to obtain the second resource quantity correction coefficient.

[0116] Among them, the weights of weighted fusion are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.

[0117] Alternatively, the second resource quantity correction coefficient may be determined in the following manner.

[0118] Part 4: The server determines a third resource amount correction coefficient based on the action matching information and the voice matching information.

[0119] In a possible implementation, the server performs feature extraction on the action matching information and the voice matching information to obtain the action matching feature of the action matching information and the voice matching feature of the voice matching information. The server fuses the action matching feature and the voice matching feature to obtain the target matching feature. The server performs full connection and normalization on the target matching feature to obtain the third resource quantity correction coefficient.

[0120] Part 5: The server uses the first resource quantity correction coefficient, the second resource quantity correction coefficient and the third resource quantity correction coefficient to correct the initial resource quantity to obtain the target resource quantity to be distributed to the target person on the target date.

[0121] In a possible implementation manner, the server multiplies the first resource quantity correction coefficient, the second resource quantity correction coefficient, and the third resource quantity correction coefficient with the initial resource quantity to obtain the target resource quantity.

[0122] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0123] Through the technical solution provided in the embodiment of the present application, the personnel video, personnel information, position information, position resource issuance information and work information of the target work of the target personnel when participating in the target work on the target date are obtained. Action recognition and voice recognition are performed on the personnel video to determine the action information and voice information of the target personnel at work. Based on the action information, voice information and the work content information of the target personnel at work, the action matching information and voice matching information of the target personnel are determined. Based on the position information, the personnel information, the position resource issuance information, the work results information, the action matching information and the voice matching information, the target resource amount to be issued to the target personnel on the target date is determined, and the target resource amount is automatically determined, and the information of multiple dimensions is combined in the determination process, thereby improving the accuracy of the determined target resource amount.

[0124] Figure 4 This is a schematic diagram of a resource distribution determination system based on personnel behavior provided in an embodiment of the present application. Figure 4 The system includes: an acquisition module 401, an identification module 402, a matching information determination module 403 and a resource quantity determination module 404.

[0125] Acquisition module 401 is used to obtain personnel video, personnel information, position information, position resource distribution information and work information of the target job when the target personnel participates in the target job on the target date. The position information is used to describe the job responsibilities of the position where the target personnel is located, and the position resource distribution information is used to describe the resource distribution standards of the target personnel. The work information includes work content information and work results information.

[0126] The recognition module 402 is used to perform action recognition and voice recognition on the person video to determine the action information and voice information of the target person at work.

[0127] The matching information determination module 403 is used to determine the action matching information and voice matching information of the target person based on the action information, voice information and the work content information of the target person at work. The action matching information is used to indicate the matching degree between the action information and the work content information, and the voice matching information is used to indicate the matching degree between the voice information and the work content information.

[0128] The resource quantity determination module 404 is used to determine the target resource quantity to be distributed to the target person on the target date based on the position information, the person information, the position resource distribution information, the work result information, the action matching information and the voice matching information.

[0129] In a possible implementation, the action information includes multiple actions of the target person, the voice information includes multiple voice contents, the work content information includes work type, work project and work target, and the matching information determination module 403 is used to determine the first action matching information and the second action matching information of the target person based on the multiple actions, the work type and the work target, the first action matching information is used to indicate the matching degree between the multiple actions and the work type, and the second action matching information is used to indicate the matching degree between the multiple actions and the work target. Based on the first action matching information and the second action matching information, the action matching information of the target person is determined. Based on the multiple voice contents, the work type and the work project, the first voice matching information and the second voice matching information of the target person are determined, the first voice matching information is used to indicate the matching degree between the multiple voice contents and the work type, and the second voice matching information is used to indicate the matching degree between the multiple voice contents and the work project. Based on the first voice matching information and the second voice matching information, the voice matching information of the target person is determined.

[0130] In a possible implementation, the matching information determination module 403 is used to cluster temporally adjacent actions among the multiple actions to obtain multiple action sets. Feature extraction is performed on the multiple action sets to obtain action set features of each of the action sets. Based on the action set features of each of the action sets and the work type, the first action matching information of the target person is determined. Based on the action set features of each of the action sets and the work goal, the second action matching information of the target person is determined.

[0131] The matching information determination module 403 is used to cluster the temporally adjacent voice contents in the multiple voice contents to obtain multiple voice content segments. Feature extraction is performed on the multiple voice content segments to obtain voice content segment features of each of the voice content segments. Based on the voice content segment features of each of the voice content segments and the job type, the first voice content matching information of the target person is determined. Based on the voice content segment features of each of the voice content segments and the job project, the second voice content matching information of the target person is determined.

[0132] In a possible implementation, the matching information determination module 403 is used to encode the first action matching information and the second action matching information respectively based on the attention mechanism to obtain a first action matching feature of the first action matching information and a second action matching feature of the second action matching information. The first action matching feature and the second action matching feature are fused to obtain a first fused matching feature. The first fused matching feature is iteratively decoded for multiple rounds based on the attention mechanism to obtain the action matching information of the target person.

[0133] The matching information determination module 403 is used to encode the first voice matching information and the second voice matching information respectively based on the attention mechanism to obtain a first voice matching feature of the first voice matching information and a second voice matching feature of the second voice matching information. The first voice matching feature and the second voice matching feature are fused to obtain a second fused matching feature. The second fused matching feature is iteratively decoded for multiple rounds based on the attention mechanism to obtain the voice matching information of the target person.

[0134] In a possible implementation, the resource quantity determination module 404 is used to determine the initial resource quantity based on the position information, the position resource issuance information and the work result information. A first resource quantity correction coefficient is determined based on the position resource issuance information, the action matching information and the personnel information. A second resource quantity correction coefficient is determined based on the position resource issuance information, the voice matching information and the personnel information. A third resource quantity correction coefficient is determined based on the action matching information and the voice matching information. The initial resource quantity is corrected using the first resource quantity correction coefficient, the second resource quantity correction coefficient and the third resource quantity correction coefficient to obtain the target resource quantity issued to the target person on the target date.

[0135] In a possible implementation, the resource quantity determination module 404 is used to determine the achievement resource description information of the position based on the position information and the position resource release information, and the achievement resource description information is used to indicate that the position completes the target work and reaches the resource release standard corresponding to different achievements. Based on the achievement resource description information and the work achievement information, the initial resource quantity is determined.

[0136] In a possible implementation, the personnel information includes physiological parameters and personnel attributes, and the resource quantity determination module 404 is used to determine a first reference correction coefficient based on the post resource release information and the action matching information. A second reference correction coefficient is determined based on the action matching information and the personnel information. The first reference correction coefficient and the second reference correction coefficient are merged to obtain the first resource quantity correction coefficient.

[0137] In a possible implementation, the resource quantity determination module 404 is used to determine a third reference correction coefficient based on the job resource release information and the voice matching information. A fourth reference correction coefficient is determined based on the voice matching information and the personnel information. The third reference correction coefficient and the fourth reference correction coefficient are merged to obtain the second resource quantity correction coefficient.

[0138] In a possible implementation, the resource quantity determination module 404 is used to extract features from the action matching information and the voice matching information to obtain action matching features of the action matching information and voice matching features of the voice matching information. The action matching features and the voice matching features are fused to obtain target matching features. The target matching features are fully connected and normalized to obtain the third resource quantity correction coefficient.

[0139] It should be noted that: the above embodiment provides a resource issuance determination system based on personnel behavior, when making a resource issuance determination, only uses the division of the above functional modules as an example. In actual applications, the above functional distribution can be completed by different functional modules as needed, that is, the internal structure of the server is divided into different functional modules to complete all or part of the functions described above. In addition, the above embodiment provides a resource issuance determination system based on personnel behavior and a resource issuance determination method embodiment based on personnel behavior, which belong to the same concept. The specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0140] Through the technical solution provided in the embodiment of the present application, the personnel video, personnel information, position information, position resource issuance information and work information of the target work of the target personnel when participating in the target work on the target date are obtained. Action recognition and voice recognition are performed on the personnel video to determine the action information and voice information of the target personnel at work. Based on the action information, voice information and the work content information of the target personnel at work, the action matching information and voice matching information of the target personnel are determined. Based on the position information, the personnel information, the position resource issuance information, the work results information, the action matching information and the voice matching information, the target resource amount to be issued to the target personnel on the target date is determined, and the target resource amount is automatically determined, and the information of multiple dimensions is combined in the determination process, thereby improving the accuracy of the determined target resource amount.

[0141] Figure 5 This is a schematic diagram of the structure of a server provided in an embodiment of the present application. The server 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 501 and one or more memories 502, wherein at least one computer program is stored in the one or more memories 502, and the at least one computer program is loaded and executed by the one or more processors 501 to implement the methods provided in the above-mentioned various method embodiments. Of course, the server 500 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The server 500 may also include other components for implementing device functions, which will not be described in detail here.

[0142] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a computer program, and the computer program can be executed by a processor to complete a method for determining resource issuance based on personnel behavior in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0143] In an exemplary embodiment, a computer program product or a computer program is also provided, which includes a program code, and the program code is stored in a computer-readable storage medium. A processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned method for determining resource issuance based on personnel behavior.

[0144] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected by a communication network. Multiple computer devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.

[0145] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0146] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for determining resource distribution based on personnel behavior, characterized in that: The method comprises: Obtaining personnel video, personnel information, position information, position resource distribution information, and work information of the target job when the target personnel participates in the target job on the target date, wherein the position information is used to describe the job responsibilities of the position of the target personnel, the position resource distribution information is used to describe the resource distribution standards of the target personnel, and the work information includes work content information and work results information; Performing action recognition and voice recognition on the personnel video to determine the action information and voice information of the target personnel at work; Based on the action information, voice information and work content information of the target person at work, determine the action matching information and voice matching information of the target person, wherein the action matching information is used to indicate the matching degree between the action information and the work content information, and the voice matching information is used to indicate the matching degree between the voice information and the work content information; Based on the position information, the personnel information, the position resource issuance information, the work achievement information, the action matching information and the voice matching information, a target resource amount to be issued to the target personnel on the target date is determined.

2. The method according to claim 1, characterized in that The action information includes multiple actions of the target person, the voice information includes multiple voice contents, the work content information includes work type, work project and work goal, and determining the action matching information and voice matching information of the target person based on the action information, voice information and work content information of the target person at work includes: Based on the multiple actions, the work type and the work target, determine first action matching information and second action matching information of the target person, wherein the first action matching information is used to indicate the matching degree between the multiple actions and the work type, and the second action matching information is used to indicate the matching degree between the multiple actions and the work target; Determining the action matching information of the target person based on the first action matching information and the second action matching information; Based on the multiple voice contents, the work type and the work project, determine first voice matching information and second voice matching information of the target person, wherein the first voice matching information is used to indicate the matching degree between the multiple voice contents and the work type, and the second voice matching information is used to indicate the matching degree between the multiple voice contents and the work project; Based on the first voice matching information and the second voice matching information, the voice matching information of the target person is determined.

3. The method according to claim 2, characterized in that The determining, based on the multiple actions, the work type and the work goal, the first action matching information and the second action matching information of the target person includes: Clustering temporally adjacent actions among the multiple actions to obtain multiple action sets; extracting features from the multiple action sets to obtain action set features of each action set; determining first action matching information of the target person based on the action set features of each action set and the work type; determining second action matching information of the target person based on the action set features of each action set and the work goal; The determining, based on the multiple voice contents, the work type, and the work item, the first voice matching information and the second voice matching information of the target person includes: Clustering the temporally adjacent voice contents among the multiple voice contents to obtain multiple voice content segments; performing feature extraction on the multiple voice content segments to obtain voice content segment features of each of the voice content segments; determining first voice content matching information of the target person based on the voice content segment features of each of the voice content segments and the work type; determining second voice content matching information of the target person based on the voice content segment features of each of the voice content segments and the work project.

4. The method according to claim 2, characterized in that: The determining the action matching information of the target person based on the first action matching information and the second action matching information includes: Encode the first action matching information and the second action matching information respectively based on the attention mechanism to obtain a first action matching feature of the first action matching information and a second action matching feature of the second action matching information; Fusing the first action matching feature and the second action matching feature to obtain a first fused matching feature; Perform multiple rounds of iterative decoding on the first fused matching features based on an attention mechanism to obtain action matching information of the target person; The determining the voice matching information of the target person based on the first voice matching information and the second voice matching information includes: Encode the first voice matching information and the second voice matching information respectively based on the attention mechanism to obtain a first voice matching feature of the first voice matching information and a second voice matching feature of the second voice matching information; Fusing the first voice matching feature and the second voice matching feature to obtain a second fused matching feature; Based on the attention mechanism, the second fusion matching feature is iteratively decoded for multiple rounds to obtain the voice matching information of the target person.

5. The method according to claim 1, characterized in that The determining, based on the position information, the personnel information, the position resource issuance information, the work achievement information, the action matching information, and the voice matching information, of a target resource amount to be issued to the target personnel on the target date includes: Determine an initial resource amount based on the position information, the position resource release information, and the work result information; Determine a first resource quantity correction coefficient based on the post resource issuance information, the action matching information and the personnel information; Determining a second resource quantity correction coefficient based on the job resource issuance information, the voice matching information and the personnel information; Determining a third resource amount correction coefficient based on the action matching information and the voice matching information; The initial resource quantity is corrected by using the first resource quantity correction coefficient, the second resource quantity correction coefficient, and the third resource quantity correction coefficient to obtain the target resource quantity to be distributed to the target person on the target date.

6. The method according to claim 5, characterized in that The determining of the initial resource amount based on the position information, the position resource release information and the work result information includes: Based on the position information and the position resource release information, determining the achievement resource description information of the position, wherein the achievement resource description information is used to indicate the resource release standards corresponding to different achievements achieved by the position in completing the target work; The initial resource quantity is determined based on the result resource description information and the work result information.

7. The method according to claim 5, characterized in that The personnel information includes physiological parameters and personnel attributes, and determining the first resource quantity correction coefficient based on the post resource issuance information, the action matching information and the personnel information includes: Determining a first reference correction coefficient based on the job resource issuance information and the action matching information; Determining a second reference correction coefficient based on the action matching information and the personnel information; The first reference correction coefficient and the second reference correction coefficient are combined to obtain the first resource quantity correction coefficient.

8. The method according to claim 5, characterized in that The determining of the second resource quantity correction coefficient based on the post resource release information, the voice matching information and the personnel information includes: Determining a third reference correction coefficient based on the job resource release information and the voice matching information; Determining a fourth reference correction coefficient based on the voice matching information and the personnel information; The third reference correction coefficient and the fourth reference correction coefficient are combined to obtain the second resource quantity correction coefficient.

9. The method according to claim 5, characterized in that The determining a third resource amount correction coefficient based on the action matching information and the voice matching information includes: Performing feature extraction on the action matching information and the voice matching information to obtain action matching features of the action matching information and voice matching features of the voice matching information; The action matching feature and the voice matching feature are fused to obtain a target matching feature; The target matching features are fully connected and normalized to obtain the third resource quantity correction coefficient.

10. A resource distribution determination system based on personnel behavior, characterized in that: An acquisition module is used to acquire a personnel video, personnel information, position information, position resource distribution information, and work information of a target person when he / she participates in a target job on a target date. The position information is used to describe the job responsibilities of the position where the target person is located. The position resource distribution information is used to describe the resource distribution standard of the target person. The work information includes work content information and work achievement information. A recognition module, used to perform action recognition and voice recognition on the personnel video, and determine the action information and voice information of the target personnel at work; A matching information determination module, used to determine the action matching information and voice matching information of the target person based on the action information, voice information and work content information of the target person at work, wherein the action matching information is used to indicate the matching degree between the action information and the work content information, and the voice matching information is used to indicate the matching degree between the voice information and the work content information; A resource quantity determination module is used to determine the target resource quantity to be distributed to the target personnel on the target date based on the position information, the personnel information, the position resource distribution information, the work result information, the action matching information and the voice matching information.

Images