Task list processing method and system

Abstract

The invention provides a task list processing method and system, and the method comprises the steps: obtaining historical access data of a historical user, each piece of historical access data comprising an exposure task list and click data of each exposure task list; a vector set of the historical access data is determined, vectors in the vector set comprise personnel parameter values, task single parameter values and click parameter values, and the click parameter values are determined based on the click data; determining a first model based on the vector set; obtaining user access information of the target user, wherein the user access information comprises user identity information corresponding to the target user; determining a second model corresponding to the target user based on the user access information and the first model; and based on the second model, determining a recommendation task list corresponding to the target user. According to the method and the device, the presentation and / or recommendation of the task list of each user can be more accurate and personalized according to the preference of the user group and / or the user individual, so that the working efficiency of the user is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a task list processing method and system. Background Art

[0002] Engineering projects usually involve the allocation and dispatch of tasks. Task-related personnel (such as task dispatchers, order takers, etc.) often need to view the task orders in the task order list and perform corresponding operations on the task orders. However, there are often many tasks in engineering projects, and the division of labor among personnel at various levels or positions is also relatively detailed. After the source tasks are dispatched layer by layer, a large number of task orders will be generated. There may also be many task orders in the task order list of task-related personnel. How to quickly find the target task order is of great significance to the execution of the task. At the same time, different personnel pay attention to different tasks or task orders, and the actual situation of the task (such as whether it needs to be handled urgently, rectified, whether it is overdue, etc.) is also different. It is inefficient for users to manually sort the task orders. Sorting the task orders according to fixed task order information (such as task start time, task order status, etc.) lacks flexibility and is difficult to meet actual needs. Therefore, how to present (such as sorting) and / or recommend task orders is an urgent problem to be solved.

[0003] Therefore, a task order processing method and system are provided, which can flexibly sort the task list in combination with the characteristics of personnel, the characteristics of the task order and its actual situation, so that the presentation and / or recommendation of the task order is more personalized and targeted. Summary of the invention

[0004] The present invention provides a task order processing method, comprising: obtaining historical access data of historical users, each of the historical access data comprising an exposed task order and click data of each of the exposed task orders; determining a vector set of the historical access data, the vectors in the vector set comprising a personnel parameter value, a task order parameter value and a click parameter value, wherein the click parameter value is determined based on the click data; determining a first model based on the vector set; obtaining user access information of a target user, the user access information comprising user identity information corresponding to the target user; determining a second model corresponding to the target user based on the user access information and the first model; and determining a recommended task order corresponding to the target user based on the second model.

[0005] The present invention provides a task order processing system, comprising: a first acquisition module, configured to acquire historical access data of historical users, each of the historical access data comprising an exposed task order and click data of each of the exposed task orders; a vector set determination module, configured to determine a vector set of the historical access data, the vectors in the vector set comprising a personnel parameter value, a task order parameter value and a click parameter value, wherein the click parameter value is determined based on the click data; a first model determination module, configured to determine a first model based on the vector set; a second acquisition module, configured to acquire user access information of a target user, the user access information comprising user identity information corresponding to the target user; a second model determination module, configured to determine a second model corresponding to the target user based on the user access information and the first model; and a recommendation module, configured to determine a recommended task order corresponding to the target user based on the second model.

[0006] The beneficial effects that may be brought about by the present invention include but are not limited to: (1) through a large amount of historical access data corresponding to a large number of historical users, a first model that meets the preferences of most people can be obtained to quickly and comprehensively sort the task orders of any individual; (2) based on the first model, a second model is generated in combination with the information of the target individual, so that the sequence of the automatically generated task order list is more personalized and accurate; (3) the first model and / or the second model are updated regularly, so that the updated first model and / or the second model can adapt to the access situation of the task order information in the actual application scenario and the change of information of different users, so that the sorting of the task order can be more in line with the actual needs; (4) through the design of effective parameter value combinations and their influencing factors, the influencing factors in the task order that trigger the user to click on the task order can be more accurately analyzed, so that the first model and / or the second model can sort the task order more accurately, thereby meeting the needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The present invention will be further described in the form of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 is a schematic diagram of an application scenario of a task order processing system according to some embodiments of the present invention; Figure 2 is a module diagram of a task order processing system according to some embodiments of the present invention; Figure 3 is an exemplary flow chart of a task order processing method according to some embodiments of the present invention; Figure 4a is an exemplary flow chart of a method for determining a first model according to some embodiments of the present invention; Figure 4b is a schematic diagram of a process of determining a first model according to some embodiments of the present invention; Figure 5a is a schematic diagram of a process of determining a second model according to some embodiments of the present invention; Figure 5b is a schematic diagram of a process of updating a second model according to some embodiments of the present invention; Figure 6 is a schematic diagram of a process of determining a recommended task list according to some embodiments of the present invention. DETAILED DESCRIPTION

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present invention. For ordinary technicians in this field, the present invention can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0009] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0010] Unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0011] The present invention uses a flow chart to illustrate the operations performed by the system according to an embodiment of the present invention. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0012] Figure 1 It is a schematic diagram of an application scenario of a task order processing system according to some embodiments of the present invention.

[0013] like Figure 1As shown, the application scenario 100 of the task order processing system includes a processing device 110 , a network 120 , a terminal 130 , a storage device 140 and a task information database 150 .

[0014] The processing device 110 can process the data and / or information obtained from the terminal 130, the storage device 140 and the task information database 150. For example, the processing device 110 can obtain the task-related information from the task information database 150 and generate a task list based on the task-related information. For another example, the processing device 110 can generate a first model and / or a second model corresponding to the target user based on the task-related information to sort the task list. More information about the first model and the second model can be found elsewhere in the present invention (for example, Figure 3 ).

[0015] In some embodiments, processing device 110 can be a single server or a server group. In some embodiments, processing device 110 can be local or remote. Processing device 110 can be directly connected to terminal 130, storage device 140 and task information database 150 to access stored or acquired information and / or data. In some embodiments, processing device 110 can be implemented on a cloud platform. Just as an example, a cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc. or any combination thereof. In some embodiments, processing device 110 can be a distributed server group, which can include multiple server nodes.

[0016] The network 120 may include any suitable network that facilitates information and / or data exchange of the application scenario 100 of the task order processing system. In some embodiments, one or more components of the application scenario 100 of the task order processing system (e.g., the terminal 130, the processing device 110, the storage device 140, or the task information database 150) may communicate information and / or data with one or more other components of the application scenario 100 of the task order processing system via the network 120. For example, the processing device 110 may obtain task-related information from the storage device 140 and / or the task information database 150 via the network 120.

[0017] In some embodiments, the network 120 may be any one or more of a wired network or a wireless network. In some embodiments, the network may be a point-to-point, shared, centralized, or other topological structure or a combination of multiple topological structures.

[0018] The terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, etc., or any combination thereof. In some embodiments, the terminal 130 may interact with other components in the application scenario 100 of the task order processing system through the network 120. In some embodiments, the terminal 130 may receive information and / or instructions input by a user, and send the received information and / or instructions to the processing device 110 via the network 120. For example, the terminal 130 may receive instructions from a user (such as a dispatcher or a receiver), obtain task-related information from the storage device 140 and / or the task information database 150 through the network 120, and present one or more task orders in the form of a list.

[0019] In some embodiments, the application scenario 100 of the task list processing system also includes a preset client application, which can be a software or application installed on the terminal 130, for example, it can be a mobile phone application installed on the mobile device 130-1, a desktop application on the laptop 130-3, etc. In some embodiments, the client application includes a visual user interface to present task related information to the user. For example, the user interface presents a task list, which contains multiple task lists sorted based on a sorting algorithm. Among them, the sorting algorithm can be preset / default, or specified by the user. In some embodiments, the sorting algorithm can also be implemented based on the first model or the second model.

[0020] In some embodiments, the client application can also interact with users (such as dispatchers, order takers, etc.) through the user interface. For example, the client application can respond to the user's operation on the task list presented on the user interface and generate a corresponding operation event record. Exemplarily, it can respond to the user's click event on the task list (such as a mouse single / double-click event, a gesture click event, etc.) and generate a click event record. The click event record includes but is not limited to the operator ID, the task list ID of the clicked task list, the click time, the number of clicks, etc. In some embodiments, the client application can send the operation event record (such as the click event record) to the storage device 140 and / or the task information database 150 via the network 120 for storage. In some embodiments, the operation event record can be generated by the processing device 110.

[0021] In some embodiments, the processing device 110 can perform deduplication processing on the click event records. For example, the processing device 110 can obtain multiple click event records from the task information database 150, and perform deduplication processing based on the operator ID, task order ID, click time and preset deduplication time threshold in the multiple click event records. Exemplarily, in response to the operator ID and task order ID in the multiple click event records being the same, and the difference between the multiple click times being within the deduplication time threshold (such as 2s), the processing device 110 only stores one of the click event records (such as the earliest click event record).

[0022] In some embodiments, the client application can perform deduplication processing on the click event based on the deduplication time threshold (such as 2 seconds). For example, when the user performs a click operation multiple times (such as 3 times) within the deduplication time threshold, the client application only sends the click event record corresponding to the first click event to the storage device 140 for storage. By performing deduplication processing on the click event in the client application, data transmission can be reduced, and the data processing pressure of the processing device 110 can be reduced.

[0023] The storage device 140 may store data and / or instructions. In some embodiments, the storage device 140 may store data obtained from the processing device 110, the terminal 130, and / or the task information database 150. For example, the storage device 140 may store data (such as task-related information) obtained from the task information database 150, etc. In some embodiments, the storage device 140 may store data and / or instructions used by the processing device 110 to execute the exemplary methods described in the present invention. For example, the storage device 140 may store instructions for the processing device 110 to execute the methods shown in the various flow charts. In some embodiments, the storage device 140 may include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 140 may be implemented on a cloud platform. In some embodiments, the storage device 140 may be part of the processing device 110.

[0024] The task information database 150 refers to a source for providing task-related data. For example, it can be a database of an operating entity (such as an enterprise, a construction unit, etc.) and / or a third-party service platform (such as an information service provider). In some embodiments, the task information database 150 can be used to provide various types of task-related information. For example, task-related information includes the values ​​of task order parameters (such as task ID, creation time, dispatch time, order acceptance time, associated subtasks, task execution status, etc.), and the values ​​of personnel parameters (such as the id, name, position, etc. of the dispatcher, the receiver, the person in charge, the executor, etc.). For another example, task-related information may include historical access information of the task order (such as the number of times it has been queried or clicked, etc.). It should be noted that task-related information may include various forms of information such as text, graphics, sound, and video.

[0025] In some embodiments, the task information database 150 can interact with other components in the application scenario 100 of the task order processing system through the network 120. For example, it can send task-related information (such as task order information, personnel information, etc.) to the processing device 110 through the network 120, so that the processing device 110 performs analysis and / or processing of the task order. For another example, the task information database 150 can send task-related information to the terminal 130 through the network 120, so that the terminal 130 presents the task-related information to the user. In some embodiments, the task information database 150 can be integrated or deployed in the storage device 140.

[0026] The above description is for illustrative purposes only, and actual application scenarios may vary.

[0027] It should be noted that the application scenario 100 of the task list processing system is provided for illustrative purposes only and is not intended to limit the scope of the present invention. For those of ordinary skill in the art, various modifications or variations can be made based on the description of the present invention. However, these variations and modifications will not deviate from the scope of the present invention.

[0028] Figure 2 is a module diagram of a task order processing system according to some embodiments of the present invention.

[0029] like Figure 2 As shown, the task order processing system 200 may include a first acquisition module 210 , a vector set determination module 220 , a first model determination module 230 , a second acquisition module 240 , a second model determination module 250 , and a recommendation module 260 .

[0030] The first acquisition module 210 is configured to acquire historical access data of historical users, each of which includes an exposure task list and click data of each exposure task list.

[0031] The vector set determination module 220 is configured to determine a vector set of historical access data, where the vectors in the vector set include personnel parameter values, task order parameter values, and click parameter values, wherein the click parameter values ​​are determined based on the click data.

[0032] In some embodiments, the vector set includes a vector corresponding to each exposure task order of each historical access data. The vector set determination module 220 is further configured to: for each exposure task order of each historical access data, determine a task order sub-vector based on the task order parameter value of the exposure task order; determine a personnel sub-vector based on the personnel parameter value of the exposure task order; determine a click sub-vector based on the click parameter value of the exposure task order; and determine a vector corresponding to the exposure task order based on the task order sub-vector, the personnel sub-vector and the click sub-vector.

[0033] The first model determination module 230 is configured to determine a first model based on the vector set.

[0034] The first model reflects the influence of the task order related parameter values ​​on the click events of the historical users.

[0035] In some embodiments, the first model determination module 230 is further configured to: determine multiple valid parameters based on the vectors in the vector set; determine multiple valid parameter value combinations based on the multiple valid parameters, each valid parameter value combination including at least one personnel parameter value and at least one task order parameter value; determine the influencing factor of each valid parameter value combination based on the vectors in the vector set; determine the first model based on each valid parameter value combination and its influencing factor.

[0036] In some embodiments, the first model determination module 230 is further configured to: determine the occurrence frequency of each parameter based on the vectors in the vector set; and determine multiple valid parameters based on the occurrence frequency of each parameter and a preset frequency threshold.

[0037] In some embodiments, the first model determination module 230 is further configured to: determine a valid vector for each vector in the vector set based on multiple valid parameters and their corresponding activation ranges; and determine at least one valid parameter value combination based on the valid vector, wherein the multiple valid parameter value combinations include at least one valid parameter value combination corresponding to each vector.

[0038] In some embodiments, the first model determination module 230 is further configured to determine an influence factor based on the number of occurrences and / or the number of clicks of each valid parameter value combination.

[0039] The second acquisition module 240 is configured to acquire user access information of the target user, where the user access information includes user identity information corresponding to the target user.

[0040] The second model determination module 250 is configured to determine the second model corresponding to the target user based on the user access information and the first model.

[0041] The second model reflects the degree of influence of the task order related parameter values ​​on the click event of the target user.

[0042] In some embodiments, the second model determination module 250 is also configured to: determine the dedicated parameter value combination and its influencing factor corresponding to the target user based on the user identity information corresponding to the target user and the personnel parameter value in each valid parameter value combination in the first model; and determine the second model based on the dedicated parameter value combination and its influencing factor.

[0043] In some embodiments, the second model determination module 250 is further configured to: obtain the target exposure task list corresponding to the target user and the target click data of the target user on the target exposure task list; and update the second model based on the target exposure task list and the target click data.

[0044] In some embodiments, the second model determination module 250 is further configured to: obtain identity change information of the target user; determine the update-specific parameter value combination and its influencing factors corresponding to the target user based on the identity change information; and update the second model corresponding to the target user based on the update-specific parameter value combination and its influencing factors.

[0045] The recommendation module 260 is configured to determine a recommended task list corresponding to the target user based on the second model.

[0046] In some embodiments, the recommendation module 260 is also configured to: determine the target task order parameter value corresponding to each target task order based on the target task order information corresponding to the target task order related to the target user; determine multiple target parameter value combinations corresponding to each target task order based on the target task order parameter value and the target personnel parameter value of the target user; determine the recommendation weight of each target task order based on the multiple target parameter value combinations and the second model; and determine the recommended task order based on the recommendation weight.

[0047] In some embodiments, the first model determination module 230 is also configured to group the vector set to obtain multiple vector groups; determine the first model based on each vector group to obtain multiple first models; and the second model determination module 250 is further configured to select a target first model corresponding to the target user from the multiple first models; and determine the second model based on the target first model and user access information.

[0048] It should be noted that the above description of the task order processing system 200 and its modules is only for the convenience of description and does not limit the present invention to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the various modules or form a subsystem to connect with other modules without deviating from this principle. For example, the first acquisition module 210, the vector set determination module 220, the first model determination module 230, the second acquisition module 240, the second model determination module 250, and the recommendation module 260 can be different modules in the system, or one module can realize the functions of two or more of the above modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the protection scope of the present invention.

[0049] Figure 3 is an exemplary flow chart of a task order processing method according to some embodiments of the present invention.

[0050] In some embodiments, process 300 may be performed by a task order processing system. Figure 3 As shown, process 300 includes the following steps.

[0051] Step S310, obtaining historical access data of historical users, each historical access data includes an exposure task list and click data of each exposure task list.

[0052] Historical users refer to users who have accessed task order data through client applications. For example, historical users input query or retrieval instructions through client applications, and the client application can present the corresponding task order list for historical users to access. Users refer to personnel related to tasks, and can be any personnel in the operating entity (such as enterprises, construction units). Users can include task dispatchers, order takers, supervisors, responsible persons, executors, etc. Among them, tasks refer to matters that need to be performed or handled by users, which can be represented in the form of task documents (i.e. task orders). Task orders can include one or more combinations of various types of task-related information. For more information about task-related information, see Figure 1 and its description.

[0053] Historical access data refers to the record of historical users' access events to task order data in the past period of time, including various types of information related to the access event. For example, the past period of time can be within the past month, three months, six months, etc. Information related to access events includes access time, access content, historical users' interaction with access content (such as clicks), etc.

[0054] An exposed task order refers to a task order that has been presented to historical users for browsing. For example, it can be one or more task orders presented in the field of view of historical users on the interface of the client application. Exemplarily, when a historical user inputs a query or retrieval instruction through the client application, the corresponding task order list can be presented to the historical user for access. When there are many task orders in the task order list, the task orders need to be displayed in pages. For multiple task orders displayed in pages, multiple task orders in the current page and the browsed pages are exposed task orders, and task orders that have not been browsed by historical users are unexposed task orders.

[0055] In some embodiments, the processing device 110 may mark the exposure status of the task order (eg, 1 for exposed, 0 for unexposed) through a client application to determine a set of exposed task orders corresponding to each historical user.

[0056] The click data of the exposure task sheet is related to the click events of the historical users on the exposure task sheet, and includes the click parameter values ​​of the click parameters. The click parameters include the click time, the number of clicks, the ID of the historical person who executed the click event, the ID of the exposure task sheet that was clicked, etc. In some embodiments, the click data of the exposure task sheet includes the click event records after deduplication processing. Exemplarily, if the historical person ID and the exposure task sheet ID in multiple click event records are the same, and the difference between multiple click times is within the deduplication time threshold (such as 2s), then these multiple click event records will only retain one click event record, and accordingly, the corresponding multiple click events will be recorded as one click event.

[0057] In some embodiments, the historical access data also includes other relevant data of the exposure task order, such as personnel data (such as personnel parameter values) and task order data (such as task order parameters). For more information about personnel parameters and task order parameters, please refer to the description of step S320.

[0058] Step S320, determining a vector set of historical access data, where the vectors in the vector set include personnel parameter values, task order parameter values, and click parameter values.

[0059] The vector set of historical access data refers to a set constructed from multiple vectors generated from historical access data, which can reflect the data features of historical access data. Each vector in the vector set includes the type or parameter of the data feature (such as name, department) and the feature value or parameter value corresponding to the data feature (such as "Xiao Qiang", "Administrative Department").

[0060] In some embodiments, the task order processing system can generate a vector corresponding to each historical access data (such as a database record) based on the historical access data. The vector includes the values ​​of multiple elements, each element represents a parameter, and the value of the element represents a specific parameter value. Exemplarily, the vector Vector is (101, Xiaoqiang, Administration Department, Minister), where the parameters of the vector are personnel number, name, department, and position, and the parameter values ​​are 101, Xiaoqiang, Administration Department, and Minister.

[0061] It can be understood that different access data contain different information and their corresponding vectors may be different. The task order processing system may generate a vector set based on multiple vectors corresponding to multiple historical access data.

[0062] In some embodiments, the vector set includes a vector corresponding to each exposure task order of each historical access data. For each exposure task order of each historical access data, the task order processing system can determine a task order sub-vector based on the task order parameter value of the exposure task order; determine a personnel sub-vector based on the personnel parameter value of the exposure task order; determine a click sub-vector based on the click parameter value of the exposure task order; and determine a vector corresponding to the exposure task order based on the task order sub-vector, the personnel sub-vector, and the click sub-vector. Hereinafter, the vector corresponding to the exposure task order is referred to as the exposure task order vector.

[0063] The task order parameter value refers to the value of the task order parameter of the exposure task order. Exemplary task order parameters include task ID, creation time, dispatch time, order acceptance time, associated subtasks, task execution status, review status, the level to which the task order belongs, etc. The task order processing system can determine the task order subvector according to the task order parameters and parameter values ​​corresponding to each exposure task order. In some embodiments, the task order parameter value is determined based on the task order data in the historical access data obtained in step S310. In some embodiments, the task order processing system obtains the task order parameter value of the exposure task order by querying the task information database 150.

[0064] The personnel parameter value refers to the value of the personnel parameter of the exposure task order. Exemplary personnel parameters include the ID, name, age, department, position, etc. of the task-related personnel such as the dispatcher, the receiver, the person in charge, and the executor. The task order processing system can determine the personnel sub-vector according to the personnel parameters and parameter values ​​corresponding to each exposure task order. In some embodiments, the task order parameter value is determined based on the personnel data in the historical access data obtained in step S310. In some embodiments, the task order processing system obtains the personnel parameter value of the exposure task order by querying the task information database 150.

[0065] The click parameter value refers to the value of the click parameter of the exposure task order. The task order processing system can determine the click sub-vector according to the click parameter and parameter value of each exposure task order. The click parameter value can be determined based on the click data in the historical access data obtained in step S310.

[0066] It should be noted that the vector corresponding to the exposure task order may also include other parameter information, for example, parameter values ​​of access parameters (such as query time, number of queries, etc.) of the exposure task order by relevant personnel.

[0067] In some embodiments, the task list processing system may perform splicing processing on the task list sub-vector, the personnel sub-vector and the click sub-vector to obtain an exposure task list vector. For example, the exposure task list vector V is (F task 、F member 、F tap ), where F task 、F member 、F tap They represent the task list sub-vector, personnel sub-vector and click sub-vector respectively.

[0068] It should be noted that information extraction can be performed on the exposure task order according to preset rules to determine the exposure task order vector. In some embodiments, the task order processing system can determine the task order sub-vector, the personnel sub-vector and the click sub-vector according to the parameter weights of each parameter in the task order parameter, the personnel parameter and the click parameter, and then determine the exposure task order vector.

[0069] Taking the personnel sub-vector as an example, the parameter weights of personnel parameters such as user ID, work number, name, etc. can be set larger (such as greater than a threshold), and the parameter weights of personnel parameters such as age and address can be set smaller (such as less than a threshold). The task order processing system can generate a personnel sub-vector based on the personnel parameters whose parameter weights are greater than the threshold.

[0070] In some embodiments of the present invention, by setting the parameter weight of each parameter in the task order parameter, personnel parameter, and click parameter to determine the exposure task order vector, some irrelevant parameters can be eliminated according to actual needs, thereby reducing the subsequent data processing amount.

[0071] Step S330: determining a first model based on the vector set.

[0072] The first model refers to a general task order ranking model generated based on historical access data (or vector sets) of multiple historical users.

[0073] The first model can reflect the first degree of influence of task order related parameter values ​​(such as personnel parameters, task order parameters) on historical users' click events, and can therefore be used to sort task orders. For example, the first degree of influence can be represented by a numerical value in the interval [0,1]. The larger the value, the more likely the task order related parameter value is to cause historical users to click on the exposed task order. In actual scenarios (such as accessing task orders through an application), if the first degree of influence corresponding to the value of a task order parameter of a task order is large, the task order can be placed at the front of the task order list to facilitate users to view or click on it.

[0074] In some embodiments, the task order processing system can analyze the historical access data of multiple historical users to generate a first model. For example, for multiple exposure task orders of historical access data, the task order processing system can count the frequency of occurrence of different task order parameter values ​​or combinations, and the number of clicks on the exposure task orders corresponding to the task order parameter values ​​or combinations. The more clicks, the greater the first degree of influence corresponding to the task order parameter value or combination. Furthermore, the first model can sort the multiple exposure task orders according to the first degrees of influence corresponding to different task order parameter values ​​or combinations and the task order parameters in the multiple exposure task orders.

[0075] In some embodiments, the task order processing system can process the vector set to determine a first model. For more information about the first model, see Figure 4a and its description.

[0076] In some embodiments, multiple historical users can be divided into multiple different user groups, and the task order processing system can generate multiple different first models for multiple different user groups. The user group can be determined according to actual needs. For example, the user group can include but is not limited to multiple different user groups divided by department, gender, position, type of work, etc. As an example, the first model can include a first model for construction personnel, a first model for administrative personnel, etc.

[0077] In some embodiments, the task order processing system may group the vector set to obtain a plurality of vector groups; and determine the first model based on each vector group to obtain a plurality of first models. Figure 4a and its description.

[0078] Step S340: obtaining user access information of the target user, where the user access information includes user identity information corresponding to the target user.

[0079] The target user refers to a single user individual who has a need to access the task order data. For example, the target user may be a user in the administrative department. The user identity information corresponding to the target user includes but is not limited to work number, name, position, etc. In some embodiments, the user identity information also includes the user's login information (such as user ID, account number, etc.).

[0080] Step S350: Determine a second model corresponding to the target user based on the user access information and the first model.

[0081] The second model refers to a task order sorting model for target users, which can be called a dedicated sorting model. The second model can reflect the second degree of influence of task order related parameter values ​​(such as personnel parameters, task order parameters) on the target user's click event. For example, the second degree of influence can be a numerical representation in the [0,1] interval, and the larger the value, the more likely the task order related parameter value is to cause the target user to click on the exposed task order. In actual scenarios, if the second degree of influence corresponding to the value of the task order parameter of a certain task order is larger, the task order can be placed at the front of the task order list to facilitate the target user to view or click.

[0082] In some embodiments, the task order processing system may determine the second model corresponding to the target user based on the identity information of the target user and the first model.

[0083] In some embodiments, the task order processing system can select a target first model corresponding to the target user from multiple first models; and determine the second model based on the target first model and user access information. For more information about the second model, see Figure 5a and its description.

[0084] Step S360: Determine the recommended task list corresponding to the target user based on the second model.

[0085] Recommended tasks refer to tasks that the target user may be concerned about or interested in, which may trigger a click event of the target user.

[0086] In some embodiments, the task list processing system may determine the recommendation weights of multiple target tasks related to the target user based on the second model, and determine the recommended task list based on the recommendation weights. For more information about the recommended task list, see Figure 6 and its description.

[0087] In some embodiments, the task list processing system can present the recommended task list corresponding to the target user through the user interface of the client application, and can also remind the target user through text, voice, highlight display, etc.

[0088] In some embodiments, the task order processing system may update the first model according to a preset update cycle. The preset update cycle may be monthly, quarterly, etc. In some embodiments, the task order processing system may obtain the extended historical access data of historical users within the preset update cycle, and determine an extended vector set based on the extended historical access data, and then determine to update the first model based on the current vector set and the extended vector set.

[0089] Among them, the expanded historical access data refers to the historical access data of historical users newly added within the preset update cycle. Exemplarily, at the beginning of each month, the task order processing system can obtain the historical access data of historical users of the previous month as the expanded historical access data based on step S310. The expanded vector set refers to a vector set generated based on the expanded historical access data. For example, the expanded vector set can be generated for the expanded historical access data of the previous month based on step S320. Furthermore, the task order processing system can generate a new vector set from the current vector set and the expanded vector set, and generate an updated first model based on the method of step S330.

[0090] In some embodiments of the present invention, it is possible to combine the historical access data (such as clicks) of a large number of historical users to obtain a task order sorting model (i.e., the first model) with strong universality, which can be used to recommend task orders for each user. At the same time, it is possible to obtain a task order sorting model (i.e., the second model) with strong pertinence based on the user access information of the target user, thereby achieving personalized task order recommendations to meet the needs of the target user. In addition, through the first model and / or the second model, when the number of task orders is relatively large, it is possible to accurately filter and present the task orders that the user is concerned about, thereby improving work efficiency.

[0091] Figure 4a is an exemplary flow chart of a method for determining a first model according to some embodiments of the present invention.

[0092] In some embodiments, process 400 may be performed by a task order processing system. Figure 4a As shown, process 400 includes the following steps.

[0093] Step S410: determining a plurality of valid parameters based on the vectors in the vector set.

[0094] The valid parameters refer to parameters that are usually included in the task order. For example, the valid parameters include task order parameters and / or personnel parameters that appear frequently (eg, greater than a threshold) in the exposure task order.

[0095] In some embodiments, the task order processing system may determine the occurrence frequency of each parameter based on the vectors in the vector set, and determine a plurality of valid parameters based on the occurrence frequency of each parameter and a preset frequency threshold.

[0096] In some embodiments, the task order processing system can determine the frequency of occurrence of each parameter based on the number of occurrences of each parameter in the vector set and the total number of parameters in the vector set. For example, the task order processing system can count the total number of parameters contained in the vectors in the vector set and the number of occurrences of each parameter, and use the ratio of the number of occurrences of each parameter (i.e., the number of vectors containing the parameter) to the total number of parameters as the frequency of occurrence of the parameter. If the frequency of occurrence of a parameter is greater, it means that the task order usually contains this parameter (i.e., most task orders contain this parameter); conversely, it means that the task order usually does not contain this parameter (i.e., only a small number of task orders contain this parameter). Just as an example, vector A contains n1 parameters, vector B contains n2 parameters, and vector C contains n3 parameters. The frequency of occurrence of a parameter is the ratio of the number of occurrences of the parameter in vector AC to (n1+n2+n3).

[0097] In some embodiments of this specification, considering that the data structure of a task order (such as an exposure task order) (such as the parameters it contains) is usually pre-set, the data structures of different task orders (such as task orders of different departments and types of work) may be different. In addition, the generation rules of the vectors corresponding to the task orders may also be different. For example, when the user does not set a certain task order parameter, the vector corresponding to the task order may not contain the parameter, or the parameter value may be replaced by a default value to include the parameter in the vector. The frequency of occurrence of each parameter is determined by the number of occurrences of each parameter in the vector set and the total number of parameters, which can accurately reflect the overall distribution of each parameter in the vector set (such as the use situation), so as to adapt to different application scenarios (such as vector generation rules).

[0098] In some embodiments, the task order processing system may determine the frequency of occurrence of each parameter based on the number of occurrences of each parameter in the vector set and the total number of vectors in the vector set. For example, for each parameter, the task order processing system may count the total number of vectors N in the vector set. vector (such as 100), and the number of times the parameter appears in multiple vectors in the vector set N para , and based on N para With N vector The ratio of is taken as the frequency of occurrence of the parameter. For example, the total number of vectors in the vector set N vector is 100, a parameter P i The number of occurrences in the vector set N para is 80, which means that the parameter P appears in 80 of the 100 vectors in the vector set. i , then the frequency of occurrence of the parameter Pi is 80 / 100=0.8.

[0099] In some embodiments of the present specification, the frequency of occurrence of each parameter is determined by the number of occurrences of each parameter in the vector set and the total number of vectors in the vector set, which can reflect the actual distribution of each parameter in the vector (task list) and reduce the amount of calculation.

[0100] For each parameter, in response to the occurrence frequency of the parameter being greater than or equal to the frequency threshold, the task order processing system determines it as a valid parameter. In response to the occurrence frequency of the parameter being lower than the frequency threshold, the task order processing system determines it as an invalid parameter.

[0101] In some embodiments of the present invention, valid parameters are determined by the frequency of occurrence of the parameters. In subsequent analysis, only common parameters in the task list need to be analyzed, and invalid parameters that rarely appear need not be analyzed, so as to reduce the amount of calculation.

[0102] In some embodiments, the task order processing system may also determine the effective parameters according to the frequency of occurrence of each parameter and its corresponding parameter weight. The parameter weight may be pre-set according to different types of work, positions, etc. For example, for types of work on a construction site, parameters related to industrial accidents (such as whether an accident occurred, the number of injured, etc.) may be set with a larger parameter weight.

[0103] In some embodiments of the present invention, considering that some parameters may occur less frequently but are more important or require more attention, by setting parameter weights, some parameters that are occasional but very important in practical applications can be retained, thereby making the effective parameters more complete and in line with the actual situation.

[0104] In some embodiments, the task order processing system can be divided into multiple vector groups based on the vector set, where each vector group can be divided for a certain user group (such as all construction personnel, administrative personnel group, etc.). Exemplarily, according to the personnel parameters of each vector in the vector set (such as the department ID of the task executor), vectors with the same department ID can be classified into one vector group, thereby obtaining multiple vector groups corresponding to user groups in different departments. The task order processing system can process each vector group as a vector set in step S410 to determine the valid parameters corresponding to each vector group, and obtain the first model corresponding to multiple different user groups based on the following steps S420 to S440.

[0105] Step S420: Based on the multiple valid parameters, multiple valid parameter value combinations are determined, each valid parameter value combination including at least one personnel parameter value and at least one task order parameter value.

[0106] The valid parameter value combination refers to the combination of parameter values ​​corresponding to the valid parameters in the valid parameter combination. For example, for the valid parameter combination [department, execution status], the corresponding valid parameter value combination may be [Administrative Department, Not Started], [Finance Department, Completed], [Construction Department, In Progress], etc. Among them, Administrative Department, Finance Department and Construction Department are parameter values ​​corresponding to department parameters, and Not Started, Completed and In Progress are parameter values ​​corresponding to execution status parameters.

[0107] Figure 4b is a schematic diagram of determining a first model according to some embodiments of the present invention.

[0108] like Figure 4b As shown, the effective parameters 401 include multiple effective parameters {P 1 , P 2 , P 3 , ..., P n}. Multiple valid parameters can be combined into 2 (such as [P 1 , P 2 ]), a combination of three (such as [P 1 , P 2 , P 3 ]), ..., n combinations (such as [P 1 , P 2 , P 3 , ..., P n ]). In some embodiments, each valid parameter combination includes at least one personnel parameter and at least one task list parameter. As an example only, for the valid parameter combination [P 1 , P 2 , P 3 ], parameter P 1 is a personnel parameter (such as personnel id), parameter P 2 , P 3 Task order parameters (such as task name and estimated completion time).

[0109] In some embodiments, for each vector in the vector set, the task order processing system can determine whether each parameter in the vector is a valid parameter. In response to a parameter being an invalid parameter, the parameter and its parameter value in the vector are removed, and a valid vector is generated based on the remaining parameters and their parameter values ​​in the vector. Based on this method, multiple valid vectors corresponding to the vector set are determined. This method can be called a removal method.

[0110] For each vector in the vector set, the task order processing system can determine the valid vector 412 based on the valid parameters 401 and the vector set. Figure 4b As shown, the effective vector includes (V 11 , V 21 , V31 )、(V 11 , V 22 ,NULL),(V 12 , V 22 , V 31 ), (V 11 ,NULL,V 31 It should be noted that the above NULL is used to represent the parameters and parameter values ​​removed from each vector in the original vector set, and the corresponding valid vector obtained is (V 11 , V 21 , V 31 )、(V 11 , V 22 ), (V 12 , V 22 , V 31 ), (V 11 , V 31 ). V ij Represents parameter P i The jth value of .

[0111] In some embodiments, for each vector in the vector set, the task order processing system may filter out multiple parameters corresponding to multiple valid parameters from the vector, and then generate a valid vector based on the filtered parameters and parameter values ​​corresponding to the multiple parameters in the vector. Then, multiple valid vectors are determined based on the multiple vectors in the vector set. This method may be called a screening method.

[0112] Exemplarily, for each vector in the vector set, the task order processing system may determine the multiple valid parameters contained in the vector according to the valid parameter set (such as the valid parameter 401), wherein the multiple valid parameters may be represented in the form of a valid parameter combination, for example, [P 1 , P 2 ]、[P 1 , P 2 , P 3 ]、[P 1 , P 2 , P 3 , ..., P n ]. The task order processing system can obtain the parameter value corresponding to each valid parameter of the valid parameter combination from the vector to obtain the valid vector (V 11 , V 21 )、(V 11 , V 21 , V 31 )、(V 11 , V 21 , V 31 , ..., V n1 ). Among them, V 11 、V21 、V 31 ,……,V n1 Represents a single valid parameter P 1 , P 2 , P 3 ,……,P n Corresponding parameter values. For example, V 11 Represents parameter P 1 (such as department) corresponding to the parameter value (such as administrative department), V 21 Represents parameter P 2 (such as execution status) corresponding to the parameter value (such as not started), and the same applies to the others. Taking the first vector of the vector set as an example, when the vector contains only P in the valid parameter 401 1 , P 2 and P 3 When P 1 , P 2 and P 3 The corresponding parameter value V 11 、V 21 、V 31 To generate a vector (V 11 , V 21 , V 31 ), as a valid vector. Other vectors in the vector set are processed in a similar manner, thereby obtaining multiple valid vectors 412 corresponding to the vector set.

[0113] In some embodiments, for each vector in the vector set, the task order processing system can determine the valid vector based on multiple valid parameters and their corresponding activation ranges; and determine at least one valid parameter value combination based on the valid vector, wherein the multiple valid parameter value combinations include at least one valid parameter value combination corresponding to each vector.

[0114] The activation range is used to indicate the valid range of a parameter value. When the parameter value is within the activation range, it means that the parameter value is valid. The activation range can be used to reflect whether the parameter value has an impact on the user's click event. For example, when the parameter value corresponding to a parameter is outside the activation range, the task sheet containing the parameter value (such as the exposure task sheet) is usually not clicked by the user.

[0115] In some embodiments, the activation range can be preset according to the type of parameter (such as effective parameter). For example, the activation range corresponding to the task parameter "completion time" can be set to the past six months to the current time. If the completion time of a task list is six months ago, it means that it is not within the activation range. The probability of viewing or clicking on this task list is very small, or the user will not view or click on it.

[0116] In some embodiments, for each vector in the vector set, the task order processing system determines the initial valid vector corresponding to the vector based on multiple valid parameters. For example, the task order processing system processes the vector based on the elimination method or screening method described above to obtain an initial valid vector composed of valid parameters and parameter values ​​in the vector (i.e., invalid parameters and corresponding parameter values ​​will be eliminated). Furthermore, the task order processing system can further determine the valid vector based on the activation range corresponding to each valid parameter in the initial valid vector.

[0117] In some embodiments, if the parameter value of a valid parameter in the initial valid vector is not within its corresponding activation range, the valid parameter and its parameter value will be removed from the initial valid vector; the remaining valid parameters and their parameter values ​​will generate a valid vector. Based on this principle, the task order processing system can obtain multiple valid vectors based on multiple initial valid vectors and the activation range corresponding to each valid parameter.

[0118] In other embodiments, for each of the initial valid vectors, the task order processing system may also determine whether the parameter value corresponding to each valid parameter in the initial valid vector is within the activation range. In response to a valid parameter value not being within the activation range, the task order processing system may directly remove the initial valid vector. Based on this principle, multiple initial valid vectors are processed to obtain multiple valid vectors (i.e., the remaining initial valid vectors will be used as valid vectors).

[0119] In some embodiments, the task order processing system may further determine the valid parameter value combination 413 based on the parameter value corresponding to each valid parameter of each valid vector in the valid vector 412. Figure 4b As shown, valid parameter value combinations 413 include [V 11 , V 21 ]、[V 11 , V 22 ], ..., [V 11 , V 22 , ..., V nk ], k refers to the effective parameter P n For example, the task order processing system first determines the valid parameter value combination corresponding to each valid vector 412, and then removes duplicates from these valid parameter value combinations to obtain the final valid parameter value combination 413. Figure 4b , the valid parameter value combinations corresponding to the first valid vector include [V 11 , V 21 ]、[V 11 , V 31 ]、[V 21 , V 31 ]、[V 11, V 21 , V 31 ], the valid parameter value combinations corresponding to the second valid vector include [V 11 , V 22 ], these valid parameter combinations do not overlap and will all be regarded as part of the valid parameter value combination 413.

[0120] Step S430: determining the impact factor of each valid parameter value combination based on the vectors in the vector set.

[0121] The influence factor is used to reflect the influence of the effective parameter value combination on the number of users clicking on the task order. For example, the influence factor can be in the form of a numerical value, and the smaller the value, the smaller the influence, otherwise the greater the influence.

[0122] In some embodiments, the task order processing system may determine the impact factor based on the number of occurrences and / or clicks of each valid parameter value combination, wherein the number of occurrences of the valid parameter value combination refers to the number of times the valid parameter value combination appears in the valid vector 412 .

[0123] For a certain valid parameter value combination, the more times it appears, the more users access the task sheet containing this valid parameter value combination, and the greater the value of its corresponding impact factor. For other users, the greater the probability of the task sheet containing this valid parameter value combination being accessed, the more likely it should be presented to the user.

[0124] For a certain valid parameter value combination, the more clicks a task sheet containing the valid parameter value combination has, the greater its corresponding impact factor. For other users, the greater the probability that a task sheet containing the valid parameter value combination is clicked, the more priority it will be presented to the user.

[0125] In some embodiments, the task list processing system can determine the impact factor of each valid parameter value combination based on the first reference weight corresponding to the number of occurrences and the second reference weight corresponding to the number of clicks. The second reference weight is greater than the first reference weight. For example, for two valid parameter value combinations with the same number of occurrences, the valid parameter value combination with a larger number of clicks has a relatively larger corresponding impact factor.

[0126] like Figure 4b As shown, the task order processing system can obtain an impact factor 414 based on each effective parameter combination based on the effective parameter value combination 413. The impact factor 414 includes the impact factors F corresponding to the m effective parameter value combinations. 1 、Influence Factor F 2 ,……, Impact Factor F m .

[0127] In some embodiments of the present invention, through multiple effective parameter value combinations and their corresponding influencing factors, it is possible to comprehensively evaluate the visits and / or clicks on task orders by a large number of users from a large amount of historical access data, thereby reflecting the preferences or attention levels of multiple users or user groups to task orders, thereby providing a good basis for the priority presentation (such as sorting) of task orders.

[0128] In some embodiments, for each valid parameter value combination in the plurality of valid parameter value combinations, the task order processing system may respectively determine the number of times T the valid parameter value combination appears in the vector set corresponding to the historical access data (eg, the exposure task order). i times, and the click results of the exposure task order containing the valid parameter value combination (such as the number of clicks D i ), and D i With T i The ratio of is used as the influence factor corresponding to the combination of valid parameter values, and then the influence factors 414 corresponding to the multiple valid parameter value combinations 413 are obtained.

[0129] Step S440: Determine a first model based on each valid parameter value combination and its influencing factor.

[0130] The first model can be in various forms such as mathematical models. In some embodiments, the task order processing system can construct a ranking factor pair based on each valid parameter value combination and its corresponding influencing factor, and generate the first model based on the ranking factor pair set obtained based on multiple ranking factors. For example, the first model can be expressed as {G 1 :F 1 , ..., G m :F m}, where m is a positive integer greater than or equal to 1. Figure 4b , G 1 Indicates the valid parameter value combination [V 11 , V 21 ],F 1 Represents G 1 The corresponding impact factor, G m Indicates the valid parameter value combination [V 11 , V 21 , ..., V nk ],F m Represents G m The task order processing system may generate a first model 405 based on the valid parameter value combination 413 and the influence factor 414 .

[0131] In some embodiments, the task order processing system can determine the sorting strategy of multiple task orders of any user based on the first model, and obtain a sorted task order list based on the sorting strategy, wherein the task orders ranked higher in the task order list represent task orders that the user may be more interested in or pay more attention to. For example, a user expects to obtain multiple original task orders that need to be processed from the server through an application, and the task order processing system can perform multiple rounds of processing on the original task orders based on the first model to determine a sorted task order list.

[0132] As an example, we can first obtain the ranking factor pair F with the largest influencing factor in the first model. max , and obtain the effective parameter value combination G corresponding to the sorting factor max Then, match the valid parameter value combination G from multiple original task orders max The plurality of candidate task orders are selected and the priority of the plurality of candidate task orders is set to be larger, and the larger the priority, the higher the ranking. Based on the plurality of sorting factors, the plurality of original task orders and / or candidate task orders are processed respectively based on the above process, so as to determine the sorted task order list.

[0133] The task order processing system then presents the task order list on the interface of the application program, so that the user can process (eg, view, click and operate) the task order more quickly.

[0134] Some embodiments of the present invention can quickly generate a sorted task list of any user through the first model, so that the user can more quickly locate, view and / or process the task lists that he may be more concerned about, thereby improving work efficiency.

[0135] Figure 5a is a schematic diagram of determining a second model according to some embodiments of the present invention.

[0136] In some embodiments, the task order processing system can determine the dedicated parameter value combination corresponding to the target user and its corresponding influencing factor based on the user identity information corresponding to the target user and the personnel parameter value in each valid parameter value combination in the first model, and determine the second model based on the dedicated parameter value combination and its influencing factor.

[0137] In some embodiments, the first model may be a first model for all users. In other embodiments, the first model may be a target first model corresponding to a target user selected from a plurality of first models. The plurality of first models may be first models corresponding to a plurality of different user groups. The task order processing system may further determine a dedicated parameter value combination corresponding to the target user and its corresponding influencing factor based on the user identity information corresponding to the target user and the personnel parameter value in each valid parameter value combination in the target first model, and determine the second model based on the dedicated parameter value combination and its influencing factor. More information about the first model may be found elsewhere in the present invention (e.g. Figure 3 and Figure 4a ).

[0138] The dedicated parameter value combination refers to a valid parameter value combination associated with the target user. For example, at least one personal parameter value in the dedicated parameter value combination is the same as the personal parameter value of the target user.

[0139] like Figure 5a As shown, the task order processing system can determine the dedicated parameter value combination 503 based on the valid parameter value combination 413 and the user identity information 501 of the target user. The user identity information 501 of the target user can include, but is not limited to, a combination of one or more of the target user's user ID (such as 10), work number (such as 9001), name (such as Xiaoqiang), department (such as the administrative department), and other personnel parameter values.

[0140] In some embodiments, the task order processing system may match one or more personnel parameter values ​​in the user identity information 501 of the target user with the personnel parameter values ​​in the valid parameter value combination 413, thereby filtering out h dedicated parameter value combinations 503 from the valid parameter value combination 413, where h is less than or equal to m, which is understandable. The dedicated parameter value combination 503 is a subset of the valid parameter value combination 413. The task order processing system may further determine the influencing factor 504 of the filtered dedicated parameter value combination 503. For example, the dedicated parameter value combination g h Indicates a combination of dedicated parameter values ​​[V 11 , V 21 , ..., V nh ], impact factor f h Indicates a special parameter value combination g hCorresponding influencing factors. Among them, in each of the selected special parameter value combinations 503, there is at least one personnel parameter value that is the same as the personnel parameter value of the target user. For example, the value of the "department" parameter in each special parameter value combination 503 is the same as the value of the "department" parameter of the target user. Further, the task order processing system can determine the second model 505 based on the special parameter value combination 503 and the influencing factor 504. The representation of the second model 505 is similar to that of the first model 405, which will not be repeated here. For example, the second model can be represented as {g 1 :f 1 , ..., g h :f m}.

[0141] The impact factor of the dedicated parameter value combination is used to reflect the degree of influence of the dedicated effective parameter value combination on the target user's clicking on the task order. Since the dedicated parameter value combination is screened from the effective parameter value combination of the first model, the impact factor corresponding to the dedicated parameter combination in the first model can be directly used as the impact factor of the second model. In other words, the impact factor of the dedicated parameter value combination in the first model and the second model is the same.

[0142] In some embodiments, the task order processing system may also obtain historical access data corresponding to the target user (such as the exposure task order corresponding to the target user), and determine the impact factor corresponding to the dedicated parameter value combination based on the historical access data corresponding to the target user in the same manner as the impact factor of the first model. In this case, the impact factor of the dedicated parameter value combination in the first model is different from the impact factor in the second model.

[0143] In some embodiments of the present invention, the target user's identity information is used to match the dedicated parameter value combination corresponding to the target user's identity information from the first model, so that the second model conforms to the target user's identity. In addition, considering the target user's corresponding historical access data, the second model is also more consistent with the target user's preference for the task list, thereby making the second model more accurate and more targeted.

[0144] In some embodiments, the task order processing system may update the second model corresponding to the target user based on the identity change information of the target user. For example, the task order processing system may obtain the identity change information of the target user; determine the update-specific parameter value combination corresponding to the target user and its corresponding influencing factor based on the identity change information; and then update the second model corresponding to the target user based on the update-specific parameter value combination and the influencing factor.

[0145] The identity change information includes the personnel parameters of the target user that have changed and their updated personnel parameter values. For example, if the department of the target user has changed, the identity change information may include the updated department of the target user. In some embodiments, the task order processing system may match the updated personnel parameter values ​​with the personnel parameter values ​​in the valid parameter value combination, thereby filtering out the updated dedicated parameter value combination from the valid parameter value combination. The method for determining the updated dedicated parameter value combination is similar to the method for determining the dedicated parameter value combination, which will not be repeated here.

[0146] In some embodiments, Figure 5b As shown, the task order processing system can also obtain the target exposure task order 506 corresponding to the target user and the target click data 507 of the target user for the target exposure task order, and update the second model 505 based on the target exposure task order 506 and the target click data 507 to obtain an updated second model 508.

[0147] The target exposure task list refers to the task list presented to the target user for browsing. The target click data is related to the target click event of the target user on the target exposure task list. The target click data includes the click parameter value of the target click event. The target click data and the target click event are similar to the click data and click event described in step S310, and will not be repeated here.

[0148] In some embodiments, the task list processing system can determine the target effective vector corresponding to the vector of the clicked target exposure task list based on the target click data, and determine the target effective parameter value combination based on the target effective vector. The determination method of the target effective vector and the target effective parameter value combination is similar to the determination method of the effective vector and the effective parameter value combination described above. Further, if the target effective parameter value combination is included in the dedicated parameter value combination of the second model 505, the task list processing system can increase the influence factor of the dedicated parameter value combination, thereby obtaining its corresponding updated influence factor. If the target effective parameter value combination is not included in the dedicated parameter value combination of the second model 505, the task list processing system can supplement the target effective parameter value combination as a new dedicated parameter value combination, and determine its influence shadow based on its number of occurrences and / or number of clicks. In some embodiments, if the target effective parameter value combination is not included in the dedicated parameter value combination of the second model 505, the task list processing system can supplement the target effective parameter value combination as a new dedicated parameter value combination only when the number of target click events corresponding to the target effective parameter value combination exceeds the threshold.

[0149] In some embodiments, the task order processing system can set an update cycle to update the second model based on the target exposure task order and the target click data. The update cycle can be the same day, a week, etc. In some embodiments of the present invention, by setting an update cycle, the second model can be updated regularly while reducing the data processing load.

[0150] Figure 6 It is a schematic diagram of determining a recommended task list corresponding to a target user according to some embodiments of the present invention.

[0151] like Figure 6 As shown, the task order processing system can determine the target task order parameter value 602 corresponding to each target task order based on the target task order information corresponding to the target task order 601 related to the target user.

[0152] The target task list 601 may be one or more task lists queried or retrieved by the target user. Alternatively, the target task list 601 may be a task list to be processed by the target user. The target task list information includes various types of information related to the target task list 601. The target task list parameter value 602 refers to the value of the task list parameter of the target task list 601. The target person parameter value refers to the value of the person parameter of the target user. For more information about task list parameters and person parameters, see Figure 3 and its description.

[0153] In some embodiments, the task order processing system may determine multiple target parameter value combinations 603 corresponding to each target task order based on the target task order parameter value 602 and the target personnel parameter value of the target user. The target parameter value combination 603 may be a combination of at least one task order parameter value in the target task order parameter value 602 and at least one personnel parameter value in the target personnel parameter value, for example, a combination of 2 parameter values, a combination of 3 parameter values, etc.

[0154] In some embodiments, the task order processing system may determine a recommendation weight for each target task order based on the target parameter value combination 603 and the second model 505 .

[0155] The recommendation weight can reflect the priority of the target task list. The larger the recommendation weight, the more preferentially the target task list is presented to the target user. In some embodiments, the task list processing system can determine the recommendation weight of each target task list based on the target parameter value combination 603, through each dedicated valid parameter value combination in the second model 505 and its corresponding influencing factor. For example, for each target task list, the task list processing system can match its corresponding target parameter value combination with the dedicated valid parameter value combination in the second model, determine which dedicated valid parameter value combinations are the same as the target parameter value combination; and then determine the recommendation weight based on the influencing factors of these dedicated valid parameter value combinations. If the corresponding influencing factors of these dedicated valid parameter value combinations are larger, the recommendation weight is larger.

[0156] like Figure 6 As shown, the task order processing system can determine the recommendation weight 604 based on the second model 505 and the target parameter value combination 603. For example, the recommendation weight 604 includes the recommendation weight 1 of the target task order 1, the recommendation weight 2 of the target task order 2, ..., the recommendation weight n of the target task order n. Further, the recommendation weights of multiple target task orders can be sorted to obtain multiple recommended task orders 605. Figure 6 The recommended task list 1, ..., recommended task list m are shown. In some embodiments, the plurality of recommended task lists 605 may be target task lists with a recommendation weight greater than a weight threshold. In some embodiments, the plurality of recommended task lists 605 are target task lists with a recommendation weight in the top m.

[0157] Some embodiments of the present invention can quickly determine the recommended task list corresponding to the target user through the second model, thereby improving the target user's efficiency in processing the task list in actual work.

[0158] It should be noted that the above description of the relevant process is only for example and explanation, and does not limit the scope of application of the present invention. For those skilled in the art, various modifications and changes can be made to the process under the guidance of the present invention. However, these modifications and changes are still within the scope of the present invention.

[0159] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present invention. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and corrections to the present invention. Such modifications, improvements and corrections are suggested in the present invention, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present invention.

[0160] At the same time, the present invention uses specific words to describe the embodiments of the present invention. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present invention. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in the present invention does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of the present invention can be appropriately combined.

[0161] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in the present invention, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of the present invention. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of the present invention. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0162] Similarly, it should be noted that in order to simplify the description of the present invention and thus facilitate the understanding of one or more embodiments of the invention, in the above description of the embodiments of the invention, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the subject matter of the invention requires more features than those mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0163] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of the present invention are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0164] Each patent, patent application, patent application publication, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this invention are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this invention, documents that limit the broadest scope of the claims of this invention (currently or later attached to this invention) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the accompanying materials of this invention are inconsistent or conflicting with the contents of this invention, the descriptions, definitions, and / or use of terms in this invention shall prevail.

[0165] Finally, it should be understood that the embodiments described in the present invention are only used to illustrate the principles of the embodiments of the present invention. Other variations may also fall within the scope of the present invention. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present invention may be considered consistent with the teachings of the present invention. Accordingly, the embodiments of the present invention are not limited to the embodiments explicitly introduced and described in the present invention.

Claims

1. A task order processing method, characterized in that: include: Acquire historical access data of historical users, each of the historical access data includes an exposure task list and click data of each of the exposure task lists; Determine a vector set of the historical access data, wherein the vectors in the vector set include personnel parameter values, task order parameter values, and click parameter values, wherein the click parameter values ​​are determined based on the click data; Based on the vector set, determining a first model; Acquire user access information of a target user, wherein the user access information includes user identity information corresponding to the target user; Determine a second model corresponding to the target user based on the user access information and the first model; Based on the second model, a recommended task list corresponding to the target user is determined.

2. The method according to claim 1, characterized in that The determining the first model based on the vector set comprises: Determining a plurality of effective parameters based on the vectors in the vector set; Based on the multiple valid parameters, determine multiple valid parameter value combinations, each of the valid parameter value combinations includes at least one of the personnel parameter value and at least one of the task order parameter value; Determining an impact factor of each of the valid parameter value combinations based on the vectors in the vector set; The first model is determined based on each of the valid parameter value combinations and their influencing factors.

3. The method according to claim 2, characterized in that The determining of a plurality of valid parameters based on the vectors in the vector set comprises: Determining the frequency of occurrence of each parameter based on the vectors in the vector set; The plurality of valid parameters are determined based on the occurrence frequency of each of the parameters and a preset frequency threshold.

4. The method according to claim 2, characterized in that: The determining, based on the multiple valid parameters, multiple valid parameter value combinations comprises: For each vector in the vector set, Determining an effective vector based on the plurality of effective parameters and their corresponding activation ranges; At least one of the valid parameter value combinations is determined based on the valid vector, wherein the multiple valid parameter value combinations include at least one of the valid parameter value combinations corresponding to each of the vectors.

5. The method according to claim 2, characterized in that: The determining, based on the vectors in the vector set, the influencing factor of each of the valid parameter value combinations comprises: The influence factor is determined based on the number of occurrences and / or clicks of each of the valid parameter value combinations.

6. The method according to claim 2, characterized in that The determining, based on the user access information and the first model, a second model corresponding to the target user comprises: Determine the dedicated parameter value combination corresponding to the target user and its influencing factor based on the user identity information corresponding to the target user and the personnel parameter value in each of the valid parameter value combinations in the first model; Based on the combination of dedicated parameter values ​​and their influencing factors, the second model is determined.

7. The method according to claim 6, characterized in that The method further comprises: Obtaining a target exposure task list corresponding to the target user and target click data of the target user on the target exposure task list; Based on the target exposure task list and the target click data, the second model is updated.

8. The method according to claim 6, characterized in that The method further comprises: Obtaining identity change information of the target user; Based on the identity change information, determining an update-specific parameter value combination and its influencing factor corresponding to the target user; Based on the update-specific parameter value combination and its influencing factors, the second model corresponding to the target user is updated.

9. The method according to claim 3, characterized in that: The determining, based on the second model, a recommended task list corresponding to the target user comprises: Based on the target task order information corresponding to the target task order related to the target user, determining the target task order parameter value corresponding to each target task order; Based on the target task list parameter value and the target personnel parameter value of the target user, determining a plurality of target parameter value combinations corresponding to each target task list; Determining a recommendation weight of each of the target task orders based on the multiple target parameter value combinations and the second model; Based on the recommendation weight, the recommended task list is determined.

10. A task order processing system, characterized in that: include: A first acquisition module is configured to acquire historical access data of historical users, each of the historical access data includes an exposure task list and click data of each of the exposure task lists; A vector set determination module, configured to determine a vector set of the historical access data, wherein the vectors in the vector set include a personnel parameter value, a task order parameter value, and a click parameter value, wherein the click parameter value is determined based on the click data; A first model determination module is configured to determine a first model based on the vector set; A second acquisition module is configured to acquire user access information of a target user, wherein the user access information includes user identity information corresponding to the target user; A second model determination module is configured to determine a second model corresponding to the target user based on the user access information and the first model; The recommendation module is configured to determine a recommended task list corresponding to the target user based on the second model.

Images