Project processing method and device

Abstract

The embodiments of this specification provide a project processing method and device, wherein the project processing method includes: based on the project attributes of the initial project and the user attributes of the project users under the initial project, classifying the project users into categories to generate at least two classification results, splitting the project indicators of the initial project into multiple sub-project indicators, and determining the initial indicator value and target indicator value of the target project user contained in the target classification result in the multiple sub-project indicators, wherein the target classification result is one of the at least two classification results, determining the target processing path of the target project user based on the initial indicator value and the target indicator value, and processing the multiple sub-project indicators of the target project user based on the target processing path.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of computer technology, and more particularly to a project processing method. One or more embodiments of this specification also relate to a project processing device, a computing device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of the economy and the increasing diversity of economic development methods, more and more financial instruments have emerged. Many companies or users can borrow from different financial instruments to alleviate short-term economic pressure.

[0003] At present, loan funds, as an important part of the loan projects of various financial institutions, often have a great impact on the cash flow of financial institutions. The daily loan quota is an important goal and decision-making reference for financial institutions in daily management. However, due to the uncertainty of user behavior and market environment, the loan quota of financial institutions often fluctuates. In addition, financial institutions are currently unable to reach users according to the needs of different users and conduct long-term serial operations, resulting in a low conversion rate of user resources. Therefore, an effective method is urgently needed to solve such problems. Summary of the invention

[0004] In view of this, an embodiment of this specification provides a project processing method. One or more embodiments of this specification also relate to a project processing device, a computing device, and a computer-readable storage medium to solve the technical defects existing in the prior art.

[0005] According to a first aspect of an embodiment of this specification, a project processing method is provided, including:

[0006] Based on the project attributes of the initial project and the user attributes of the project users under the initial project, the project users are classified and at least two classification results are generated;

[0007] Splitting the project indicator of the initial project into multiple sub-project indicators, and determining the initial indicator value and the target indicator value of the target project user included in the target division result in the multiple sub-project indicators, wherein the target division result is one of the at least two division results;

[0008] A target processing path of the target project user is determined based on the initial indicator value and the target indicator value, and the multiple sub-project indicators of the target project user are processed based on the target processing path.

[0009] Optionally, the determining of the initial indicator value and the target indicator value of the target project user in the plurality of sub-project indicators included in the target division result includes:

[0010] Determine the target project user included in the target division result, and input the user attribute of the target project user and the multiple sub-project indicators into the indicator value prediction model for processing to generate the target indicator value of the target project user in the multiple sub-project indicators; and

[0011] Determine the initial index value of the target project user included in the target division result in the plurality of sub-project indicators.

[0012] Optionally, determining a target processing path of the target project user based on the initial indicator value and the target indicator value includes:

[0013] Determining an initial key point corresponding to the target project user according to the initial indicator value, and determining a target key point of the target project user based on the target indicator value;

[0014] Determine the distance between the initial key point and the target key point according to the initial index value and the target index value;

[0015] Constructing a directed relationship network based on the initial key point, the target key point and the distance;

[0016] According to the target path algorithm, a target processing path between an initial key point corresponding to the target project user and the target key point in the directed relationship network is determined.

[0017] Optionally, determining the target processing path between the initial key point corresponding to the target project user and the target key point in the directed relationship network according to the target path algorithm includes:

[0018] The initial index values ​​of multiple sub-project indicators corresponding to the initial key points in the directed relationship network and the target index values ​​of multiple sub-project indicators corresponding to the target key points in the directed relationship network are input into the target path algorithm for calculation to generate a target processing path between each initial key point and the target key point in the directed relationship network.

[0019] Optionally, determining the target processing path between the initial key point corresponding to the target project user and the target key point in the directed relationship network according to the target path algorithm includes:

[0020] Based on the target path algorithm and the directed relationship network, an initial adjacency matrix corresponding to the initial key point and the target key point is generated, wherein the elements in the initial adjacency matrix are used to represent the initial distances between different key points;

[0021] At least one of the initial key point and the target key point is used as an intermediate key point, and based on the intermediate key point, the elements in the initial adjacency matrix are updated to generate a target adjacency matrix;

[0022] According to the target adjacency matrix, a target distance between the initial key point and the target key point is determined, and according to the target distance, a target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point is determined.

[0023] Optionally, determining a target processing path of the target project user based on the initial indicator value and the target indicator value includes:

[0024] Determine, based on the initial indicator value and the target indicator value, an initial processing path for at least two target project users in the target division result;

[0025] Determining the similarity between the initial processing paths of the at least two target project users;

[0026] When the similarity is greater than or equal to a preset similarity threshold, the initial processing path of any one of the at least two target project users is used as the target processing path of the at least two target project users.

[0027] Optionally, after processing the multiple sub-project indicators of the target project user based on the target processing path, the method further includes:

[0028] Detecting the indicator growth value of the target project user in the multiple sub-project indicators within a preset time interval;

[0029] In the case where it is determined that the indicator growth value of the target sub-project indicator is lower than the first preset growth threshold, the processing strategy corresponding to the pre-generated target sub-project indicator and the user attribute of the target project user are input into the strategy recommendation model for processing to generate a target processing strategy corresponding to the target project user;

[0030] The target sub-project indicators of the target project user are processed based on the target processing strategy.

[0031] Optionally, the classifying the project users based on the project attributes of the initial project and the user attributes of the project users under the initial project to generate at least two classification results includes:

[0032] Discretize the project attributes of the initial project and the user attributes of the project users under the initial project to obtain multiple enumeration values;

[0033] Combining at least one enumeration value of the project attribute and at least one enumeration value of the user attribute to generate at least two combination results;

[0034] The project users are classified according to the enumeration values ​​respectively included in the at least two combination results, and a classification result corresponding to the at least two combination results is generated.

[0035] Optionally, the project processing method further includes:

[0036] Detecting a first growth value of the project indicator of the target project user within a preset time interval;

[0037] Acquire a reference user whose user attribute similarity with the target project user is greater than a preset similarity threshold, and determine a second growth value of the reference user in the project indicator within the preset time interval;

[0038] Determining whether the first growth value is greater than the second growth value;

[0039] If not, the indicator value prediction model is optimized by using the user attributes of the reference user and the multiple sub-item indicators.

[0040] According to a second aspect of an embodiment of this specification, there is provided a project processing device, including:

[0041] A classification module is configured to classify the project users into categories based on the project attributes of the initial project and the user attributes of the project users under the initial project, and generate at least two classification results;

[0042] a splitting module configured to split the project indicator of the initial project into a plurality of sub-project indicators, and determine the initial indicator value and the target indicator value of the target project user included in the target splitting result in the plurality of sub-project indicators, wherein the target splitting result is one of the at least two splitting results;

[0043] The processing module is configured to determine a target processing path of the target project user based on the initial indicator value and the target indicator value, and process the multiple sub-project indicators of the target project user based on the target processing path.

[0044] According to a third aspect of an embodiment of this specification, a computing device is provided, including:

[0045] Memory and processor;

[0046] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions:

[0047] Based on the project attributes of the initial project and the user attributes of the project users under the initial project, the project users are classified and at least two classification results are generated;

[0048] Splitting the project indicator of the initial project into multiple sub-project indicators, and determining the initial indicator value and the target indicator value of the target project user included in the target division result in the multiple sub-project indicators, wherein the target division result is one of the at least two division results;

[0049] A target processing path of the target project user is determined based on the initial indicator value and the target indicator value, and the multiple sub-project indicators of the target project user are processed based on the target processing path.

[0050] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the project processing method are implemented.

[0051] An embodiment of the present specification categorizes the project users based on the project attributes of the initial project and the user attributes of the project users under the initial project to generate at least two categorization results, splits the project indicators of the initial project into multiple sub-project indicators, and determines the initial indicator value and target indicator value of the target project users included in the target categorization result in the multiple sub-project indicators, wherein the target categorization result is one of the at least two categorization results, determines the target processing path of the target project user based on the initial indicator value and the target indicator value, and processes the multiple sub-project indicators of the target project user based on the target processing path.

[0052] The embodiments of this specification classify project users into categories, and based on the multiple sub-project indicators obtained by splitting the project indicators, determine the processing path between the initial indicator value and the target indicator value for project users in different categories, so as to perform long-term operational processing on project users based on the processing path, thereby achieving user reach according to the needs of different users and conducting long-term serialized operations, which is beneficial to improving the conversion rate of users in the initial project. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a processing flow chart of a project processing method provided by an embodiment of this specification;

[0054] Figure 2 It is a process flow chart of a project processing method provided by an embodiment of this specification;

[0055] Figure 3is a schematic diagram of a project processing device provided by an embodiment of this specification;

[0056] Figure 4 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0057] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0058] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0059] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0060] In this specification, a project processing method is provided. This specification also involves a project processing device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0061] In the field of lending, differentiated operations based on different groups of people are a commonly used operational means to increase the scale of operations. Classic group operations mainly divide the groups based on the different performances of target users in project indicators, and perform differentiated operations on the results of group division. However, this method is a short-chain operation centered on traffic, which often has problems such as short operation cycles, large amounts of time and manpower, and operational results that cannot meet demand. In addition, with the economic effects brought about by user traffic, it is imperative to build a serialized operational means based on long-term goals. However, how to achieve differentiated long-chain operations is a more complex issue. How to divide the population grid, build differentiated paths, and produce a white-box user advancement map to achieve the purpose of long-chain operations to increase the balance scale is an issue that needs to be solved urgently.

[0062] Based on this, the project processing method provided in the embodiments of this specification divides project users into categories and determines the processing path between the initial indicator value and the target indicator value for project users in different categories based on the multiple sub-project indicators obtained by splitting the project indicators, so as to perform long-term operational processing on project users based on the processing path, thereby achieving user reach according to the needs of different users and conducting long-term serialized operations to improve the user conversion rate of the initial project.

[0063] Figure 1 A processing flow chart of a project processing method provided according to an embodiment of the present specification is shown, including steps 102 to 106.

[0064] Step 102: based on the project attributes of the initial project and the user attributes of the project users under the initial project, the project users are classified and at least two classification results are generated.

[0065] Specifically, the initial projects, i.e., the projects to be processed, include but are not limited to loan projects, resource trading projects, resource leasing projects, etc.

[0066] In the embodiments of this specification, the initial project is taken as an example of a loan project. The specific implementation of the resource trading project and the resource leasing project is similar to the specific implementation of the loan project. The specific implementation of the loan project can be referred to and will not be repeated here.

[0067] The project attributes refer to the attribute information related to the initial project; for example, when the initial project is a loan project, its project attributes may be risk type, risk level, risk assessment index, etc.; the project users refer to the users who participate in the initial project, for example, users who participate in the loan project are the loan users of the loan project; the user attributes refer to the attribute information of the project users, including but not limited to the user's age, gender, occupation, education level, consumption habits and credit record, etc.

[0068] In an embodiment of the present specification, when it is determined that the project to be processed is the initial project, the project attributes of the initial project and the user attributes of the project users under the initial project can be obtained, and then the project users are classified based on the project attributes and user attributes to obtain corresponding category classification results, so as to determine the target processing path for the project users in each category classification result.

[0069] In a specific implementation, based on the project attributes of the initial project and the user attributes of the project users under the initial project, the project users are classified to generate at least two classification results, which can be specifically implemented in the following manner:

[0070] Discretize the project attributes of the initial project and the user attributes of the project users under the initial project to obtain multiple enumeration values;

[0071] Combining at least one enumeration value of the project attribute and at least one enumeration value of the user attribute to generate at least two combination results;

[0072] The project users are classified according to the enumeration values ​​respectively included in the at least two combination results, and a classification result corresponding to the at least two combination results is generated.

[0073] Specifically, if project attributes and user attributes are taken as two fixed types, namely two enumerations, and one enumeration corresponds to several defined values, namely enumeration values, then for project attributes, the enumeration values ​​include but are not limited to risk type, risk level, risk assessment indicators, etc.; for user attributes, the enumeration values ​​include but are not limited to age, gender, occupation, education level, consumption habits and credit record, etc.

[0074] After obtaining the project attributes of the initial project and the user attributes of the project users under the initial project, if the project attributes and / or user attributes are continuous data, they can be discretized to obtain multiple enumeration values ​​corresponding to the project attributes and / or user attributes; then one, two or more enumeration values ​​corresponding to the project attributes and one, two or more enumeration values ​​corresponding to the user attributes can be randomly combined or arranged to obtain two or more combination results; then the project users are classified according to the enumeration values ​​contained in each combination result, and one combination result is one category. The project users are classified according to the combination results to generate classification results corresponding to each combination result.

[0075] Taking the initial project as a loan project as an example, the enumeration value corresponding to the project attribute can be a risk assessment indicator, specifically a risk assessment indicator for overdue return of loan resources, and the enumeration value corresponding to the user attribute can be occupation, consumption habits and credit record; the above four enumeration values ​​are combined, and the combination results obtained can be combination 1 and combination 2, wherein combination 1 includes three enumeration values ​​of risk assessment indicator, occupation and consumption habits; combination 2 includes three enumeration values ​​of risk assessment indicator, occupation and credit record; the specific information corresponding to the three enumeration values ​​in combination 1 is 10%, freelance, and online purchase; the specific information corresponding to the three enumeration values ​​in combination 2 is 3%, office worker, and good credit; then the loan users can be classified according to the specific information corresponding to the enumeration values ​​contained in combination 1 and combination 2, and the loan users who meet the information corresponding to the three enumeration values ​​in combination 1 are classified into one category, and the loan users who meet the information corresponding to the three enumeration values ​​in combination 2 are classified into another category.

[0076] In addition, the lending users can also be categorized based on user needs. For example, the user needs in the lending project include but are not limited to enumeration values ​​such as consumption, transfer, and repayment; similarly, the enumeration values ​​corresponding to the project attributes, the enumeration values ​​corresponding to the user attributes, and the enumeration values ​​corresponding to the user needs can be aggregated to obtain two or more combination results; the lending users are then classified according to the enumeration values ​​contained in each combination result, and one combination result is one category. The project users are categorized according to the combination results, that is, all lending users are divided into groups based on user attributes + user needs + business attributes (such as: credit record + consumption + risk assessment indicators), and the division results corresponding to each combination result are generated.

[0077] Step 104: split the project indicator of the initial project into multiple sub-project indicators, and determine the initial indicator values ​​and target indicator values ​​of the target project users included in the target division result in the multiple sub-project indicators.

[0078] Specifically, the target division result is one of at least two division results, and the target project user is the project user included in the target division result.

[0079] After determining the initial project to be processed, you can continue to determine the project indicators of the initial project, and then you can decompose the project indicators of all users into n operational dimensions through high dimensions based on the project attributes of the initial project, that is, split the project indicators into multiple sub-project indicators, and decompose the goal of improving the indicator value of the project indicator into improving the indicator values ​​of multiple sub-project indicators. Then, based on the initial indicator values ​​and target indicator values ​​of the sub-project indicators of the target project users contained in each division result, plan the operation paths for different project users.

[0080] For example, the initial project may be a lending project, and the project indicator of the initial project may be the total amount of lending resources of the project user within a preset time interval; after determining the project indicator, a certain operation strategy needs to be adopted for the lending user so that the amount of lending resources of the lending user within the preset time interval is greater than or equal to the total amount of lending resources, and the operation strategy may split the project indicator according to the project attributes of the lending project, and split it into three sub-project indicators, namely, the number of loans, the amount of lending resources each time, and the number of days the lending resources are used; and then, based on these three sub-project indicators, the operation strategy (operation path) corresponding to the lending user is determined.

[0081] In specific implementation, determining the initial indicator value and target indicator value of the target project user included in the target division result in the multiple sub-project indicators can be specifically achieved in the following ways:

[0082] Determine the target project user included in the target division result, and input the user attribute of the target project user and the multiple sub-project indicators into the indicator value prediction model for processing to generate the target indicator value of the target project user in the multiple sub-project indicators; and

[0083] Determine the initial index value of the target project user included in the target division result in the plurality of sub-project indicators.

[0084] Specifically, after classifying the project users into categories and obtaining corresponding classification results, the project users included in each classification result can be determined, and then the user attributes of the project users in each classification result and multiple sub-project indicators are respectively input into the indicator value prediction model. The indicator value prediction model determines the target indicator values ​​of the project users in each classification result in multiple sub-project indicators based on the user attributes of each project user in the classification result, and the target indicator values ​​of the multiple sub-project indicators of the multiple project users in each classification result are the same.

[0085] In addition, after the users are classified and the corresponding classification results are obtained, the current index values ​​of the project users in each classification result in multiple sub-project indicators can be used as the initial index values.

[0086] After determining the aforementioned initial indicator values ​​and target indicator values, the operation paths for different project users can be planned based on the two.

[0087] Among them, determining the target indicator value is to construct a stable point (i.e., advanced target) for the division result, so that the fluctuation of the user's indicators in multiple sub-items approaches 0 over time. That is, when the user is in a certain high-dimensional state, he will continue to maintain the periodic behavior for a period of time in the future rather than a one-time behavior (for example, once the user develops a certain borrowing habit, he will continue to maintain the borrowing habit for a period of time in the future).

[0088] Step 106: determine a target processing path for the target project user based on the initial indicator value and the target indicator value, and process the multiple sub-project indicators of the target project user based on the target processing path.

[0089] In a specific implementation, determining the target processing path of the target project user based on the initial indicator value and the target indicator value can be implemented in the following ways:

[0090] Determining an initial key point corresponding to the target project user according to the initial indicator value, and determining a target key point of the target project user based on the target indicator value;

[0091] Determine the distance between the initial key point and the target key point according to the initial index value and the target index value;

[0092] Constructing a directed relationship network based on the initial key point, the target key point and the distance;

[0093] According to the target path algorithm, a target processing path between an initial key point corresponding to the target project user and the target key point in the directed relationship network is determined.

[0094] Further, according to the target path algorithm, the target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point is determined, which can be specifically implemented in the following manner:

[0095] The initial index values ​​of multiple sub-project indicators corresponding to the initial key points in the directed relationship network and the target index values ​​of multiple sub-project indicators corresponding to the target key points in the directed relationship network are input into the target path algorithm for calculation to generate a target processing path between each initial key point and the target key point in the directed relationship network.

[0096] Specifically, after determining the initial indicator values ​​and target indicator values ​​of the target project users in the target division results in multiple sub-project indicators through the aforementioned steps, the initial key points can be determined based on the initial indicator values ​​of each user in the target division results in multiple sub-project indicators, and the target key points can be determined based on the target indicator values ​​of the multiple sub-project indicators. Wherein, when the number of sub-project indicators is 2 or 3, the coordinate points corresponding to 2 or 3 sub-project indicators in the two-dimensional or three-dimensional coordinate system can be used as the initial key points. When the number of sub-project indicators is 2, the horizontal and vertical coordinates of the initial key points are the indicator values ​​corresponding to the 2 sub-project indicators. When the number of sub-project indicators is 3, the horizontal, vertical and vertical coordinates of the initial key points are the indicator values ​​corresponding to the 3 sub-project indicators. When the number of sub-project indicators is greater than 3, any point in the two-dimensional plane can be used as the initial key point corresponding to any target project user in the target division results, and then the initial key points of other target project users are determined based on the distance between the initial indicator values ​​of other target project users and the initial indicator value of the target project user.

[0097] The distance between the initial index values ​​of other target project users and the initial index value of the target project user can be determined by constructing a one-dimensional vector based on the index values ​​of each target project user in multiple sub-project indicators in the target partition result, and then calculating the Euclidean distance between each one-dimensional vector. Similarly, the distance between initial key points and initial key points in the target partition result, and the distance between initial key points and target key points can also be determined by constructing a one-dimensional vector and calculating the Euclidean distance between vectors.

[0098] After determining the distances between the initial key points and the initial key points and between the initial key points and the target key points in the target segmentation results, a directed relationship network can be constructed based on the initial key points, the target key points and the distances. The nodes in the directed relationship network are the initial key points or the target key points, and the edges between the nodes represent the distances between the two connected nodes.

[0099] After the directed relationship network is constructed, the target processing path between the initial key point and the target key point in the directed relationship network can be determined according to the target path algorithm. The target processing path is a path with a smaller distance among multiple paths between the initial key point and the target key point.

[0100] In practical applications, the target path algorithm can be the Freud algorithm. Therefore, the Freud algorithm can be used to process the initial indicator values ​​of multiple sub-project indicators corresponding to the initial key points in the directed relationship network, the target indicator values ​​of multiple sub-project indicators corresponding to the target key points in the directed relationship network, and the user attributes of the target project users to generate a target processing path between each initial key point in the directed relationship network and the target key point.

[0101] Alternatively, according to the target path algorithm, determining the target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point can also be achieved in the following manner:

[0102] Based on the target path algorithm and the directed relationship network, an initial adjacency matrix corresponding to the initial key point and the target key point is generated, wherein the elements in the initial adjacency matrix are used to represent the initial distances between different key points;

[0103] At least one of the initial key point and the target key point is used as an intermediate key point, and based on the intermediate key point, the elements in the initial adjacency matrix are updated to generate a target adjacency matrix;

[0104] According to the target adjacency matrix, a target distance between the initial key point and the target key point is determined, and according to the target distance, a target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point is determined.

[0105] Specifically, the target path algorithm can be the Floyd algorithm. When the Floyd algorithm is used to determine the target processing path, since many intermediate nodes are passed in the process from the initial key point to the target key point, these intermediate nodes are other initial key points. Therefore, when determining the target processing path from the initial key point to the target key point, a corresponding initial adjacency matrix can be generated based on a directed relationship network, wherein the elements in the initial adjacency matrix are used to characterize the initial distances between different key points; then, one other initial key point can be used as an intermediate node between an initial key point and the target key point, and the elements in the initial adjacency matrix are updated. After the update is completed, the number of intermediate nodes is increased by 1, and then the elements in the initial adjacency matrix are updated again until the number of intermediate nodes is equal to the number of initial key points in the target partitioning result minus 1, then the update is stopped to generate the target adjacency matrix.

[0106] After the target adjacency matrix is ​​generated according to the above process, the distance between the initial key point and the target key point can be determined according to the size of the element value in the target adjacency matrix, and then the target processing path between the two can be determined according to the distance.

[0107] In the embodiments of the present specification, the target processing path can be expressed as a sequence. Since the initial key points and the target key points are determined by the indicator values ​​of multiple sub-project indicators, the target processing path can be composed of multiple sub-project indicators. For example, the initial project is a loan project, and the target indicator values ​​of the target loan users in the target division results are: average monthly expenditure times 1 time, average monthly expenditure amount 3,000 yuan, average expenditure time 30 days, then the target processing path can be: operation loan times -> loan resource quota each time -> loan resource usage days. The meaning of this path is to perform operation processing on the project users in turn according to the operation strategies corresponding to the loan times, loan resource quota each time, and loan resource usage days, so that the user's loan data reaches the target indicator value.

[0108] In addition, determining the target processing path of the target project user based on the initial indicator value and the target indicator value may also be achieved in the following manner:

[0109] Determine, based on the initial indicator value and the target indicator value, an initial processing path for at least two target project users in the target division result;

[0110] Determining the similarity between the initial processing paths of the at least two target project users;

[0111] When the similarity is greater than or equal to a preset similarity threshold, the initial processing path of any one of the at least two target project users is used as the target processing path of the at least two target project users.

[0112] Specifically, after determining the initial indicator value and the target indicator value, the initial processing paths of at least two target project users in the target division results can be determined based on the initial indicator value and the target indicator value. The specific determination process is similar to the aforementioned processing process of determining the target processing path based on the initial indicator value and the target indicator value, and will not be repeated here.

[0113] After determining the initial processing path of the target project user, the similarity between the initial processing paths of at least two target project users in the target division result can be calculated, and when it is determined that the similarity is greater than or equal to a preset similarity threshold, the initial processing paths of the at least two target projects can be merged. Specifically, the initial processing path of any one of the at least two target project users can be used as the target processing path of the at least two target project users, or the initial processing paths of the at least two target project users can be fine-tuned to make their initial processing paths closer or more consistent.

[0114] Alternatively, the similarity between the initial processing paths of the target project users in different target division results may be calculated, and when it is determined that the similarity is greater than or equal to a preset similarity threshold, the initial processing paths of the target project users in each target division result may be adjusted so that their initial processing paths are differentiated.

[0115] By adjusting the population division in this way, the advancement stability points and advancement paths of different groups of people can be differentiated.

[0116] In a specific implementation, after processing the multiple sub-project indicators of the target project user based on the target processing path, the method further includes:

[0117] Detecting the indicator growth value of the target project user in the multiple sub-project indicators within a preset time interval;

[0118] In the case where it is determined that the indicator growth value of the target sub-project indicator is lower than the first preset growth threshold, the processing strategy corresponding to the pre-generated target sub-project indicator and the user attribute of the target project user are input into the strategy recommendation model for processing to generate a target processing strategy corresponding to the target project user;

[0119] The target sub-project indicators of the target project user are processed based on the target processing strategy.

[0120] Specifically, after processing the target project user using the target processing path, the indicator growth values ​​of the target project user in multiple sub-project indicators within a preset time interval can also be detected experimentally to determine whether the current operation processing strategy, that is, the current target processing path, can achieve the expected effect. When it is determined that the indicator growth value of a certain target sub-project indicator is lower than the first preset growth threshold, it is determined that the processing strategy of the target sub-project indicator does not meet the conditions. In this case, the pre-generated processing strategy corresponding to the target sub-project indicator and the user attributes of the target project user can be input into the strategy recommendation model for processing to generate a target processing strategy corresponding to the target project user; then the target sub-project indicator of the target project user can be re-processed based on the target processing strategy.

[0121] In addition, after processing multiple sub-project indicators of the target project user based on the target processing path, the first growth value of the target project user in the project indicator within a preset time interval can also be detected;

[0122] Acquire a reference user whose user attribute similarity with the target project user is greater than a preset similarity threshold, and determine a second growth value of the reference user in the project indicator within the preset time interval;

[0123] Determining whether the first growth value is greater than the second growth value;

[0124] If not, the indicator value prediction model is optimized by using the user attributes of the reference user and the multiple sub-item indicators.

[0125] Specifically, after the target project user is processed using the target processing path, a series of rationality verifications can be performed. Specifically, other reference users who are relatively similar to the target project user can be obtained, that is, reference users whose user attributes have a similarity greater than a preset similarity threshold with the target project user, and the second growth value of the reference user in the project indicator within a preset time interval is determined. Then, when it is determined that the first growth value is greater than the second growth value, it indicates that the target growth value of the user in the project indicator under the current target processing path does not meet expectations and needs to be optimized. That is, the indicator value prediction model needs to be optimized through the user attributes of the reference user and the multiple sub-project indicators to make the output result of the indicator value prediction model more accurate.

[0126] In addition, for multiple sub-project indicators, if the project users are operated according to the operation strategies corresponding to each sub-project indicator, some operation strategies corresponding to a certain sub-project indicator may be insufficient for the entire project user, so some supplements can be made to the operation strategies in this direction. And usually, a strategy recommendation model can be trained based on the consumption or borrowing habits of the project users and / or some preferences of the project users, so as to use the strategy recommendation model to recommend operation strategies.

[0127] An embodiment of the present specification categorizes the project users based on the project attributes of the initial project and the user attributes of the project users under the initial project to generate at least two categorization results, splits the project indicators of the initial project into multiple sub-project indicators, and determines the initial indicator value and target indicator value of the target project users included in the target categorization result in the multiple sub-project indicators, wherein the target categorization result is one of the at least two categorization results, determines the target processing path of the target project user based on the initial indicator value and the target indicator value, and processes the multiple sub-project indicators of the target project user based on the target processing path.

[0128] The embodiments of this specification classify project users into categories, and based on the multiple sub-project indicators obtained by splitting the project indicators, determine the processing path between the initial indicator value and the target indicator value for project users in different categories, so as to perform long-term operational processing on project users based on the processing path, thereby achieving user reach according to the needs of different users and conducting long-term serialized operations, which is beneficial to improving the conversion rate of users in the initial project.

[0129] The following combination Figure 2 , taking the application of the project processing method provided in this specification in a lending scenario as an example, the project processing method is further explained. Figure 2 A flowchart of a process of applying a project processing method provided by an embodiment of the present specification to a lending scenario is shown, and the specific steps include step 202 to step 222.

[0130] Step 202: based on the project attributes of the loan project and the user attributes of the project users under the loan project, the project users are classified and at least two classification results are generated.

[0131] Step 204: split the project indicator of the loan project into multiple sub-project indicators.

[0132] Step 206, determining the target lending user included in the target segmentation result, and inputting the user attributes of the target lending user and the multiple sub-item indicators into the indicator value prediction model for processing to generate the target indicator value of the target lending user in the multiple sub-item indicators.

[0133] The target segmentation result is one of the at least two segmentation results.

[0134] Step 208: determining the initial index values ​​of the target lending users included in the target division result in the multiple sub-item indexes.

[0135] Step 210: determining the initial key point corresponding to the target lending user according to the initial indicator value, and determining the target key point of the target lending user based on the target indicator value.

[0136] Step 212: Determine the distance between the initial key point and the target key point according to the initial index value and the target index value.

[0137] Step 214: construct a directed relationship network based on the initial key points, the target key points and the distances.

[0138] Step 216, determining an initial processing path between an initial key point corresponding to the target lending user in the directed relationship network and the target key point according to the Floyd algorithm.

[0139] Step 218: Determine the similarity between the initial processing paths of the at least two target lending users.

[0140] Step 220: When the similarity is greater than or equal to a preset similarity threshold, the initial processing path of any one of the at least two target lending users is used as the target processing path of the at least two target lending users.

[0141] Step 222: Process the multiple sub-item indicators of the target lending user based on the target processing path.

[0142] The embodiments of this specification classify the borrowing users under the lending project into categories, and determine the operation path between the initial total borrowing amount and the target total borrowing amount for the borrowing users in different categories based on the multiple sub-project indicators obtained by splitting the total borrowing amount indicator, so as to perform long-term operation processing on the borrowing users based on the operation path, thereby achieving user contact and long-term serial operation according to the needs of different users. This method is conducive to improving the stickiness between users and the lending project, and is conducive to improving the conversion rate of users' lending resources under the lending project.

[0143] Corresponding to the above method embodiment, this specification also provides an item processing device embodiment, Figure 3 FIG. 1 is a schematic diagram of a project processing device provided by an embodiment of the present specification. Figure 3 As shown, the device comprises:

[0144] The classification module 302 is configured to classify the project users into categories based on the project attributes of the initial project and the user attributes of the project users under the initial project, and generate at least two classification results;

[0145] A splitting module 304 is configured to split the project indicator of the initial project into multiple sub-project indicators, and determine the initial indicator value and the target indicator value of the target project user included in the target division result in the multiple sub-project indicators, wherein the target division result is one of the at least two division results;

[0146] The processing module 306 is configured to determine a target processing path of the target project user based on the initial indicator value and the target indicator value, and process the multiple sub-project indicators of the target project user based on the target processing path.

[0147] Optionally, the splitting module 304 is further configured to:

[0148] Determine the target project user included in the target division result, and input the user attribute of the target project user and the multiple sub-project indicators into the indicator value prediction model for processing to generate the target indicator value of the target project user in the multiple sub-project indicators; and

[0149] Determine the initial index value of the target project user included in the target division result in the plurality of sub-project indicators.

[0150] Optionally, the processing module 306 includes:

[0151] A first determination submodule is configured to determine an initial key point corresponding to the target project user according to the initial indicator value, and determine a target key point of the target project user based on the target indicator value;

[0152] A second determination submodule is configured to determine the distance between the initial key point and the target key point according to the initial index value and the target index value;

[0153] A construction submodule, configured to construct a directed relationship network based on the initial key point, the target key point and the distance;

[0154] The third determination submodule is configured to determine, according to a target path algorithm, a target processing path between an initial key point corresponding to the target project user in the directed relationship network and the target key point.

[0155] Optionally, the third determining submodule is further configured to:

[0156] The initial index values ​​of multiple sub-project indicators corresponding to the initial key points in the directed relationship network and the target index values ​​of multiple sub-project indicators corresponding to the target key points in the directed relationship network are input into the target path algorithm for calculation to generate a target processing path between each initial key point and the target key point in the directed relationship network.

[0157] Optionally, the third determining submodule is further configured to:

[0158] Based on the target path algorithm and the directed relationship network, an initial adjacency matrix corresponding to the initial key point and the target key point is generated, wherein the elements in the initial adjacency matrix are used to represent the initial distances between different key points;

[0159] At least one of the initial key point and the target key point is used as an intermediate key point, and based on the intermediate key point, the elements in the initial adjacency matrix are updated to generate a target adjacency matrix;

[0160] According to the target adjacency matrix, a target distance between the initial key point and the target key point is determined, and according to the target distance, a target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point is determined.

[0161] Optionally, the processing module 306 is further configured to:

[0162] Determine, based on the initial indicator value and the target indicator value, an initial processing path for at least two target project users in the target division result;

[0163] Determining the similarity between the initial processing paths of the at least two target project users;

[0164] When the similarity is greater than or equal to a preset similarity threshold, the initial processing path of any one of the at least two target project users is used as the target processing path of the at least two target project users.

[0165] Optionally, the item processing device further includes a first detection module configured to:

[0166] Detecting the indicator growth value of the target project user in the multiple sub-project indicators within a preset time interval;

[0167] In the case where it is determined that the indicator growth value of the target sub-project indicator is lower than the first preset growth threshold, the processing strategy corresponding to the pre-generated target sub-project indicator and the user attribute of the target project user are input into the strategy recommendation model for processing to generate a target processing strategy corresponding to the target project user;

[0168] The target sub-project indicators of the target project user are processed based on the target processing strategy.

[0169] Optionally, the division module 302 is further configured to:

[0170] Discretize the project attributes of the initial project and the user attributes of the project users under the initial project to obtain multiple enumeration values;

[0171] Combining at least one enumeration value of the project attribute and at least one enumeration value of the user attribute to generate at least two combination results;

[0172] The project users are classified according to the enumeration values ​​respectively included in the at least two combination results, and a classification result corresponding to the at least two combination results is generated.

[0173] Optionally, the item processing device further includes a second detection module configured to:

[0174] Detecting a first growth value of the project indicator of the target project user within a preset time interval;

[0175] Acquire a reference user whose user attribute similarity with the target project user is greater than a preset similarity threshold, and determine a second growth value of the reference user in the project indicator within the preset time interval;

[0176] Determining whether the first growth value is greater than the second growth value;

[0177] If not, the indicator value prediction model is optimized by using the user attributes of the reference user and the multiple sub-item indicators.

[0178] The above is a schematic scheme of a project processing device of this embodiment. It should be noted that the technical scheme of the project processing device and the technical scheme of the project processing method described above are of the same concept, and the details not described in detail in the technical scheme of the project processing device can be referred to the description of the technical scheme of the project processing method described above.

[0179] Figure 4 The block diagram of a computing device 400 according to one embodiment of the present specification is shown. The components of the computing device 400 include but are not limited to a memory 410 and a processor 420. The processor 420 is connected to the memory 410 via a bus 430, and the database 450 is used to store data.

[0180] The computing device 400 also includes an access device 440 that enables the computing device 400 to communicate via one or more networks 460. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 440 may include one or more of any type of network interface (e.g., a network interface card (NIC)) wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0181] In one embodiment of the present specification, the above components of the computing device 400 and Figure 4 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 4 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0182] Computing device 400 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. Computing device 400 may also be a mobile or stationary server.

[0183] The memory 410 is used to store computer executable instructions, and the processor 420 is used to execute the following computer executable instructions:

[0184] Based on the project attributes of the initial project and the user attributes of the project users under the initial project, the project users are classified and at least two classification results are generated;

[0185] Splitting the project indicator of the initial project into multiple sub-project indicators, and determining the initial indicator value and the target indicator value of the target project user included in the target division result in the multiple sub-project indicators, wherein the target division result is one of the at least two division results;

[0186] A target processing path of the target project user is determined based on the initial indicator value and the target indicator value, and the multiple sub-project indicators of the target project user are processed based on the target processing path.

[0187] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned project processing method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned project processing method.

[0188] An embodiment of the present specification also provides a computer-readable storage medium storing computer instructions, which are used to implement the steps of the project processing method when executed by a processor.

[0189] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned project processing method belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned project processing method.

[0190] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0191] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0192] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0193] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0194] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A project processing method, comprising: Discretize the project attributes of the initial project and the user attributes of the project users under the initial project to obtain multiple enumeration values; Combining at least one enumeration value of the project attribute and at least one enumeration value of the user attribute to generate at least two combination results; Classify the project users according to the enumeration values ​​respectively included in the at least two combination results, and generate classification results corresponding to the at least two combination results; Splitting the project indicator of the initial project into multiple sub-project indicators, and determining the initial indicator value and the target indicator value of the target project user included in the target division result in the multiple sub-project indicators, wherein the target division result is one of the at least two division results; Determining an initial key point corresponding to the target project user according to the initial indicator value, and determining a target key point of the target project user based on the target indicator value; Determine the distance between the initial key point and the target key point according to the initial index value and the target index value; Constructing a directed relationship network based on the initial key point, the target key point and the distance; According to the target path algorithm, a target processing path between an initial key point corresponding to the target project user and the target key point in the directed relationship network is determined, and the multiple sub-project indicators of the target project user are processed based on the target processing path.

2. The project processing method according to claim 1, wherein the determining of the initial index value and the target index value of the target project user in the plurality of sub-project indexes included in the target division result comprises: Determine the target project user included in the target division result, and input the user attribute of the target project user and the multiple sub-project indicators into the indicator value prediction model for processing to generate the target indicator value of the target project user in the multiple sub-project indicators; and Determine the initial index value of the target project user included in the target division result in the plurality of sub-project indicators.

3. The project processing method according to claim 1, wherein determining the target processing path between the initial key point corresponding to the target project user and the target key point in the directed relationship network according to the target path algorithm comprises: The initial index values ​​of multiple sub-project indicators corresponding to the initial key points in the directed relationship network and the target index values ​​of multiple sub-project indicators corresponding to the target key points in the directed relationship network are input into the target path algorithm for calculation to generate a target processing path between each initial key point and the target key point in the directed relationship network.

4. The project processing method according to claim 1, wherein determining the target processing path between the initial key point corresponding to the target project user and the target key point in the directed relationship network according to the target path algorithm comprises: Based on the target path algorithm and the directed relationship network, an initial adjacency matrix corresponding to the initial key point and the target key point is generated, wherein the elements in the initial adjacency matrix are used to represent the initial distances between different key points; At least one of the initial key point and the target key point is used as an intermediate key point, and based on the intermediate key point, the elements in the initial adjacency matrix are updated to generate a target adjacency matrix; According to the target adjacency matrix, a target distance between the initial key point and the target key point is determined, and according to the target distance, a target processing path between the initial key point corresponding to the target project user in the directed relationship network and the target key point is determined.

5. The project processing method according to claim 1, wherein the step of determining the target processing path of the target project user based on the initial indicator value and the target indicator value comprises: Determine, based on the initial indicator value and the target indicator value, an initial processing path for at least two target project users in the target division result; Determining the similarity between the initial processing paths of the at least two target project users; When the similarity is greater than or equal to a preset similarity threshold, the initial processing path of any one of the at least two target project users is used as the target processing path of the at least two target project users.

6. The project processing method according to claim 1, after processing the multiple sub-project indicators of the target project user based on the target processing path, further comprising: Detecting the indicator growth value of the target project user in the multiple sub-project indicators within a preset time interval; In the case where it is determined that the indicator growth value of the target sub-project indicator is lower than the first preset growth threshold, the processing strategy corresponding to the pre-generated target sub-project indicator and the user attribute of the target project user are input into the strategy recommendation model for processing to generate a target processing strategy corresponding to the target project user; The target sub-project indicators of the target project user are processed based on the target processing strategy.

7. The project processing method according to claim 2, further comprising: Detecting a first growth value of the project indicator of the target project user within a preset time interval; Acquire a reference user whose user attribute similarity with the target project user is greater than a preset similarity threshold, and determine a second growth value of the reference user in the project indicator within the preset time interval; Determining whether the first growth value is greater than the second growth value; If not, the indicator value prediction model is optimized by using the user attributes of the reference user and the multiple sub-item indicators.

8. A project processing device, comprising: A partitioning module is configured to discretize the project attributes of the initial project and the user attributes of the project users under the initial project to obtain multiple enumeration values; Combining at least one enumeration value of the project attribute and at least one enumeration value of the user attribute to generate at least two combination results; classifying the project users according to the enumeration values ​​respectively included in the at least two combination results, and generating classification results corresponding to the at least two combination results; a splitting module configured to split the project indicator of the initial project into a plurality of sub-project indicators, and determine the initial indicator value and the target indicator value of the target project user included in the target splitting result in the plurality of sub-project indicators, wherein the target splitting result is one of the at least two splitting results; The processing module is configured to determine the initial key point corresponding to the target project user according to the initial indicator value, and determine the target key point of the target project user based on the target indicator value; determine the distance between the initial key point and the target key point according to the initial indicator value and the target indicator value; construct a directed relationship network based on the initial key point, the target key point and the distance; determine the target processing path between the initial key point corresponding to the target project user and the target key point in the directed relationship network according to the target path algorithm, and process the multiple sub-project indicators of the target project user based on the target processing path.

9. A computing device comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the steps of the project processing method described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the steps of the project processing method described in any one of claims 1 to 7.

Images