CN116993396B - Risk early warning method based on vehicle user tag and computer equipment - Google Patents
Risk early warning method based on vehicle user tag and computer equipment Download PDFInfo
- Publication number
- CN116993396B CN116993396B CN202311252821.XA CN202311252821A CN116993396B CN 116993396 B CN116993396 B CN 116993396B CN 202311252821 A CN202311252821 A CN 202311252821A CN 116993396 B CN116993396 B CN 116993396B
- Authority
- CN
- China
- Prior art keywords
- user
- information
- sample
- value
- user value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000004044 response Effects 0.000 claims abstract description 16
- 238000001514 detection method Methods 0.000 claims description 105
- 238000012795 verification Methods 0.000 claims description 36
- 238000005192 partition Methods 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000012546 transfer Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000006835 compression Effects 0.000 claims description 7
- 238000007906 compression Methods 0.000 claims description 7
- 238000012512 characterization method Methods 0.000 claims 1
- 238000013500 data storage Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000000638 solvent extraction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The embodiment of the disclosure discloses a risk early warning method and computer equipment based on a vehicle user tag. One embodiment of the method comprises the following steps: classifying the user vehicle circulation information set to obtain a user vehicle circulation information set; inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence; user vehicle circulation information meeting initial early warning conditions is selected from the user vehicle circulation information set to serve as user vehicle circulation information to be detected; inputting the user vehicle circulation information group to be detected into a user value information prediction model to obtain a user value information group; and generating user early warning information in response to determining that the user value information meets the early warning condition, and sending the user early warning information to the associated user early warning terminal for early warning. The embodiment improves the business transaction value of the user and avoids the loss of the user.
Description
Technical Field
The embodiment of the disclosure relates to the field of vehicle information early warning, in particular to a risk early warning method based on a vehicle user tag and computer equipment.
Background
With the development of the automotive industry, automobile users are also becoming more and more. Currently, when selecting a car user or pushing car information to a user, the following methods are generally adopted: the information of the user is detected only by the staff, so that different vehicle information is pushed to different users. However, by detecting the information of the user by the staff, the risk information of the user is difficult to detect, and the vehicle information is pushed to the risk user, so that the traffic pushing resources are wasted; in addition, the information of the user is detected by the staff, the detection accuracy is low, and the detection time is long; when the value information of the user is changed, abnormal user value information is difficult to detect in time, and traffic pushing resources are further wasted.
Disclosure of Invention
The disclosure is in part intended to introduce concepts in a simplified form that are further described below in the detailed description. The disclosure is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
Some embodiments of the present disclosure propose a risk early warning method, a computer device and a computer-readable storage medium based on a vehicle user tag to solve the technical problems mentioned in the background section above.
In a first aspect, some embodiments of the present disclosure provide a risk early warning method based on a vehicle user tag, the method comprising: acquiring a user vehicle circulation information set in a preset historical time period, wherein the user vehicle circulation information in the user vehicle circulation information set comprises: vehicle type, vehicle value information, vehicle circulation type and value circulation node information; classifying the user vehicle circulation information set according to the vehicle type and the vehicle circulation type included in the user vehicle circulation information set to obtain a user vehicle circulation information set; inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence; selecting user vehicle circulation information meeting initial early warning conditions from the user vehicle circulation information set as user vehicle circulation information to be detected to obtain a user vehicle circulation information set to be detected, wherein the initial early warning conditions are as follows: the vehicle circulation type included in the user vehicle circulation information is a target vehicle circulation type; inputting the user vehicle circulation information group to be detected into a pre-trained user value information prediction model to obtain a user value information group, wherein one user vehicle circulation information to be detected corresponds to one user value information; for each piece of user value information in the user value information, generating user early warning information in response to determining that the user value information meets the early warning condition, and sending the user early warning information to an associated user early warning terminal for early warning.
In a second aspect, the present disclosure also provides a computer device comprising a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, implements a method as described in any of the implementations of the first aspect.
In a third aspect, the present disclosure also provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method as described in any of the implementations of the first aspect.
The above embodiments of the present disclosure have the following advantageous effects: by the risk early warning method based on the vehicle user tag, which is disclosed by the embodiment of the invention, the risk user can be detected, so that the waste of traffic pushing resources can be reduced. First, a user vehicle circulation information set in a preset history period is acquired. Wherein, the user vehicle circulation information in the user vehicle circulation information set includes: vehicle type, vehicle value information, vehicle circulation type, and value circulation node information. Thus, data support is provided for determining recommended vehicle information. And secondly, classifying the user vehicle circulation information set according to the vehicle type and the vehicle circulation type included in the user vehicle circulation information set to obtain a user vehicle circulation information set. Thus, a reasonable recommended vehicle information sequence can be deduced according to different types of users. And then, inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence. Therefore, the recommended vehicle information sequence is conveniently predicted according to the vehicle purchasing habits of all users. Therefore, the accuracy of the recommended vehicle information is ensured. And then, selecting the user vehicle circulation information meeting the initial early warning condition from the user vehicle circulation information set as user vehicle circulation information to be detected, and obtaining a user vehicle circulation information set to be detected. Wherein, the initial early warning condition is: the user vehicle circulation information comprises a vehicle circulation type which is a target vehicle circulation type. And then, inputting the user vehicle circulation information group to be detected into a pre-trained user value information prediction model to obtain a user value information group. Wherein, the vehicle circulation information of a user to be detected corresponds to user value information. Thus, the user value of each user can be detected. And finally, for each piece of user value information in the user value information, generating user early warning information in response to determining that the user value information meets the early warning condition, and sending the user early warning information to an associated user early warning terminal for early warning. Therefore, the risk user (user early warning information) is detected, and the vehicle information can be prevented from being pushed to the risk user, so that the waste of traffic pushing resources is reduced. The business transaction value of the user is improved, and the user loss is avoided.
Drawings
The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale.
FIG. 1 is a flow chart of some embodiments of a vehicle user tag-based risk early warning method according to the present disclosure;
fig. 2 is a schematic block diagram of a computer device provided in an embodiment of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure.
It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings. Embodiments of the present disclosure and features of embodiments may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.
It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.
The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.
The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 is a flow chart of some embodiments of a vehicle user tag-based risk early warning method according to the present disclosure. A flow 100 of some embodiments of a vehicle user tag-based risk early warning method according to the present disclosure is shown. The risk early warning method based on the vehicle user tag comprises the following steps:
step 101, acquiring a user vehicle circulation information set in a preset historical time period.
In some embodiments, a subject (e.g., computing device) performing risk early warning based on a vehicle user tag may obtain a set of user vehicle circulation information over a preset historical period from a preset vehicle information database. Wherein, the user vehicle circulation information in the user vehicle circulation information set includes: vehicle type, vehicle value information, vehicle circulation type, and value circulation node information. The vehicle type may represent a specific type model of the vehicle. The vehicle value information may represent the value of the vehicle (vehicle unit price). The vehicle circulation type may represent a type of vehicle value transfer, and may include a full value transfer type (full-money purchase) and a segment value transfer type (segment purchase). The value transfer node information may indicate whether the value of the vehicle is transferred to completion, and whether there is a staged value transfer to completion (advance payoff) or deferred (deferred payoff). For example, the value transfer node information may be full value transfer completion. The value circulation node information can also be that the segmentation value transfer is completed, and the segmentation value is transferred in advance. The value flow node information can also be that the segment value transfer is incomplete and the segment value is delayed. The historical time period may be 2022, 1-2022, 6-30.
Step 102, classifying the user vehicle circulation information set according to the vehicle type and the vehicle circulation type included in the user vehicle circulation information set to obtain a user vehicle circulation information set.
In some embodiments, the executing body may classify the user vehicle circulation information set according to a vehicle type and a vehicle circulation type included in the user vehicle circulation information set, so as to obtain a user vehicle circulation information set. The user vehicle circulation information set can be obtained by gathering the user vehicle circulation information of the same vehicle type and the vehicle circulation type into one type.
And step 103, inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence.
In some embodiments, the executing entity may input the set of user vehicle circulation information sets into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence. The user vehicle recommendation model may be a pre-trained neural network model with the user vehicle circulation information set as input and the recommended vehicle information sequence as output. For example, the user vehicle recommendation model may be a convolutional neural network model. That is, each recommended vehicle information corresponds to one user vehicle circulation information group. The recommended vehicle information may contain basic information of the vehicle. For example, the recommended vehicle information may include a vehicle name, vehicle value information, a vehicle type.
Step 104, selecting the user vehicle circulation information meeting the initial early warning condition from the user vehicle circulation information set as the user vehicle circulation information to be detected, and obtaining a user vehicle circulation information set to be detected.
In some embodiments, the executing body may select, from the set of user vehicle circulation information, user vehicle circulation information that meets an initial early warning condition as user vehicle circulation information to be detected, to obtain a user vehicle circulation information set to be detected. Wherein, the initial early warning condition is: the user vehicle circulation information comprises a vehicle circulation type which is a target vehicle circulation type.
Further, an initial user value information prediction model file is obtained. Wherein the initial user value information prediction model file includes tree structure parameters of nested relationships, and the initial user value information prediction model in the initial user value information prediction model file is a pre-trained model (for example, a convolutional neural network model) for generating user value information. The tree structure may be a binary tree structure. The nested relationship may be a nested relationship in a binary tree. The user value information may be information representing the value of the user (e.g., the value of the user purchasing the vehicle again).
Further, format conversion processing is carried out on the initial user value information prediction model file, and a converted initial user value information prediction model file is generated. The format of the converted initial user value information predictive model file may be xml (Extensible Markup Language ) format. The initial user value information predictive model file may be formatted using a preset transformation tool DumpViewer dump viewer.
And further, carrying out dimension compression processing on the conversion initial user value information prediction model file to obtain a compression initial user value information prediction model file. That is, the model parameters in the conversion-initial user value information prediction model file may be subjected to dimensional compression (dimension reduction).
Further, a user verification data set is obtained. The user verification data may include vehicle browsing information of the user and information of purchasing the vehicle. The user verification data set may be used to verify whether the accuracy of the initial user value information prediction model file is the same as the accuracy of the compressed initial user value information prediction model file.
Further, the user verification data set is input into the initial user value information prediction model file, and a first verification value corresponding to the user verification data set is obtained.
Further, the user verification data set is input into the compressed initial user value information prediction model file, and a second verification value corresponding to the user verification data set is obtained.
Further, a check result is generated by the first check value and the second check value. Firstly, respectively carrying out downward rounding processing on the first check value and the second check value to obtain a rounded first check value and a rounded second check value. And then, when the rounding first check value and the rounding second check value are the same, generating a check result representing that the check passes. And when the rounding first check value and the rounding second check value are different, generating a check result representing that the check fails.
Further, in response to determining that the verification result characterizes verification, determining a model in the compressed initial user value information prediction model file as a user value information prediction model.
For the background technology, the detection of the information of the user by the staff is low in detection accuracy and long in detection time. ". The method can be solved by the following steps: first, an initial user value information prediction model file is acquired. The initial user value information prediction model file comprises tree structure parameters of nested relations, and an initial user value information prediction model in the initial user value information prediction model file is a pre-trained model for generating user value information. And secondly, carrying out format conversion processing on the initial user value information prediction model file to generate a conversion initial user value information prediction model file, and then carrying out dimension compression processing on the conversion initial user value information prediction model file to obtain a compression initial user value information prediction model file. Thereby, the dimension of the compressed model can be reduced. Thus, the calculation speed of the model is improved. Then, acquiring a user verification data set; inputting the user verification data set into the initial user value information prediction model file to obtain a first verification value corresponding to the user verification data set; inputting the user verification data set into the compressed initial user value information prediction model file to obtain a second verification value corresponding to the user verification data set; and generating a verification result through the first verification value and the second verification value. And finally, determining the model in the compressed initial user value information prediction model file as a user value information prediction model in response to determining that the verification result represents that verification passes. Thereby, the dimension of the compressed model can be reduced. Thus, the calculation speed of the model is improved. And the model is verified by the user verification data set, so that the accuracy of user information detection is improved, and the detection time is shortened.
And 105, inputting the user vehicle circulation information group to be detected into a pre-trained user value information prediction model to obtain a user value information group.
In some embodiments, the executing body may input the to-be-detected user vehicle circulation information set into a pre-trained user value information prediction model to obtain a user value information set. Wherein, the vehicle circulation information of a user to be detected corresponds to user value information. And inputting the vehicle circulation information of each user to be detected into the user value information prediction model to generate user value information, so as to obtain a user value information group. That is, the user value information prediction model may be a convolutional neural network model trained in advance, with user vehicle flow information to be detected as input, and user value information as output. The user value information may represent a value of continuing to serve the user. The higher the value, the lower the risk of the user, and the stronger the purchasing ability. The lower the value, the higher the risk of the user and the worse the purchasing ability.
And 106, for each piece of user value information in the user value information, generating user early warning information in response to determining that the user value information meets the early warning condition, and sending the user early warning information to the associated user early warning terminal for early warning.
In some embodiments, the executing entity may generate, for each piece of user value information, user early warning information in response to determining that the user value information satisfies an early warning condition, and send the user early warning information to an associated user early warning terminal for early warning. The early warning condition may be: the value degree of the user value information representation is smaller than or equal to the preset value degree. When the user value information meets the early warning condition, the execution subject can automatically generate the user early warning information. The user early warning terminal can be a computer terminal of a vehicle customer service person. The user early warning information may be information for early warning of a risk user. For example, the user early warning information may be that the risk of the user a is high, please carefully push information to the user a or sell the car.
Further, for each user vehicle circulation information in the user vehicle circulation information set, the following processing steps are executed:
first, determining whether value flow node information included in the user vehicle flow information meets value early warning conditions. The value early warning condition may be that the value circulation node information represents a segment value delay.
And secondly, responding to the fact that the value circulation node information meets the value early warning condition, and carrying out value early warning on the user corresponding to the user vehicle circulation information. That is, the user account corresponding to the user vehicle circulation information may be marked as a risk account, so as to remind the staff.
Further, for each piece of user value information in the user value information, in response to determining that the user value information does not meet the pre-warning condition, vehicle circulation policy information corresponding to the user value information is selected from a preset vehicle circulation policy information group, and the vehicle circulation policy information is sent to a corresponding customer service terminal. Wherein, the vehicle circulation policy information in the vehicle circulation policy information group includes: vehicle circulation policy information corresponding to a first vehicle circulation type and vehicle circulation policy information corresponding to a second vehicle circulation type. Here, the vehicle circulation policy information group may be policy information (may include a manner of pushing vehicle information and a manner of introducing vehicles by customer service personnel) that is preset to push information to a user and provide services. Each vehicle circulation strategy information sets a corresponding value degree interval. That is, different vehicle circulation policy information is selected for users of different user value information. The first vehicle flow type may represent a full value transfer type (full-money order). The second vehicle flow type may represent a segment value transfer type (segment purchase).
Further, for each piece of the user value information, the recommended vehicle information sequence is transmitted to the user terminal corresponding to the user value information in response to determining that the user value information does not satisfy the pre-warning condition. The user terminal corresponding to the user value information may be a mobile phone terminal of the user corresponding to the user value information.
Further, user value fluctuation information corresponding to each user vehicle circulation information in the user vehicle circulation information set is obtained, and a user value fluctuation information set is obtained.
In some embodiments, the executing body may acquire user value fluctuation information corresponding to each user vehicle circulation information in the user vehicle circulation information set, to obtain a user value fluctuation information set. That is, the user value fluctuation information may be obtained from each user terminal by means of wired connection or wireless connection, to obtain a user value fluctuation information set. The user value fluctuation information may represent asset fluctuation information of the user.
Optionally, a user value fluctuation sample set is obtained.
In some embodiments, the executive may obtain a sample set of user value fluctuations. Wherein the user value fluctuation sample set is a sample set stored in the form of a data block. The user value fluctuation samples in the user value fluctuation sample set may include asset transition information for the user.
Optionally, obtaining a historical sample division result corresponding to each data block in the historical user value fluctuation sample set, and obtaining a historical sample division result set.
In some embodiments, the executing body may obtain a historical sample division result corresponding to each data block in the historical user value fluctuation sample set, so as to obtain a historical sample division result set. The historical sample division results may be historical sample division results corresponding to respective data storage spaces in the data block. The historical sample partitioning result may be a result of partitioning the historical user value fluctuation sample set according to the data storage location. And storing the divided target user value fluctuation sample group corresponding to the historical sample division result in a corresponding data storage position. The historical user value fluctuation sample set may be a historical acquired user value fluctuation sample set.
Optionally, determining a current sample division result of the data block corresponding to the user value fluctuation sample set.
In some embodiments, the executing entity may determine a current sample division result of the data block corresponding to the user value fluctuation sample set. The current sample dividing result can represent sample storage conditions of the corresponding data blocks of the user value fluctuation sample set corresponding to the data storage positions.
In an actual application scenario, the execution body may determine a current sample division result of the data block corresponding to the user value fluctuation sample set by:
and a first step of determining a sample attribute set corresponding to the user value fluctuation sample set. The sample attribute may be an attribute of a user value fluctuation sample, among others. For example, sample attributes include: user value fluctuation time (user asset change time), user value fluctuation value (asset increment and decrement value), and the like.
And a second step of determining first attribute division information corresponding to each sample attribute in the sample attribute set based on the user value fluctuation sample set. The first attribute classification information may be determination information of an attribute for performing sample classification on the user value fluctuation sample set. The judgment information may be a judgment parameter of the partitionable attribute. For example, the judgment information may be the difference between the maximum value and the minimum value of the sample attribute in the user value fluctuation sample set. First, judgment information corresponding to each sample attribute in the sample attribute set may be determined. For example, the judgment information may be the difference between the maximum value and the minimum value of the corresponding sample attribute. And then, determining judgment information corresponding to each sample attribute according to the user value fluctuation sample set. That is, the judgment information may be a difference between a maximum value and a minimum value of the attribute in the above-described user value fluctuation sample set.
And thirdly, determining the sample attribute of which the first attribute partition information corresponding to the sample attribute set does not meet the partition attribute condition as a first sample attribute, and obtaining a first sample attribute group. Wherein each first sample attribute in the first sample attribute set may be a partitionable attribute for a user value fluctuation sample set. The user value fluctuation sample set may be partitioned according to partitionable attributes. The partitioning attribute condition may be that a value corresponding to the first attribute partitioning information is smaller than a preset attribute threshold corresponding to the first attribute partitioning attribute.
Fourth, selecting a first target sample attribute from the first sample attribute group. One first sample attribute may be randomly selected from the first sample attribute group as the first target sample attribute.
Fifthly, dividing the user value fluctuation sample set based on the first target sample attribute to obtain a user value fluctuation sample set. The median of the first target sample attribute in the user value fluctuation sample set may be determined. And then, uniformly dividing the user value fluctuation sample set through the median to obtain a user value fluctuation sample set.
Sixth, for each user value fluctuation sample group in the user value fluctuation sample group set, performing the steps of:
first, according to the user value fluctuation sample group, second attribute division information corresponding to each sample attribute in the sample attribute set is determined. The generation of the second attribute division information may refer to a manner of generating the first attribute division information.
And secondly, selecting the sample attribute of which the corresponding second attribute division information does not accord with the division attribute condition from the sample attribute set as a second sample attribute, and obtaining a second sample attribute group. The generation of the second sample property group may refer to the manner in which the first sample property group is generated.
Next, a second target sample attribute is selected from the second set of sample attributes. A second set of sample attributes may be randomly selected from the second set of sample attributes as the second target sample attribute.
And then, dividing the user value fluctuation sample group according to the second target sample attribute to obtain an alternative user value fluctuation sample group set. Reference may be made to the manner in which the set of user value fluctuation samples is partitioned.
Then, for each of the set of candidate user value fluctuation sample groups, third attribute classification information corresponding to the candidate user value fluctuation sample group in the sample attribute set is determined. The generation manner of the third attribute division information may be referred to the generation manner of the first attribute division information.
And finally, determining a sample division result corresponding to the candidate user value fluctuation sample group set as a sample division result corresponding to the user value fluctuation sample group in response to the fact that the obtained third attribute division information meets the corresponding division attribute condition and the number of samples included in the candidate user value fluctuation sample group set is smaller than a preset value.
And seventh, determining each sample division result as the current sample division result of the data block corresponding to the user value fluctuation sample set.
Optionally, determining partition difference information between each historical sample partition result in the historical sample partition result set and the current sample partition result to obtain a partition difference information set.
In some embodiments, the executing body may determine partition difference information between each of the historical sample partition result sets and the current sample partition result to obtain a partition difference information set. The partition difference information may characterize sample difference information between a sample set corresponding to the historical sample partition result and a sample set corresponding to the current sample partition result. In practice, the partition difference information may include: number of sample differences, sample difference content.
Alternatively, the target difference information is generated according to the above-described divided difference information group.
In some embodiments, the executing entity may generate the target difference information according to the divided difference information group. The target difference information may be data information of a data storage location where a difference in sample set between each data block and a corresponding data block of the user value fluctuation sample set satisfies a difference condition. And each data storage position corresponding to the data block has a corresponding relation with each data storage position corresponding to the data block corresponding to the user value fluctuation sample set. The difference condition may be that the difference between the sample sets is greater than a difference threshold. First, partition difference information corresponding to difference information greater than a difference threshold value may be selected from the above-described partition difference information group as target partition difference information. Then, storage location information of a data storage location corresponding to the target division difference information is determined as target difference information.
Optionally, each user value fluctuation sample with the corresponding storage position information in the corresponding data block of the user value fluctuation sample set as the target difference information is determined as a target user value fluctuation sample group.
In some embodiments, the executing entity may determine each user value fluctuation sample whose corresponding storage location information in the user value fluctuation sample set corresponding data block is the target difference information as the target user value fluctuation sample group.
Optionally, determining a sample weight corresponding to each target user value fluctuation sample in the target user value fluctuation sample group corresponding to the initial value detection information prediction model, so as to obtain a sample weight set.
In some embodiments, the executing entity may determine a sample weight corresponding to each target user value fluctuation sample in the target user value fluctuation sample group corresponding to the initial value detection information prediction model, so as to obtain a sample weight set. The user value detection information generation model includes: historical value detection information predicts a model sequence. Wherein the sample weights may characterize the degree to which the model learns the sample semantics. The sample weights may be values of 0-1. The sample weight of the initial value detection information prediction model may be a weight that characterizes a degree of importance of the initial value detection information prediction model to learn sample semantics. The initial value detection information prediction model may be an untrained XGBoost model. The historical value detection information prediction model sequence may be a sequence of individual historical value detection information prediction models (XGBoost models) that are trained according to training time.
In an actual application scenario, the executing body may determine a sample weight corresponding to each target user value fluctuation sample in the target user value fluctuation sample group, where the sample weight corresponds to the initial value detection information prediction model, through the following steps:
first, for the historical value detection information prediction model in the above-described sequence of historical value detection information prediction models, the following generation steps are performed:
first, in response to determining that a model that is located before the historical value detection information prediction model at a corresponding model position exists in the sequence of historical value detection information prediction models, a latest historical value detection information prediction model that is located before the historical value detection information prediction model is determined as a target historical value detection information prediction model. For example, the historical value detection information prediction model sequence may include: the system comprises a first historical value detection information prediction model, a second historical value detection information prediction model and a third historical value detection information prediction model. The historical value detection information prediction model is a second historical value detection information prediction model. The most recent historical value detection information prediction model is the first historical value detection information prediction model.
And then, determining a sample weight corresponding to each target user value fluctuation sample in the target user value fluctuation sample group corresponding to the target historical value detection information prediction model as an adjacent sample weight to obtain an adjacent sample weight set.
And then, generating a sample weight set corresponding to the historical value detection information prediction model according to the adjacent sample weight set. For example, the execution subject may update a formula using the sample weight of Adaboost to generate a sample weight set corresponding to the historical value detection information prediction model.
And secondly, determining a sample weight set corresponding to the historical value detection information prediction model positioned at the target position in the historical value detection information prediction model sequence as a target sample weight set. The historical value detection information prediction models in the historical value detection information prediction model sequence are ordered by the sequence of training time, and the target position can be the position of the first historical value detection information prediction model in the historical value detection information prediction model sequence.
Thirdly, generating a sample weight set corresponding to the initial value detection information prediction model according to the target sample weight set. And updating a formula by using the sample weight of Adaboost to generate a sample weight set corresponding to the initial value detection information prediction model.
Optionally, training the initial value detection information prediction model based on the target user value fluctuation sample set and the sample weight set to obtain a trained value detection information prediction model.
In some embodiments, the executing entity may train the initial value detection information prediction model based on the target user value fluctuation sample set and the sample weight set to obtain a trained value detection information prediction model. Firstly, the execution subject can take a target user value fluctuation sample group as model input data, take the sample weight group as data weight corresponding to the model input data, and perform model training on the initial value detection information prediction model through a training mode of a deep neural network to obtain a trained value detection information prediction model.
Optionally, the trained value detection information prediction model is fused with a pre-trained user value detection information generation model to generate a user value detection information prediction model.
In some embodiments, the executing entity may fuse the trained value detection information prediction model with a pre-trained user value detection information generation model to generate a user value detection information prediction model. The value detection information generation model may be a model (convolutional neural network model) that generates value detection information. For example, the value detection information generation model may detect the credit value of the user. The trained value detection information prediction model and the pre-trained user value detection information generation model can be integrated and fused to generate the user value detection information prediction model.
Further, the user value fluctuation information set is input into a pre-trained user value detection information prediction model, and a user value detection information set is obtained.
In some embodiments, the executing entity may input the user value fluctuation information set into a pre-trained user value detection information prediction model to obtain a user value detection information set. Wherein one user value detection information corresponds to one user value fluctuation information. The user value detection information prediction model may be a neural network model which is trained in advance, takes user value fluctuation information as input, and takes user value detection information as output. For example, the user value detection information prediction model may be a convolutional neural network model, or may be an integrated XGBoost model. The user value detection information may represent a recommended value degree (such as credit value, asset value, purchase value, etc.) of the user. The higher the recommendation value, the higher the value of pushing the vehicle information to the user.
Further, the user value detection information which meets the user value early warning condition in the user value detection information set is determined to be abnormal user value detection information, and an abnormal user value detection information set is obtained.
In some embodiments, the executing entity may determine the user value detection information that satisfies the user value early warning condition in the set of user value detection information as abnormal user value detection information, to obtain an abnormal user value detection information set. The user value early warning condition may be: the recommendation value degree of the user value detection information representation is smaller than or equal to the preset value degree.
Further, risk early warning is carried out on each user corresponding to the abnormal user value detection information group.
In some embodiments, the executing body may perform risk early warning on each user corresponding to the abnormal user value detection information set. For example, user information of each user may be marked as risk user information, and voice prompts may be given to customer service personnel.
For the background technology, when the value information of the user is changed, abnormal user value information is difficult to detect in time, and traffic pushing resources are further wasted. ". The method can be solved by the following steps: first, a user value fluctuation sample set is obtained. And secondly, obtaining a historical sample division result corresponding to each data block in the historical user value fluctuation sample set, and obtaining a historical sample division result set. Then, determining the current sample division result of the data block corresponding to the user value fluctuation sample set. And then determining the division difference information between each historical sample division result in the historical sample division result set and the current sample division result to obtain a division difference information set. And then, generating target difference information according to the divided difference information groups. Then, each user value fluctuation sample with the corresponding storage position information in the corresponding data block of the user value fluctuation sample set as the target difference information is determined as a target user value fluctuation sample group. Here, a sample group with changed sample distribution value fluctuation can be selected, so that the problem of sample distribution fluctuation can be effectively solved, and the situation of inaccurate value detection caused by the sample distribution fluctuation problem can be avoided. And then, determining the sample weight corresponding to each target user value fluctuation sample in the target user value fluctuation sample group corresponding to the initial value detection information prediction model, and obtaining a sample weight set. Therefore, the sample weight corresponding to each user value fluctuation sample is determined and used for carrying out targeted model training on the subsequent initial value detection information prediction model, so that the initial value detection information prediction model can learn the characteristic information of more fluctuation samples. Thus, the value information of the user can be detected more accurately. And then, training the initial value detection information prediction model based on the target user value fluctuation sample group and the sample weight group to obtain a trained value detection information prediction model. And finally, fusing the trained value detection information prediction model with a pre-trained user value detection information generation model to generate a user value detection information prediction model. Therefore, through fusion of the models, accurate value detection of the user value fluctuation information can be achieved. Therefore, abnormal user value information can be accurately and timely detected, information pushing to abnormal users is avoided, and waste of pushing resources is reduced.
Fig. 2 is a schematic block diagram of a structure of a computer device according to an embodiment of the disclosure. The computer device may be a terminal.
As shown in fig. 2, the computer device includes a processor, a memory, and a network interface connected by a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program comprises program instructions that, when executed, cause the processor to perform any one of a number of risk early warning methods based on a vehicle user tag.
The processor is used to provide computing and control capabilities to support the operation of the entire computer device.
The internal memory provides an environment for the execution of a computer program in the non-volatile storage medium, which when executed by the processor, causes the processor to perform any one of a number of vehicle user tag-based risk warning methods.
The network interface is used for network communication such as transmitting assigned tasks and the like. Those skilled in the art will appreciate that the architecture shown in fig. 2 is merely a block diagram of some of the architecture relevant to the disclosed aspects and is not limiting of the computer device to which the disclosed aspects apply, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
It should be appreciated that the processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field-programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of: acquiring a user vehicle circulation information set in a preset historical time period, wherein the user vehicle circulation information in the user vehicle circulation information set comprises: vehicle type, vehicle value information, vehicle circulation type and value circulation node information; classifying the user vehicle circulation information set according to the vehicle type and the vehicle circulation type included in the user vehicle circulation information set to obtain a user vehicle circulation information set; inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence; selecting user vehicle circulation information meeting initial early warning conditions from the user vehicle circulation information set as user vehicle circulation information to be detected to obtain a user vehicle circulation information set to be detected, wherein the initial early warning conditions are as follows: the vehicle circulation type included in the user vehicle circulation information is a target vehicle circulation type; inputting the user vehicle circulation information group to be detected into a pre-trained user value information prediction model to obtain a user value information group, wherein one user vehicle circulation information to be detected corresponds to one user value information; for each piece of user value information in the user value information, generating user early warning information in response to determining that the user value information meets the early warning condition, and sending the user early warning information to an associated user early warning terminal for early warning.
Embodiments of the present disclosure also provide a computer readable storage medium having a computer program stored thereon, where the computer program includes program instructions, and a method implemented when the program instructions are executed may refer to various embodiments of the present disclosure risk early warning method based on a vehicle user tag.
The computer readable storage medium may be an internal storage unit of the computer device according to the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may be an external storage device of the computer device, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present disclosure are merely for description and do not represent advantages or disadvantages of the embodiments. While the invention has been described with reference to certain preferred embodiments, it will be apparent to one skilled in the art that various changes and substitutions can be made herein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (6)
1. A risk early warning method based on a vehicle user tag, the method comprising:
acquiring a user vehicle circulation information set in a preset historical time period, wherein the user vehicle circulation information in the user vehicle circulation information set comprises: vehicle type, vehicle value information, vehicle circulation type and value circulation node information;
classifying the user vehicle circulation information set according to the vehicle type and the vehicle circulation type included in the user vehicle circulation information set to obtain a user vehicle circulation information set;
inputting the user vehicle circulation information set into a pre-trained user vehicle recommendation model to obtain a recommended vehicle information sequence;
Selecting user vehicle circulation information meeting initial early warning conditions from the user vehicle circulation information set as user vehicle circulation information to be detected to obtain a user vehicle circulation information set to be detected, wherein the initial early warning conditions are as follows: the vehicle circulation type included in the user vehicle circulation information is a target vehicle circulation type;
acquiring an initial user value information prediction model file, wherein the initial user value information prediction model file comprises tree structure parameters of a nested relation, and an initial user value information prediction model in the initial user value information prediction model file is a pre-trained model for generating user value information;
performing format conversion processing on the initial user value information prediction model file to generate a converted initial user value information prediction model file;
performing dimension compression processing on the conversion initial user value information prediction model file to obtain a compression initial user value information prediction model file;
acquiring a user verification data set;
inputting the user verification data set into the initial user value information prediction model file to obtain a first verification value corresponding to the user verification data set;
Inputting the user verification data set into the compressed initial user value information prediction model file to obtain a second verification value corresponding to the user verification data set;
generating a verification result through the first verification value and the second verification value;
determining a model in the compressed initial user value information prediction model file as a user value information prediction model in response to determining that the verification result characterization verification passes;
inputting the user vehicle circulation information group to be detected into a pre-trained user value information prediction model to obtain a user value information group, wherein one user vehicle circulation information to be detected corresponds to one user value information;
for each piece of user value information in the user value information, generating user early warning information in response to determining that the user value information meets an early warning condition, and sending the user early warning information to an associated user early warning terminal for early warning;
acquiring user value fluctuation information corresponding to each user vehicle circulation information in the user vehicle circulation information set to acquire a user value fluctuation information set;
acquiring a user value fluctuation sample set, wherein the user value fluctuation sample set is a sample set stored in a data block form;
Acquiring a historical sample division result corresponding to each data block in the user value fluctuation sample set to obtain a historical sample division result group;
determining a current sample division result of a data block corresponding to the user value fluctuation sample set;
determining division difference information between each historical sample division result in the historical sample division result set and the current sample division result to obtain a division difference information set;
generating target difference information according to the divided difference information groups;
determining each user value fluctuation sample with the corresponding storage position information in the corresponding data block of the user value fluctuation sample set as the target difference information as a target user value fluctuation sample group;
determining sample weights corresponding to the initial value detection information prediction model and corresponding to each target user value fluctuation sample in the target user value fluctuation sample group to obtain sample weight groups;
training the initial value detection information prediction model based on the target user value fluctuation sample group and the sample weight group to obtain a trained value detection information prediction model;
fusing the trained value detection information prediction model with a pre-trained user value detection information generation model to generate a user value detection information prediction model;
Inputting the user value fluctuation information set into a pre-trained user value detection information prediction model to obtain a user value detection information set, wherein one user value detection information corresponds to one user value fluctuation information;
determining the user value detection information which meets the user value early warning condition in the user value detection information set as abnormal user value detection information to obtain an abnormal user value detection information set;
performing risk early warning on each user corresponding to the abnormal user value detection information group;
the method is characterized in that the determining the current sample division result of the data block corresponding to the user value fluctuation sample set comprises the following steps:
determining a sample attribute set corresponding to the user value fluctuation sample set;
determining first attribute division information corresponding to each sample attribute in the sample attribute set based on the user value fluctuation sample set;
determining sample attributes, corresponding to the sample attribute set, of which the first attribute partition information does not accord with the partition attribute conditions as first sample attributes, and obtaining a first sample attribute group;
selecting a first target sample attribute from the first sample attribute group;
Dividing the user value fluctuation sample set based on the first target sample attribute to obtain a user value fluctuation sample set;
for each user value fluctuation sample set in the user value fluctuation sample set, performing the steps of:
determining second attribute division information corresponding to each sample attribute in the sample attribute set according to the user value fluctuation sample group;
selecting a sample attribute of which the corresponding second attribute division information does not accord with the division attribute condition from the sample attribute set as a second sample attribute to obtain a second sample attribute group;
selecting a second target sample attribute from the second sample attribute group;
dividing the user value fluctuation sample group according to the second target sample attribute to obtain an alternative user value fluctuation sample group set;
for each alternative user value fluctuation sample group in the alternative user value fluctuation sample group set, determining third attribute partition information corresponding to the alternative user value fluctuation sample group in the sample attribute set;
in response to determining that the obtained third attribute partition information meets the corresponding partition attribute condition and the number of samples included in the alternative user value fluctuation sample group is smaller than a preset value, determining a sample partition result corresponding to the alternative user value fluctuation sample group set as a sample partition result corresponding to the user value fluctuation sample group;
And determining each sample division result as a current sample division result of the data block corresponding to the user value fluctuation sample set.
2. The method according to claim 1, wherein the method further comprises:
for each user vehicle circulation information in the user vehicle circulation information set, executing the following processing steps:
determining whether value transfer node information included in the user vehicle transfer information meets a value early warning condition;
and responding to the fact that the value circulation node information meets the value early warning condition, and carrying out value early warning on the user corresponding to the user vehicle circulation information.
3. The method according to claim 1, wherein the method further comprises:
for each piece of user value information in the user value information, in response to determining that the user value information does not meet the early warning condition, selecting vehicle circulation policy information corresponding to the user value information from a preset vehicle circulation policy information group, and sending the vehicle circulation policy information to a corresponding customer service terminal, wherein the vehicle circulation policy information in the vehicle circulation policy information group comprises: vehicle circulation policy information corresponding to a first vehicle circulation type and vehicle circulation policy information corresponding to a second vehicle circulation type.
4. A method according to claim 3, characterized in that the method further comprises:
and for each piece of user value information in the user value information, responding to the fact that the user value information does not meet the early warning condition, and sending the recommended vehicle information sequence to a user terminal corresponding to the user value information.
5. A computer device, wherein the computer device comprises a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, implements the steps of the method according to any of claims 1-4.
6. A computer readable storage medium, wherein the computer readable storage medium has stored thereon a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311252821.XA CN116993396B (en) | 2023-09-27 | 2023-09-27 | Risk early warning method based on vehicle user tag and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311252821.XA CN116993396B (en) | 2023-09-27 | 2023-09-27 | Risk early warning method based on vehicle user tag and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116993396A CN116993396A (en) | 2023-11-03 |
| CN116993396B true CN116993396B (en) | 2023-12-22 |
Family
ID=88526971
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311252821.XA Active CN116993396B (en) | 2023-09-27 | 2023-09-27 | Risk early warning method based on vehicle user tag and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116993396B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021004121A1 (en) * | 2019-07-05 | 2021-01-14 | 深圳壹账通智能科技有限公司 | Vehicle insurance recommendation method, apparatus and device, and computer-readable storage medium |
| CN113204577A (en) * | 2021-04-15 | 2021-08-03 | 北京沃东天骏信息技术有限公司 | Information pushing method and device, electronic equipment and computer readable medium |
| CN113407827A (en) * | 2021-06-11 | 2021-09-17 | 广州三七极创网络科技有限公司 | Information recommendation method, device, equipment and medium based on user value classification |
| CN115062231A (en) * | 2022-08-18 | 2022-09-16 | 南京三百云信息科技有限公司 | Data processing method and device suitable for vehicle source recommendation |
| CN115981522A (en) * | 2023-03-16 | 2023-04-18 | 北京北汽鹏龙汽车服务贸易股份有限公司 | Method, equipment and computer medium for processing early warning vehicle circulation information |
-
2023
- 2023-09-27 CN CN202311252821.XA patent/CN116993396B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021004121A1 (en) * | 2019-07-05 | 2021-01-14 | 深圳壹账通智能科技有限公司 | Vehicle insurance recommendation method, apparatus and device, and computer-readable storage medium |
| CN113204577A (en) * | 2021-04-15 | 2021-08-03 | 北京沃东天骏信息技术有限公司 | Information pushing method and device, electronic equipment and computer readable medium |
| CN113407827A (en) * | 2021-06-11 | 2021-09-17 | 广州三七极创网络科技有限公司 | Information recommendation method, device, equipment and medium based on user value classification |
| CN115062231A (en) * | 2022-08-18 | 2022-09-16 | 南京三百云信息科技有限公司 | Data processing method and device suitable for vehicle source recommendation |
| CN115981522A (en) * | 2023-03-16 | 2023-04-18 | 北京北汽鹏龙汽车服务贸易股份有限公司 | Method, equipment and computer medium for processing early warning vehicle circulation information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116993396A (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111275491B (en) | Data processing method and device | |
| CN107563757B (en) | Data risk identification method and device | |
| CN110554958B (en) | Graph database testing method, system, device and storage medium | |
| CN111797320B (en) | Data processing method, device, equipment and storage medium | |
| CN110991474A (en) | Machine learning modeling platform | |
| KR102109583B1 (en) | Method and Apparatus for pricing based on machine learning | |
| CN110634021A (en) | Big data based vehicle estimation method, system, device and readable storage medium | |
| CN109102206A (en) | A kind of evaluation method and relevant device of Automobile Service Factory | |
| CN112070216A (en) | Method and system for training neural network model based on graph computing system | |
| CN105989066A (en) | Information processing method and device | |
| Sun et al. | On the tradeoff between sensitivity and specificity in bus bunching prediction | |
| US20230099627A1 (en) | Machine learning model for predicting an action | |
| CN108805332B (en) | Feature evaluation method and device | |
| CN111507541B (en) | Goods quantity prediction model construction method, goods quantity measurement device and electronic equipment | |
| CN111680645B (en) | Garbage classification treatment method and device | |
| CN116993396B (en) | Risk early warning method based on vehicle user tag and computer equipment | |
| CN110751501B (en) | Commodity shopping guide method, device, equipment and storage medium | |
| CN116503092A (en) | User reservation intention recognition method and device, electronic equipment and storage medium | |
| CN113935788B (en) | Model evaluation method, device, equipment and computer readable storage medium | |
| CN111831892A (en) | Information recommendation method, information recommendation device, server and storage medium | |
| CN113239272B (en) | Intention prediction method and intention prediction device of network management and control system | |
| CN115203556A (en) | Score prediction model training method and device, electronic equipment and storage medium | |
| CN115278757A (en) | Method and device for detecting abnormal data and electronic equipment | |
| US20210168195A1 (en) | Server and method for controlling server | |
| CN110278524B (en) | User position determining method, graph model generating method, device and server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |