Historical recall data filtering method and device based on driver click behavior
By clustering and outlier filtering of driver click behavior data, the noise problem in driver click behavior is solved, the accuracy of loading and unloading points is improved, and the user experience and order service quality are improved.
Patent Information
- Application Number
- CN202411586718.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-06
- Publication Date
- 2025-02-14
AI Technical Summary
In the prior art, drivers may click in advance during the completion of orders or click in advance or leave the loading and unloading place after delaying clicking, resulting in noise in clicking in the clicking behavior data, which affects the order completion status and the accuracy of the loading and unloading place, and thus affects the Order fee calculation and user experience.
By obtaining the arrival click behavior data in the target driver's historical order record, clustering the data using the clustering algorithm, determining outliers (indicating the arrival click behavior abnormal point), and performing outliers according to preset filtering rules Filter processing.
Through clustering and outlier point filtering, the noise in click behavior data is removed, and the accuracy of loading and unloading points is improved, which is conducive to subsequent order billing, route recommendation and other services, and improve user experience.
Smart Images
Figure CN119441737A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of freight technology, and in particular to a method and device for filtering historical recall data based on driver click behavior. Background Art
[0002] Currently, when completing an order, the driver will click on the arrival location for loading / unloading, and the system will record the driver's arrival point for loading / unloading. However, since the driver's click behavior may include clicking before arriving at the loading / unloading location or clicking after leaving the loading / unloading location, the driver's behavior data contains a certain amount of noise. This makes it impossible to determine whether the order is completed, whether the loading / unloading location is incorrect, and other issues. In particular, since most orders are charged based on the order itinerary, such as moving services, the order fee is calculated based on the distance between the loading / unloading locations, the length of stay, etc., when the loading / unloading arrival point is incorrect, the order fee will be inaccurate, affecting the user experience, and not conducive to subsequent services such as route recommendations. Summary of the invention
[0003] Based on this, it is necessary to provide a method and device for filtering historical recall data based on driver click behavior in order to solve at least one problem existing in the above-mentioned prior art.
[0004] In the first aspect, a method for filtering historical recall data based on driver click behavior is provided, comprising:
[0005] Acquire historical order records of a target driver, wherein the historical order records include arrival click behavior data of the target driver when arriving at a loading and unloading location;
[0006] Clustering the arrival click behavior data;
[0007] Based on the clustering results, outliers are determined, where the outliers represent abnormal points of the click behavior;
[0008] Obtain target outlier filtering rules, and filter the outliers according to the target outlier filtering rules.
[0009] In one embodiment, before obtaining the target outlier filtering rule, the method includes: obtaining a historical order record of a target driver, wherein the historical order record includes arrival click behavior data of the target driver when arriving at a loading and unloading location;
[0010] Clustering the arrival click behavior data;
[0011] Based on the clustering results, outliers are determined, where the outliers represent abnormal points of the click behavior;
[0012] When the preset cluster filtering condition is met, the target outlier filtering rule is obtained, and the outliers are filtered according to the target outlier filtering rule.
[0013] In one embodiment, before obtaining the target outlier filtering rule, the process includes:
[0014] Determine the number of historical orders associated with the target driver;
[0015] Determining whether the number of historical orders is greater than a first preset threshold;
[0016] If the number of historical orders is greater than the first preset threshold, the preset cluster filtering condition is met.
[0017] In one embodiment, the obtaining of a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0018] Determine the user type of the target driver based on the number of historical orders associated with the target driver, where the user type includes a low-frequency user, a medium-frequency user, and a high-frequency user;
[0019] A target outlier filtering rule matching the user type is obtained, and the outliers are filtered according to the target outlier filtering rule matching the user type.
[0020] In one embodiment, the obtaining of a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0021] Determine whether the number of historical orders associated with the target driver is greater than the first preset threshold and less than / equal to the second preset threshold;
[0022] If so, determining whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a first preset ratio threshold;
[0023] If so, the outliers are filtered out.
[0024] In one embodiment, the obtaining of a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0025] Determine whether the number of historical orders associated with the target driver is greater than the second preset threshold and less than / equal to a third preset threshold;
[0026] If yes, determine whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a second preset ratio threshold;
[0027] If so, the outliers are filtered out.
[0028] In one embodiment, the obtaining of a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0029] Determining whether the number of historical orders associated with the target driver is greater than the third preset threshold;
[0030] If so, the outliers are filtered out.
[0031] In one embodiment, clustering the arrival click behavior data includes:
[0032] Calculate the density of each of the arrival click behavior data and determine the density peak point;
[0033] The density peak point is used as the cluster center point to perform density nearest neighbor clustering to obtain the clustering result.
[0034] In a second aspect, a device for filtering historical recall data based on driver click behavior is provided, the device comprising:
[0035] An arrival click behavior data acquisition unit, used to acquire a historical order record of a target driver, wherein the historical order record includes arrival click behavior data of the target driver when arriving at a loading and unloading location;
[0036] A clustering unit, used for clustering the arrival click behavior data;
[0037] An outlier and non-outlier determination unit, used to determine outliers based on clustering results, wherein the outliers represent abnormal points of the click behavior;
[0038] The outlier filtering unit obtains a target outlier filtering rule and performs filtering processing on the outliers according to the target outlier filtering rule.
[0039] In a third aspect, a computer device is provided, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the processor executes the computer-readable instructions, the steps of the method for filtering historical recall data based on driver click behavior as described above are implemented.
[0040] In a fourth aspect, a readable storage medium is provided, wherein the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the method for filtering historical recall data based on driver click behavior as described above are implemented.
[0041] The above-mentioned historical recall data filtering method, device, computer equipment and storage medium based on the driver's click behavior, and its method implementation include: obtaining the historical order records of the target driver, the historical order records include the arrival click behavior data of the target driver at the loading and unloading location; clustering the arrival click behavior data; based on the clustering results, determining the outliers, the outliers represent the abnormal points of arrival click behavior; obtaining the target outlier filtering rules, and filtering the outliers according to the target outlier filtering rules. In the embodiment of the present application, by using a clustering algorithm to cluster the click behavior data, the irregularity of the loading / unloading point group can be better adapted. And through the outlier filtering rules, the clustered outliers are filtered to remove the noise in the arrival click behavior data, improve the accuracy of the loading / unloading point, facilitate subsequent recommendations and order billing services, and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0043] Figure 1 This is a flow chart of a method for filtering historical recall data based on driver click behavior in one embodiment of the present application;
[0044] Figure 2 is a structural schematic diagram of a historical recall data filtering device based on driver click behavior in one embodiment of the present application;
[0045] Figure 3 It is a schematic diagram of a computer device in one embodiment of the present application. DETAILED DESCRIPTION
[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0047] In one embodiment, if Figure 1 As shown, a method for filtering historical recall data based on driver click behavior is provided, comprising the following steps:
[0048] In step S110, the historical order record of the target driver is obtained, and the historical order record includes the arrival click behavior data of the target driver when arriving at the loading and unloading location;
[0049] In an embodiment of the present application, historical order records may include data such as routes, order types, actual loading locations, actual unloading locations, clicked loading point arrival locations, and clicked unloading point arrival locations. Clicking the loading point arrival location means that when a user reaches the loading point, they can perform a click operation to reach the loading point through an application, mini-program, public account, etc., and obtain the location information when the click behavior to reach the loading point is performed based on the positioning system of the client. Similarly, the loading point arrival location refers to the behavioral data of the click arrival operation performed by the user when he reaches the unloading point. The click behavior data is recalled by collecting historical order records.
[0050] Among them, the historical order records can be order data within a period of time in the past, such as one month, three months, etc., which can be updated in real time according to the latest order data. In addition, classified statistics can be performed according to different drivers and loading / unloading points, such as counting the historical order data of the same driver, or counting the historical order data of the same loading / unloading point, so that the same type of click behavior data can be filtered.
[0051] In step S120, clustering the arrival click behavior data;
[0052] In an embodiment of the present application, the arrival click behavior data can be clustered by the density clustering algorithm dbscan. Specifically, an arrival click behavior data set can be constructed, and then the local reachable density of the data in the data set can be calculated using the distance matrix, a decision diagram can be generated, and the distance center can be determined. Then, the data in the data set can be sorted in descending order according to the local reachable density, and weighted allocation can be performed in turn to achieve clustering processing of the data set. It should be noted that the present application can also use other clustering algorithms, such as grid algorithms, model algorithms, partitioning methods, hierarchical methods, etc. to achieve clustering processing of arrival click behavior data.
[0053] In step S130, outliers are determined based on the clustering results, where the outliers represent abnormal points of the click behavior;
[0054] In an embodiment of the present application, each collected click-to-reach behavior data is taken as a candidate point, and the candidate points are clustered to form at least one cluster, which can be composed of multiple cluster points, and each cluster point can be connected by an edge. The cluster point can be used as a non-outlier point, which can be used to characterize the normal point of the arrival click behavior, while other discrete points can be used as outliers to characterize the abnormal point of the arrival click behavior. The outlier point and the non-outlier point can be labeled to characterize the outlier point and the non-outlier point, respectively, for example, using letters, symbols, colors, etc. for marking. This is to facilitate the subsequent distinction and quantitative statistics of outliers and non-outliers.
[0055] In an embodiment of the present application, after obtaining outlier points and non-outlier points, the number of outlier points and the total number of points can be counted. The total number of points includes the sum of the number of all outlier points and non-outlier points. When the number of outlier points accounts for more than half of the total number of points, that is, greater than 50%, it means that clustering has failed or there is an abnormality in the click behavior data. At this time, clustering processing can be performed again, or the click behavior data can be manually verified.
[0056] In step S140, a target outlier filtering rule is obtained, and the outliers are filtered according to the target outlier filtering rule.
[0057] In an embodiment of the present application, based on the number of historical order records associated with the target driver, the user type of the target driver can be determined, such as a low-frequency user, a medium-frequency user, and a high-frequency user, and then the target outlier filtering rule matching the target driver is obtained according to the user type to which the target driver belongs, and the outliers are filtered according to the target outlier filtering rule. Exemplarily, when the number of historical orders associated with the target driver is lower than the first preset single number threshold, such as 10 orders, no filtering is required. When the number of historical orders managed by the target driver is greater than the second preset single number threshold, such as greater than 50 orders, the outliers can be filtered and non-outliers can be retained for output.
[0058] In an embodiment of the present application, a method for filtering historical recall data based on driver click behavior is provided, and the method is implemented, including: obtaining the historical order records of the target driver, the historical order records including the arrival click behavior data of the target driver at the loading and unloading location; clustering the arrival click behavior data; determining outliers based on the clustering results, the outliers characterize the abnormal points of arrival click behavior; obtaining target outlier filtering rules, and filtering the outliers according to the target outlier filtering rules. In an embodiment of the present application, by clustering the click behavior data using a clustering algorithm, the irregularity of the loading / unloading point group can be better adapted. And by filtering the clustered outliers through the outlier filtering rules, the noise in the arrival click behavior data can be removed, the accuracy of the loading / unloading point can be improved, which is conducive to subsequent recommendations and order billing services, and can improve the user experience.
[0059] In an embodiment of the present application, before obtaining the target outlier filtering rule, the process includes:
[0060] Determine the number of historical orders associated with the target driver;
[0061] Determining whether the number of historical orders is greater than a first preset threshold;
[0062] If the number of historical orders is greater than the first preset threshold, the preset cluster filtering condition is met.
[0063] Optionally, the historical orders may be associated with a driver identifier, such as a driver name, ID, etc. By obtaining the target driver's identifier, the historical orders associated with the target driver are searched in all historical orders based on the identifier, and the number of historical order records is counted. If the number of historical order records is greater than a first preset threshold, such as greater than 10 orders, it can be considered that it meets the preset clustering filtering condition, otherwise, there is no need to filter the discrete points.
[0064] In an embodiment of the present application, the step of obtaining a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0065] Determine the user type of the target driver based on the number of historical orders associated with the target driver, where the user type includes a low-frequency user, a medium-frequency user, and a high-frequency user;
[0066] A target outlier filtering rule matching the user type is obtained, and the outliers are filtered according to the target outlier filtering rule matching the user type.
[0067] Optionally, based on the number of historical orders associated with the target driver, the user type of the target driver may be determined, and the user type may include a low-frequency user, a medium-frequency user, a high-frequency user, etc. Exemplarily, when the number of historical orders is greater than 10 and less than 50, the target driver may be a low-frequency user; when the number of historical orders is greater than 50 and less than 150, the target driver may be a medium-frequency user; when the number of historical orders is greater than 150, the target driver may be a high-frequency user, and different user types may correspond to different target outlier filtering rules.
[0068] In an embodiment of the present application, the step of obtaining a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0069] Determine whether the number of historical orders associated with the target driver is greater than the first preset threshold and less than / equal to the second preset threshold;
[0070] If so, determining whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a first preset ratio threshold;
[0071] If so, the outliers are filtered out.
[0072] Optionally, when the number of historical orders associated with the target driver is greater than a first preset threshold and less than / equal to a second preset threshold, for example, greater than 10 orders and less than or equal to 50 orders, the number of outliers and the number of all points can be counted based on the clustering results to determine the proportion of the number of outliers to the number of all points. For example, when it does not exceed 20%, the outliers can be filtered out and the non-outlier output can be retained.
[0073] In an embodiment of the present application, the step of obtaining a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0074] Determine whether the number of historical orders associated with the target driver is greater than the second preset threshold and less than / equal to a third preset threshold;
[0075] If yes, determine whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a second preset ratio threshold;
[0076] If so, the outliers are filtered out.
[0077] Optionally, when the number of historical orders associated with the target driver is greater than the second preset threshold and less than / equal to the third preset threshold, for example, greater than 50 orders and less than or equal to 150 orders, the number of outliers and the number of all points can be counted according to the clustering results to determine the proportion of the number of outliers to the number of all points. For example, when it does not exceed 50%, the outliers can be filtered out and the non-outlier output can be retained. Otherwise, no filtering or re-clustering can be performed.
[0078] In an embodiment of the present application, the step of obtaining a target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes:
[0079] Determining whether the number of historical orders associated with the target driver is greater than the third preset threshold;
[0080] If so, the outliers are filtered out.
[0081] Optionally, when the number of historical orders associated with the target driver is greater than a third preset threshold and less than / equal to a second preset threshold, for example, greater than 150 orders, all outliers can be filtered out and non-outlier outputs are retained.
[0082] In an embodiment of the present application, clustering the arrival click behavior data includes:
[0083] Calculate the density of each of the arrival click behaviors and determine the density peak point;
[0084] The density peak point is used as the cluster center point to perform density nearest neighbor clustering to obtain the clustering result.
[0085] Optionally, each collected click-to-reach behavior data can be used as a candidate point, and the candidate points can be clustered to form at least one cluster, which is composed of multiple cluster points, and each cluster point can be connected by an edge. The cluster point can be used as a non-outlier point, which can be used to characterize the normal point of the click-to-reach behavior, while other discrete points can be used as outliers to characterize the abnormal point of the click-to-reach behavior. The outlier point and the non-outlier point can be labeled to characterize the outlier point and the non-outlier point, respectively, for example, using letters, symbols, colors, etc. for marking. This allows for subsequent distinction and quantitative statistics of outliers and non-outliers.
[0086] In the embodiment of the present application, by using a clustering algorithm to cluster the click behavior data, the irregularity of the loading / unloading point group can be better adapted. And by filtering the clustered outliers through the outlier filtering rule, the noise in the click behavior data can be removed, the accuracy of the loading / unloading point can be improved, which is conducive to subsequent recommendations and order billing services, and can improve the user experience.
[0087] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0088] In one embodiment, a device for filtering historical recall data based on driver click behavior is provided, and the device for filtering historical recall data based on driver click behavior corresponds to the method for filtering historical recall data based on driver click behavior in the above embodiment. Figure 2 As shown, the historical recall data filtering device based on driver click behavior includes an arrival click behavior data acquisition unit 10, a clustering unit 20, an outlier and non-outlier determination unit 30 and an outlier filtering unit 40. The functional modules are described in detail as follows:
[0089] The arrival click behavior data acquisition unit 10 is used to acquire the historical order records of the target driver, wherein the historical order records include the arrival click behavior data of the target driver when arriving at the loading and unloading location;
[0090] A clustering unit 20, used for clustering the arrival click behavior data;
[0091] An outlier and non-outlier determination unit 30, for determining outliers based on clustering results, wherein the outliers represent abnormal points of the click behavior;
[0092] The outlier filtering unit 40 obtains a target outlier filtering rule, and performs filtering processing on the outliers according to the target outlier filtering rule.
[0093] In an embodiment of the present application, the device further includes: a preset cluster filtering condition judgment unit, which is used to:
[0094] Determine the number of historical orders associated with the target driver;
[0095] Determining whether the number of historical orders is greater than a first preset threshold;
[0096] If the number of historical orders is greater than the first preset threshold, the preset cluster filtering condition is met.
[0097] In one embodiment of the present application, the outlier filtering unit 40 is further configured to:
[0098] Determine the user type of the target driver based on the number of historical orders associated with the target driver, where the user type includes a low-frequency user, a medium-frequency user, and a high-frequency user;
[0099] A target outlier filtering rule matching the user type is obtained, and the outliers are filtered according to the target outlier filtering rule matching the user type.
[0100] In one embodiment of the present application, the outlier filtering unit 40 is further configured to:
[0101] Determine whether the number of historical orders associated with the target driver is greater than the first preset threshold and less than / equal to the second preset threshold;
[0102] If so, determining whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a first preset ratio threshold;
[0103] If so, the outliers are filtered out.
[0104] In one embodiment of the present application, the outlier filtering unit 40 is further configured to:
[0105] Determine whether the number of historical orders associated with the target driver is greater than the second preset threshold and less than / equal to a third preset threshold;
[0106] If yes, determine whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a second preset ratio threshold;
[0107] If so, the outliers are filtered out.
[0108] In one embodiment of the present application, the outlier filtering unit 40 is further configured to:
[0109] Based on the order frequency of the target driver, determine the number of historical orders associated with the target driver;
[0110] Determining whether the number of historical orders associated with the target driver is greater than the third preset threshold;
[0111] If so, the outliers are filtered out.
[0112] In one embodiment of the present application, the clustering unit 20 is further configured to:
[0113] Calculate the density of each of the arrival click behaviors and determine the density peak point;
[0114] The density peak point is used as the cluster center point to perform density nearest neighbor clustering to obtain the clustering result.
[0115] In the embodiment of the present application, by using a clustering algorithm to cluster the click behavior data, the irregularity of the loading / unloading point group can be better adapted. And by filtering the clustered outliers through the outlier filtering rule, the noise in the click behavior data can be removed, the accuracy of the loading / unloading point can be improved, which is conducive to subsequent recommendations and order billing services, and can improve the user experience.
[0116] For the specific definition of the historical recall data filtering device based on the driver's click behavior, please refer to the definition of the historical recall data filtering method based on the driver's click behavior above, which will not be repeated here. Each module in the above-mentioned historical recall data filtering device based on the driver's click behavior can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0117] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a historical recall data filtering method based on driver click behavior is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0118] In an embodiment of the present application, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the method for filtering historical recall data based on driver click behavior as described above are implemented.
[0119] In an embodiment of the application, a readable storage medium is provided, wherein the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the method for filtering historical recall data based on driver click behavior as described above are implemented.
[0120] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0121] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0122] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for filtering historical recall data based on driver click behavior, characterized in that: The method comprises: Acquire historical order records of a target driver, wherein the historical order records include arrival click behavior data of the target driver when arriving at a loading and unloading location; Clustering the arrival click behavior data; Based on the clustering results, outliers are determined, where the outliers represent abnormal points of the click behavior; When the preset cluster filtering condition is met, the target outlier filtering rule is obtained, and the outliers are filtered according to the target outlier filtering rule.
2. The method for filtering historical recall data based on driver click behavior according to claim 1, characterized in that: Before obtaining the target outlier filtering rule, the method includes: Determine the number of historical orders associated with the target driver; Determining whether the number of historical orders is greater than a first preset threshold; If the number of historical orders is greater than the first preset threshold, the preset cluster filtering condition is met.
3. The method for filtering historical recall data based on driver click behavior as claimed in claim 2, characterized in that: The obtaining of the target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes: Determine the user type of the target driver based on the number of historical orders associated with the target driver, where the user type includes a low-frequency user, a medium-frequency user, and a high-frequency user; A target outlier filtering rule matching the user type is obtained, and the outliers are filtered according to the target outlier filtering rule matching the user type.
4. The method for filtering historical recall data based on driver click behavior according to claim 2, characterized in that: The obtaining of the target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes: Determine whether the number of historical orders associated with the target driver is greater than the first preset threshold and less than / equal to the second preset threshold; If so, determining whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a first preset ratio threshold; If so, the outliers are filtered out.
5. The method for filtering historical recall data based on driver click behavior according to claim 4, characterized in that: The obtaining of the target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes: Determine whether the number of historical orders associated with the target driver is greater than the second preset threshold and less than / equal to a third preset threshold; If yes, determine whether the proportion of the number of outliers in the number of all cluster points is less than / equal to a second preset ratio threshold; If so, the outliers are filtered out.
6. The method for filtering historical recall data based on driver click behavior according to claim 5, characterized in that: The obtaining of the target outlier filtering rule and filtering the outliers according to the target outlier filtering rule includes: Determining whether the number of historical orders associated with the target driver is greater than the third preset threshold; If so, the outliers are filtered out.
7. The method for filtering historical recall data based on driver click behavior according to claim 1, characterized in that: The clustering of the arrival click behavior data includes: Calculate the density of each of the arrival click behavior data and determine the density peak point; The density peak point is used as the cluster center point to perform density nearest neighbor clustering to obtain the clustering result.
8. A device for filtering historical recall data based on driver click behavior, characterized in that: The device comprises: An arrival click behavior data acquisition unit, used to acquire a historical order record of a target driver, wherein the historical order record includes arrival click behavior data of the target driver when arriving at a loading and unloading location; A clustering unit, used for clustering the arrival click behavior data; An outlier and non-outlier determination unit, used to determine outliers based on clustering results, wherein the outliers represent abnormal points of the click behavior; The outlier filtering unit obtains a target outlier filtering rule and performs filtering processing on the outliers according to the target outlier filtering rule.
9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the steps of the method for filtering historical recall data based on driver click behavior as described in any one of claims 1 to 7 are implemented.
10. A readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the method for filtering historical recall data based on driver click behavior as described in any one of claims 1 to 7 are implemented.