CN114358873A - Abnormal order determining method for network appointment and related equipment - Google Patents
Abnormal order determining method for network appointment and related equipment Download PDFInfo
- Publication number
- CN114358873A CN114358873A CN202111616934.4A CN202111616934A CN114358873A CN 114358873 A CN114358873 A CN 114358873A CN 202111616934 A CN202111616934 A CN 202111616934A CN 114358873 A CN114358873 A CN 114358873A
- Authority
- CN
- China
- Prior art keywords
- order
- information
- abnormal
- user
- data set
- 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.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an abnormal order determining method and related equipment for online taxi appointment, wherein the method comprises the following steps: when an order of a network taxi appointment sent by a first terminal is received, acquiring first order information of the order, equipment information of the terminal, historical second order information related to an account corresponding to the order and user information of a user to which the account belongs; determining target parameters according to the first order information, the equipment information, the order information and the user information; and inputting the target parameters into the detection model to obtain an abnormal result of the order. In the invention, the target parameters of the input detection model comprise the installation times of an application program of an account, version parameters, login information, the travel parameters of an order, the type of a mobile phone number of a user, historical order withdrawing times, called single times and the estimated maximum order placing amount, namely whether the order is abnormal is determined in an all-around manner by combining all related information of the order, whether the order is abnormal is not determined by one kind of information, and the determination accuracy of the abnormal order of the online taxi appointment is higher.
Description
Technical Field
The invention relates to the technical field of network car booking, in particular to a method for determining an abnormal order of a network car booking and related equipment.
Background
With the combination of computer network technology and travel service, network car booking has become one of the important travel ways.
One important problem to be solved by network taxi booking is how to identify abnormal orders in travel services. Currently, an abnormal order is a single policy rule customized according to expert experience to determine whether the order is abnormal. Such as: and a plurality of users use the same IP address to carry out network car booking, and the car city is inconsistent with the IP attribution, so the travel orders of the users are abnormal orders. For another example, a large number of users complete registration and login on the same device and make a network appointment within a certain period of time, and these users may have a cheating suspicion of being subsidized by an illegal pickup platform, that is, the travel orders of these users are abnormal orders.
Therefore, the abnormal order is determined through single information, the identification accuracy rate of the abnormal order is difficult to guarantee, and the determination accuracy rate of the abnormal order of the online taxi appointment is low.
Disclosure of Invention
The invention provides a method and related equipment for determining an abnormal order of a network car booking, which are used for solving the problem of low accuracy rate of determining the abnormal order of the network car booking.
In one aspect, the invention provides a method for determining an abnormal order of a network taxi appointment, which comprises the following steps:
when a first order of a network taxi appointment sent by a first terminal is received, acquiring first order information of the first order, first equipment information of the first terminal, historical second order information related to a first account corresponding to the first order and first user information of a first user to which the first account belongs;
determining a first target parameter according to the first order information, the first device information, the second order information and the first user information, wherein the first target parameter comprises installation times, version parameters, first login information, travel parameters corresponding to the first order, a mobile phone number type of the first user, historical order withdrawal times, called single times and an estimated maximum order placing amount of the first user;
and inputting the first target parameter into a detection model to obtain an abnormal result of the first order, wherein the abnormal result is used for indicating whether the first order is abnormal or not.
In an embodiment, before the step of inputting the first target parameter into the detection model to obtain the output result, the method further includes:
obtaining each order data set, and determining a training sample corresponding to the order data set according to the order data set, wherein the order data set comprises a second order, third order information corresponding to the second order, second equipment information of a second terminal sending the second order, historical fourth order information associated with a second account corresponding to the second order, and second user information of a second user to which the second account belongs, and the training sample comprises installation times, version parameters, second login information, travel parameters corresponding to the second order, a mobile phone number type of the second user, historical order withdrawal times, number of call single times, and estimated maximum order placing amount of the second user of an application program corresponding to the second account;
and training a preset model according to each training sample to obtain the detection model.
In an embodiment, the step of determining the training sample corresponding to the order data set according to the order data set includes:
processing each order data set to obtain a second target parameter corresponding to each order data set, wherein the second target parameter comprises the installation times, the version parameter, second login information, a journey parameter corresponding to the second order, the mobile phone number type of the second user, the historical order withdrawing times, the number of called single times and the estimated maximum order placing amount of the application program corresponding to the second account;
setting a label for a second target parameter corresponding to the order data set according to the third order information in the order data set;
and generating a training sample corresponding to the order data set according to the second target parameter corresponding to the order data set and the label.
In an embodiment, the step of setting a label for a second target parameter corresponding to the order data set according to the third order information in the order data set includes:
determining whether the second order is a risk order according to the third order information in the order data set;
when the second order is a risk order, setting a first label for a second target parameter corresponding to the order data set, wherein the first label is used for indicating that the second order is a risk order;
and when the second order is not a risk order, setting a second label for a second target parameter corresponding to the order data set, wherein the second label is used for indicating that the second order is not a risk order.
In an embodiment, the step of training a preset model according to each training sample to obtain the detection model includes:
acquiring a training set and a verification set according to each training sample;
training the preset model according to the training samples in the training set to obtain a model to be determined;
inputting each training sample in the verification set into the model to be determined to obtain the accuracy and recall rate of the model to be determined;
and saving the model to be determined as the detection model when the accuracy rate is greater than or equal to a first threshold value and the recall rate is greater than or equal to a second threshold value.
In an embodiment, after the step of inputting the first target parameter into the detection model to obtain the abnormal result of the first order, the method further includes:
and when the abnormal result indicates that the first order is abnormal, adding the first account to a blacklist.
In another aspect, the present invention further provides an abnormal order determining apparatus for online taxi appointment, including:
the obtaining module is used for obtaining first order information of a first order, first equipment information of the first terminal, historical second order information related to a first account corresponding to the first order and first user information of a first user to which the first account belongs when the first order of the online taxi appointment sent by the first terminal is received;
a determining module, configured to determine a first target parameter according to the first order information, the first device information, the second order information, and the first user information, where the first target parameter includes installation times of an application corresponding to the first account, a version parameter, first login information, a travel parameter corresponding to the first order, a mobile phone number type of the first user, a historical order removal time, a call single time, and an estimated maximum order placing amount;
and the input module is used for inputting the first target parameter into a detection model to obtain an abnormal result of the first order, and the abnormal result is used for indicating whether the first order is abnormal or not.
In another aspect, the present invention further provides an abnormal order determining device for online taxi appointment, including: a memory and a processor;
the memory is to store program instructions;
the processor is used for calling the program instructions in the memory to execute the abnormal order determining method of the network appointment car.
In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon; when the computer program is executed, the abnormal order determining method of the online taxi appointment is realized.
In another aspect, the present invention further provides a computer program product, which includes a computer program, and the computer program is executed by a processor to implement the abnormal order determination method for online taxi appointment as described above.
According to the abnormal order determining method and the related equipment for the network appointment, when an order of the network appointment, which is sent by a terminal, is received, first order information of the order, equipment information of the terminal, historical second order information related to an account corresponding to the order and user information of a user to which the account belongs are obtained, target parameters are determined according to the first order information, the equipment information, the second order information and the user information, and then the target parameters are input into a detection model to obtain an abnormal result of the first order. In the invention, the target parameters of the input detection model comprise the installation times, the version parameters, the login information, the travel parameters corresponding to the order, the mobile phone number type of the user, the historical order removing times, the called single times, the estimated maximum order placing amount and the like of the application program corresponding to the account number, namely, whether the order is abnormal or not is determined in an all-around manner by combining all relevant information of the order, whether the order is abnormal or not is not determined only by one kind of information, and the determination accuracy of the abnormal order of the network taxi appointment is higher.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a system architecture diagram of an abnormal order determination method for implementing a network taxi appointment according to the present invention;
FIG. 2 is a flowchart illustrating a first embodiment of an abnormal order determination method for online taxi appointment according to the present invention;
FIG. 3 is a flowchart illustrating a second embodiment of the abnormal order determination method for online taxi appointment according to the present invention;
fig. 4 is a detailed flowchart of step S50 in the third embodiment of the method for determining an abnormal order of a network taxi appointment according to the present invention;
fig. 5 is a detailed flowchart of step S60 in the fourth embodiment of the method for determining an abnormal order of a network taxi appointment according to the present invention;
FIG. 6 is a block diagram of an abnormal order determining apparatus for online taxi appointment according to the present invention;
fig. 7 is a schematic diagram of a hardware structure of the abnormal order determining apparatus for online taxi appointment according to the present invention.
With the foregoing drawings in mind, certain embodiments of the disclosure have been shown and described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the disclosure to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
The invention provides an abnormal order determining method for a network taxi appointment, which can be realized by a system architecture diagram shown in figure 1. As shown in fig. 1, the abnormal order specifying device 100 for network appointment is communicatively connected to a plurality of mobile terminals 200. The mobile terminal 200 is loaded with a car booking program. The user can register in the abnormal order determining device 100 of the online car booking through the car booking program which can be loaded on the mobile terminal 200, and the user can log in the car booking program through the registered account number, so that the user can perform the online car booking to the abnormal order determining device 100 of the online car booking through the car booking program of the mobile terminal 200. When receiving an order for a network appointment sent by the mobile terminal 200, the abnormal order determination device 100 for a network appointment acquires current order information, order information of a user account history, device information of the mobile terminal 200, and user information to determine whether the current order is abnormal.
The following describes the technical solutions of the present invention and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present invention will be described below with reference to the accompanying drawings.
Referring to fig. 2, fig. 2 is a first embodiment of the method for determining an abnormal order of a network taxi appointment according to the present invention, and the method for determining an abnormal order of a network taxi appointment includes the following steps:
step S10, when receiving a first order of a network car appointment sent by a first terminal, obtaining first order information of the first order, first device information of the first terminal, historical second order information associated with a first account corresponding to the first order, and first user information of a first user to which the first account belongs.
In this embodiment, the execution subject is an abnormal order specifying device for a network appointment. For convenience of description, the device is hereinafter referred to as an abnormal order determination device of a network reservation car. The user can make a network car appointment to the device through the mobile terminal, namely the mobile terminal sends an order of the network car appointment to the device, the order is defined as a first order, and the user is defined as a first user. The apparatus obtains order information of a first order, the order information being defined as the first order information. The first order information includes a first account of the first user, and the first user adopts the login time of the first account and the travel parameter of the first user, such as the current location of the first user and the destination to which the first user is going.
The apparatus also obtains first device information for the first terminal. The first device information includes a model of the mobile terminal, a custom field of a name of the mobile terminal, and the like. In addition, the first account number is associated with the record of the network car appointment of the first user, so that the device can acquire historical order information associated with the first account number. The historical order information is defined as second order information. The device also requires first user information of a first user to which the user first account belongs. When the first user registers, the first user fills in own information, the device stores the filled information as first user information, and the first user information and the first account are stored in a correlated mode. The first user information includes a mobile phone number and the like.
Step S20, determining a first target parameter according to the first order information, the first device information, the second order information, and the first user information, where the first target parameter includes the installation times, version parameters, first login information, travel parameters corresponding to the first order, the mobile phone number type of the first user, the historical number of times of withdrawing orders, the number of times of calling orders, and the estimated maximum amount of money to be placed.
After the device obtains the first order information, the second order information, the first user information and the first equipment information, the first target parameter can be determined according to the first equipment information, the first order information, the second order information and the first user information. The first target parameters comprise installation times of an application program corresponding to the first account, version parameters, first login information, formation parameters corresponding to the first order, the type of a mobile phone number of the first user, historical order removal times, the number of called single times, the estimated maximum order placing amount of the first user and the like.
Specifically, the mobile terminal (first terminal) has its own intranet IP address, the intranet IP address can be obtained according to the first device information, and the application program can record the number of times that the mobile terminal corresponding to the intranet IP address is installed in the current time period (the current time point is, for example, the current day) according to its own number of times, so as to obtain the number of times of installation of the application program. If the number of times of installation is too many, it can be determined that the suspicion of abnormality of using the intranet IP address to perform network taxi appointment is large, that is, the order may be an abnormal order. The device also obtains a version parameter of the application, which may be a version number of the application. The lower the version of the application is determined from the version parameters, the greater the likelihood that the application will be cracked, and thus the greater the likelihood that an anomalous order will be sent using a cracked application.
The first login information includes a current login time point of the first account, an order placing time of the first order, and the like. The first login information can be directly obtained from the first order information.
The journey parameters comprise a distance between a position of the user when the first order is placed and a starting point in the first order, a distance between the position of the user when the first order is placed and a destination in the first order, a distance between the starting point and the destination in the first order, and the like. The journey parameter can be obtained according to the first order information, namely the first order information comprises the position of the user when placing the order.
The type of the mobile phone number is whether the mobile phone number is a virtual number or not, and whether the mobile phone number is an internet of things number section or not. The type of the mobile phone number is obtained through the first equipment information. If the mobile phone number is a virtual number and the mobile phone number is a networking number section, the suspicion that the first order is an abnormal order is large.
The historical withdrawal count may be determined by the second order information. For example, the apparatus may determine the current historical order withdrawal times of the user, and the apparatus determines the historical order withdrawal times through the current historical order information of the first user. If the number of times of historical withdrawal is too large, the first order is an abnormal order and has a large suspicion.
The estimated maximum ordering amount may be the estimated maximum ordering amount of the user on the current day. The estimated maximum ordering amount is determined according to the maximum amount paid by the first user in the network car booking in the time period. For example, the time period is a whole day, and the maximum amount paid by the first user for making a network appointment on a certain day is 200 yuan, the estimated maximum order amount of the first user on the same day is 200 yuan. The device can superpose the sum of the online taxi appointment paid by the first user on the sum of the first order, and if the superposed sum is larger than 200 yuan, the first order may be an abnormal order.
Step S30, inputting the first target parameter into the detection model to obtain an abnormal result of the first order, where the abnormal result is used to indicate whether the first order is abnormal.
The device stores a detection model. After the first target parameter is determined, the device inputs the first target parameter into the detection model, and the information output by the detection model is the abnormal result of the first order. The abnormal result includes that the first order is abnormal or the first order is normal, that is, the abnormal result is used to indicate whether the first order is abnormal.
And if the abnormal result indicates that the first order is abnormal, adding the first account number to a blacklist, so as to limit the first account number to carry out network car booking. And if the abnormal result indicates that the first order is normal, the device distributes the first order to a terminal corresponding to the driver, so that the driver takes the order to the place where the first user is located and recorded in the first order.
In the technical scheme provided by this embodiment, when an order of a network appointment sent by a terminal is received, first order information of the order, device information of the terminal, historical second order information associated with an account corresponding to the order, and user information of a user to which the account belongs are obtained, a target parameter is determined according to the first order information, the device information, the second order information, and the user information, and then the target parameter is input into a detection model to obtain an abnormal result of the first order. In the invention, the target parameters of the input detection model comprise the installation times, the version parameters, the login information, the travel parameters corresponding to the order, the mobile phone number type of the user, the historical order removing times, the called single times, the estimated maximum order placing amount and the like of the application program corresponding to the account number, namely, whether the order is abnormal or not is determined in an all-around manner by combining all relevant information of the order, whether the order is abnormal or not is not determined only by one kind of information, and the determination accuracy of the abnormal order of the network taxi appointment is higher.
Referring to fig. 3, fig. 3 is a second embodiment of the method for determining an abnormal order of a network appointment according to the present invention, and based on the first embodiment, before step S20, the method further includes:
step S40, obtaining each order data set, and determining a training sample corresponding to the order data set according to the order data set, where the order data set includes a second order, third order information corresponding to the second order, second device information of a second terminal that sends the second order, historical fourth order information associated with a second account corresponding to the second order, and second user information of a second user to which the second account belongs, and the training sample includes the installation times, version parameters, second login information, trip parameters corresponding to the second order, a mobile phone number type, historical withdrawal times, call times, and estimated maximum order placing amount of the second user corresponding to the second account.
In this embodiment, the apparatus needs to train out the detection model according to the training sample. Specifically, the device obtains each order data set. Each order data set comprises a second order, third order information corresponding to the second order, second equipment information of a second terminal sending the second order, historical fourth order information related to a second account corresponding to the second order and second user information of a second user to which the second account belongs. The type of information in the third order information is the same as the type of information in the first order information, the type of information in the fourth order information is the same as the type of information in the second order information, the type of information in the second device information is the same as the type of information in the first device information, and the type of information in the first user information is the same as the type of information in the second user information, which is not described herein again.
The device determines a training sample corresponding to each order data set according to each order data set. The training sample comprises the installation times of the application program corresponding to the second account, the version parameter, the second login information, the journey parameter corresponding to the second order, the mobile phone number type of the second user, the historical order removing times, the called single times and the estimated maximum order amount.
And step S50, training the preset model according to each training sample to obtain the detection model.
The device trains the preset model after obtaining each training sample, and if the convergence value of the preset model is not changed, the trained preset model is stored as the detection model.
In the technical solution provided in this embodiment, the apparatus obtains each order data set, and determines a training sample corresponding to each order data set, thereby obtaining a detection model according to training of each training sample.
Referring to fig. 4, fig. 4 is a third embodiment of the abnormal order determination method for online taxi appointment according to the present invention, and based on the second embodiment, step S50 includes:
step S51, processing each order data set to obtain a second target parameter corresponding to each order data set, where the second target parameter includes the installation times, version parameter, second login information, journey parameter corresponding to the second order, mobile phone number type of the second user, historical times of removing orders, number of called orders, and estimated maximum order placing amount of the application program corresponding to the second account.
After the device obtains the order data sets, the order data sets are processed to obtain second target parameters corresponding to each order data set, and the second target parameters comprise installation times, version parameters, second login information, journey parameters corresponding to a second order, a mobile phone number type of a second user, historical order withdrawing times, called single times and estimated maximum order placing amount of an application program corresponding to a second account.
Specifically, the mobile terminal (second terminal) has its own intranet IP address, the intranet IP address can be obtained according to the second device information, and the application program can record the number of times that the mobile terminal corresponding to the intranet IP address is installed in the current time period (the current time point is, for example, the current day) according to its own number of times, so as to obtain the number of times of installation of the application program. If the number of times of installation is too many, it can be determined that the suspicion of abnormality of using the intranet IP address to perform network taxi appointment is large, that is, the order may be an abnormal order. The device also obtains a version parameter of the application, which may be a version number of the application. The lower the version of the application is determined from the version parameters, the greater the likelihood that the application will be cracked, and thus the greater the likelihood that an anomalous order will be sent using a cracked application.
The second login information includes a current login time point of the second account, an order placing time of the second order, and the like. The second login information can be directly obtained from the third order information.
The journey parameters comprise the distance between the position of the user when the second order is placed and the starting point in the second order, the distance between the position of the user when the second order is placed and the destination in the second order, the distance between the starting point and the destination in the second order and the like. The journey parameter may be obtained according to the fourth order information, that is, the second order information includes a position where the user places an order.
The type of the mobile phone number is whether the mobile phone number is a virtual number or not, and whether the mobile phone number is an internet of things number section or not. And the type of the mobile phone number is acquired through the second equipment information. If the mobile phone number is a virtual number and the mobile phone number is a networking number section, the suspicion that the second order is an abnormal order is large.
The historical withdrawal count may be determined by the fourth order information. For example, the apparatus may determine a current historical order withdrawal number of the second user, and the apparatus determines the historical order withdrawal number through the current historical order information of the second user. If the number of times of historical withdrawal is too large, the second order is an abnormal order and has a large suspicion.
The estimated maximum order amount may be the estimated maximum order amount of the second user on the current day. The estimated maximum ordering amount is determined according to the maximum amount paid by the network car booking of the second user in the time period. For example, if the time period is a whole day and the maximum amount paid by the second user for making a network appointment on a certain day is 200 dollars, the estimated maximum amount paid for ordering by the second user on the same day is 200 dollars. The device can superpose the sum of the network car appointment paid by the second user on the sum of the second order, and if the superposed sum is greater than 200 yuan, the second order may be an abnormal order.
Step S52, according to the third order information in the order data set, setting a label for the second target parameter corresponding to the order data set.
The training sample comprises a second target parameter and a label. The tag may be set according to third order information in the order data set, that is, the device sets a tag for a second target parameter corresponding to the order data set.
The apparatus may display third order information for a technician to set a label for the second target parameter. I.e., the technician determines whether the second order is an abnormal order by the third order information. For example, if the second order is not paid within half an hour and the mobile phone number of the second user is a virtual number, a label marked as 1 may be set for the second target parameter, and the label marked as 1 indicates that the training sample is abnormal order data. And if the second order is a normal order, setting a label marked as 0 for the second target parameter.
Step S53, generating a training sample corresponding to the order data set according to the second target parameter corresponding to the order data set and the label.
After the label of the second target parameter is determined, the device can generate a training sample corresponding to the order data set according to the second target parameter corresponding to the order data set and the label.
In the technical scheme provided by this embodiment, the apparatus processes each order data set to obtain a second target parameter corresponding to the order data set, and then sets a label of the second target parameter according to third order information, so as to generate a training sample for training a detection model according to the second target parameter and the label.
In one embodiment, step S52 includes:
determining whether the second order is a risk order or not according to third order information in the order data set;
when the second order is a risk order, setting a first label for a second target parameter corresponding to the order data set, wherein the first label is used for indicating that the second order is the risk order;
and when the second order is not a risk order, setting a second label for a second target parameter corresponding to the order data set, wherein the second label is used for indicating that the second order is not a risk order.
In this embodiment, the apparatus may set the label for the second target parameter according to the third order information in the order data set.
Specifically, the apparatus determines whether the second order is a risk order according to third order information in the order data set. For example, the apparatus determines whether the second order completes payment within a preset duration, the preset duration being any composite number of values, e.g., the preset duration may be half an hour; if the second order is not paid within the preset time length, the order is a risk order; and if the second order completes payment within the preset time length, the second order is a normal order.
When the device determines that the second order is a risk order, setting a first label for a second target parameter corresponding to the order data set, wherein the first label is used for indicating that the second order is a risk order. And when the second order is determined not to be the risk order, setting a second label for a second target parameter corresponding to the order data set, wherein the second label is used for indicating that the second order is not the risk order.
In this embodiment, the apparatus determines whether the second order is a risk order according to the third order information in the order data set, so as to perform label setting on the second target parameter according to the determination result, without manual labeling, thereby saving the training cost of the detection model.
Referring to fig. 5, fig. 5 is a fourth embodiment of the abnormal order determination method for online taxi appointment according to the present invention, and based on the second embodiment, step S60 includes:
in step S61, a training set and a validation set are obtained from each training sample.
In this embodiment, the apparatus divides the training samples into a training set and a verification set, and the training set and the verification set both include a plurality of training samples.
And step S62, training the preset model according to the training samples in the training set to obtain the model to be determined.
And step S63, inputting each training sample in the verification set into the model to be determined, and obtaining the accuracy rate and the recall rate of the model to be determined.
The device adopts training samples in the training set to train the preset model to obtain the model to be determined. The apparatus needs to check the prediction accuracy of the model to be determined. Therefore, the device inputs each training sample in the verification set into the model to be determined, and the accuracy rate and the recall rate of the model to be determined are obtained. The precision rate is TP/(TP + FP), and the recall rate is TP/(TP + FN); wherein, TP is a sample marked as 1, and the number of samples predicted as 1 by the model to be determined; FP is a sample marked as 0, and the number of samples is predicted as 1 by the model to be determined; samples for which FN is marked 1, the number of samples predicted to be 0 by the model to be determined.
And step S64, when the precision rate is greater than or equal to the first threshold value and the recall rate is greater than or equal to the second threshold value, saving the model to be determined as the detection model.
If the accuracy rate is greater than or equal to the first threshold and the recall rate is greater than or equal to the second threshold, the prediction accuracy of the model to be determined is high, and the model to be determined is saved as the detection model. And if the accuracy rate is smaller than the first threshold and/or the recall rate is smaller than the second threshold, a new training sample is required to be adopted to train the model to be determined again.
In the technical scheme provided by this embodiment, the device divides each training sample into a training set and a verification set, trains a preset model according to the training samples in the training set to obtain a model to be determined, inputs each training model in the verification set to the model to be determined to obtain the accuracy and the recall rate of the model to be determined, and if the accuracy is greater than or equal to a first threshold and the recall rate is greater than or equal to a second threshold, the model to be determined can be stored as a detection model, that is, the training is performed to obtain a detection model with high prediction accuracy.
Referring to fig. 6, the abnormal order specification apparatus 600 for network taxi appointment includes:
the obtaining module 610 is configured to obtain first order information of a first order, first device information of the first terminal, historical second order information associated with a first account corresponding to the first order, and first user information of a first user to which the first account belongs when the first order of the online taxi appointment sent by the first terminal is received;
the determining module 620 is configured to determine a first target parameter according to the first order information, the first device information, the second order information, and the first user information, where the first target parameter includes installation times, version parameters, first login information, a trip parameter corresponding to the first order, a mobile phone number type of the first user, historical times of withdrawing orders, called times, and an estimated maximum order amount;
the input module 630 is configured to input the first target parameter into the detection model to obtain an abnormal result of the first order, where the abnormal result is used to indicate whether the first order is abnormal.
In one embodiment, the abnormal order determining apparatus 600 for online taxi appointment includes:
the obtaining module 610 is configured to obtain each order data set, and determine a training sample corresponding to the order data set according to the order data set, where the order data set includes a second order, third order information corresponding to the second order, second device information of a second terminal that sends the second order, historical fourth order information associated with a second account corresponding to the second order, and second user information of a second user to which the second account belongs, and the training sample includes installation times, version parameters, second login information, a travel parameter corresponding to the second order, a mobile phone number type of the second user, historical withdrawal times, called times, and an estimated maximum order placing amount of the second user;
and the training module is used for training the preset model according to each training sample to obtain the detection model.
In one embodiment, the abnormal order determining apparatus 600 for online taxi appointment includes:
the processing module is used for processing each order data set to obtain a second target parameter corresponding to each order data set, wherein the second target parameter comprises the installation times, the version parameter, second login information, a journey parameter corresponding to a second order, the mobile phone number type of a second user, the historical order withdrawing times, the number of called single times and the estimated maximum order placing amount of the application program corresponding to a second account;
the setting module is used for setting a label for a second target parameter corresponding to the order data set according to third order information in the order data set;
and the generating module is used for generating a training sample corresponding to the order data set according to the second target parameter corresponding to the order data set and the label.
In one embodiment, the abnormal order determining apparatus 600 for online taxi appointment includes:
a determining module 620, configured to determine whether the second order is a risk order according to the third order information in the order data set;
the setting module is used for setting a first label for a second target parameter corresponding to the order data set when the second order is a risk order, wherein the first label is used for indicating that the second order is a risk order;
and the setting module is used for setting a second label for a second target parameter corresponding to the order data set when the second order is not a risk order, wherein the second label is used for indicating that the second order is not a risk order.
In one embodiment, the abnormal order determining apparatus 600 for online taxi appointment includes:
an obtaining module 610, configured to obtain a training set and a verification set according to each training sample;
the training module is used for training a preset model according to the training samples in the training set to obtain a model to be determined;
the input module 630 is configured to input each training sample in the verification set into the model to be determined, so as to obtain an accuracy rate and a recall rate of the model to be determined;
and the determining module 620 is configured to save the model to be determined as the detection model when the accuracy rate is greater than or equal to the first threshold and the recall rate is greater than or equal to the second threshold.
In one embodiment, the abnormal order determining apparatus 600 for online taxi appointment includes:
and the adding module is used for adding the first account to the blacklist when the abnormal result indicates that the first order is abnormal.
Fig. 7 is a hardware configuration diagram illustrating an abnormal order determining apparatus for a network appointment according to an exemplary embodiment.
The abnormal order determination apparatus 700 of the network appointment vehicle may include: a processor 701, such as a CPU, a memory 702, and a transceiver 703. Those skilled in the art will appreciate that the configuration shown in fig. 7 does not constitute a definition of an anomalous order determination facility for a networked appointment, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components. The memory 702 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The transceiver 703 is used for receiving information transmitted from and transmitting information to an external device.
A non-transitory computer-readable storage medium, wherein instructions of the storage medium, when executed by a processor of an abnormal order determination device of a network appointment, enable the abnormal order determination device of the network appointment to perform the abnormal order determination method of the network appointment.
A computer program product comprising a computer program which, when executed by a processor of an abnormal order determining apparatus of a network appointment, enables the abnormal order determining apparatus of the network appointment to execute the abnormal order determining method of the network appointment.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. An abnormal order determining method for a network appointment car is characterized by comprising the following steps:
when a first order of a network taxi appointment sent by a first terminal is received, acquiring first order information of the first order, first equipment information of the first terminal, historical second order information related to a first account corresponding to the first order and first user information of a first user to which the first account belongs;
determining a first target parameter according to the first order information, the first device information, the second order information and the first user information, wherein the first target parameter comprises installation times, version parameters, first login information, travel parameters corresponding to the first order, a mobile phone number type of the first user, historical order withdrawal times, called single times and an estimated maximum order placing amount of the first user;
and inputting the first target parameter into a detection model to obtain an abnormal result of the first order, wherein the abnormal result is used for indicating whether the first order is abnormal or not.
2. The method of claim 1, wherein before the step of inputting the first target parameter into the detection model to obtain the output result, the method further comprises:
obtaining each order data set, and determining a training sample corresponding to the order data set according to the order data set, wherein the order data set comprises a second order, third order information corresponding to the second order, second equipment information of a second terminal sending the second order, historical fourth order information associated with a second account corresponding to the second order, and second user information of a second user to which the second account belongs, and the training sample comprises installation times, version parameters, second login information, travel parameters corresponding to the second order, a mobile phone number type of the second user, historical order withdrawal times, number of call single times, and estimated maximum order placing amount of the second user of an application program corresponding to the second account;
and training a preset model according to each training sample to obtain the detection model.
3. The abnormal order determination method for online taxi appointment according to claim 2, wherein the step of determining the training sample corresponding to the order data set according to the order data set comprises the following steps:
processing each order data set to obtain a second target parameter corresponding to each order data set, wherein the second target parameter comprises the installation times, the version parameter, second login information, a journey parameter corresponding to the second order, the mobile phone number type of the second user, the historical order withdrawing times, the number of called single times and the estimated maximum order placing amount of the application program corresponding to the second account;
setting a label for a second target parameter corresponding to the order data set according to the third order information in the order data set;
and generating a training sample corresponding to the order data set according to the second target parameter corresponding to the order data set and the label.
4. The method according to claim 3, wherein the step of setting a label for a second target parameter corresponding to the order data set according to the third order information in the order data set comprises:
determining whether the second order is a risk order according to the third order information in the order data set;
when the second order is a risk order, setting a first label for a second target parameter corresponding to the order data set, wherein the first label is used for indicating that the second order is a risk order;
and when the second order is not a risk order, setting a second label for a second target parameter corresponding to the order data set, wherein the second label is used for indicating that the second order is not a risk order.
5. The abnormal order determination method for online taxi appointment according to claim 2, wherein the step of training a preset model according to each training sample to obtain the detection model comprises the following steps:
acquiring a training set and a verification set according to each training sample;
training the preset model according to the training samples in the training set to obtain a model to be determined;
inputting each training sample in the verification set into the model to be determined to obtain the accuracy and recall rate of the model to be determined;
and saving the model to be determined as the detection model when the accuracy rate is greater than or equal to a first threshold value and the recall rate is greater than or equal to a second threshold value.
6. The method for determining abnormal orders for online taxi appointment according to any one of claims 1-5, wherein the step of inputting the first target parameter into the detection model to obtain the abnormal result of the first order further comprises:
and when the abnormal result indicates that the first order is abnormal, adding the first account to a blacklist.
7. An abnormal order determining apparatus for a network appointment car, comprising:
the obtaining module is used for obtaining first order information of a first order, first equipment information of the first terminal, historical second order information related to a first account corresponding to the first order and first user information of a first user to which the first account belongs when the first order of the online taxi appointment sent by the first terminal is received;
a determining module, configured to determine a first target parameter according to the first order information, the first device information, the second order information, and the first user information, where the first target parameter includes installation times of an application corresponding to the first account, a version parameter, first login information, a travel parameter corresponding to the first order, a mobile phone number type of the first user, a historical order removal time, a call single time, and an estimated maximum order placing amount;
and the input module is used for inputting the first target parameter into a detection model to obtain an abnormal result of the first order, and the abnormal result is used for indicating whether the first order is abnormal or not.
8. An abnormal order determining apparatus for a network appointment, comprising: a memory and a processor;
the memory is to store program instructions;
the processor is used for calling the program instructions in the memory to execute the abnormal order determination method of the network appointment car according to any one of claims 1-6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program; the computer program, when executed, implements the abnormal order determination method for network appointment according to any one of claims 1-6.
10. A computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method for determining an abnormal order for a networked car appointment according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111616934.4A CN114358873A (en) | 2021-12-27 | 2021-12-27 | Abnormal order determining method for network appointment and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111616934.4A CN114358873A (en) | 2021-12-27 | 2021-12-27 | Abnormal order determining method for network appointment and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114358873A true CN114358873A (en) | 2022-04-15 |
Family
ID=81103571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111616934.4A Pending CN114358873A (en) | 2021-12-27 | 2021-12-27 | Abnormal order determining method for network appointment and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114358873A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117635403A (en) * | 2023-11-08 | 2024-03-01 | 杭州一喂智能科技有限公司 | Abnormal order alarm method, device, electronic equipment and computer readable medium |
-
2021
- 2021-12-27 CN CN202111616934.4A patent/CN114358873A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117635403A (en) * | 2023-11-08 | 2024-03-01 | 杭州一喂智能科技有限公司 | Abnormal order alarm method, device, electronic equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107679674B (en) | OTA platform overseas hotel room type service defect prediction method and system | |
| CN109034660A (en) | Based on the determination method and relevant apparatus of the risk control strategy of prediction model | |
| CN106339222B (en) | A kind of service implementing method and device | |
| KR102377628B1 (en) | Apparatus and Method for Managing Performance about Artificial Intelligent Service | |
| US9307085B1 (en) | System, method, and computer program for predicting at least one reason for a current call received from a customer | |
| CN109636317A (en) | Service control method, device, system and storage medium | |
| CN103959753B (en) | Advising for the sales promotion used based on subscriber | |
| KR101672627B1 (en) | User individual information providing method and system using verifying personal identity | |
| CN114036390A (en) | Scene service recommendation method and device, electronic equipment and storage medium | |
| CN108616668B (en) | Method for monitoring dialing task, electronic device and computer readable storage medium | |
| CN114358873A (en) | Abnormal order determining method for network appointment and related equipment | |
| CN111510908A (en) | Card opening method, device, equipment and medium | |
| CN110471843B (en) | Version test method and device, storage medium and electronic equipment | |
| CN111881314A (en) | Method, device and storage medium for judging authenticity of vehicle maintenance service list | |
| CN108702334A (en) | The method and system of distributed testing for the network configuration for zero rate | |
| CN110191140B (en) | Service processing method and device and server | |
| CN108805332B (en) | Feature evaluation method and device | |
| CN108280024B (en) | Flow distribution strategy testing method and device and electronic equipment | |
| CN110111217A (en) | A kind of Destination Management method, apparatus, block chain node device and storage medium | |
| CN106612263A (en) | Method and device used for processing application access request | |
| CN108647102A (en) | Service request processing method, device and the electronic equipment of heterogeneous system | |
| CN111340635A (en) | Data checking method, equipment, server and readable storage medium | |
| CN110276628B (en) | Order processing method and device | |
| CN104378402A (en) | Method and device for updating user contact information | |
| CN110689237B (en) | Management method for vehicle manager and related products thereof |
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 |