CN115456713A - Unpaid reminding method and system for abnormal order of online taxi appointment - Google Patents
Unpaid reminding method and system for abnormal order of online taxi appointment Download PDFInfo
- Publication number
- CN115456713A CN115456713A CN202211071433.7A CN202211071433A CN115456713A CN 115456713 A CN115456713 A CN 115456713A CN 202211071433 A CN202211071433 A CN 202211071433A CN 115456713 A CN115456713 A CN 115456713A
- Authority
- CN
- China
- Prior art keywords
- unpaid
- reminding
- order
- model
- abnormal
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000012216 screening Methods 0.000 claims abstract description 18
- 238000011161 development Methods 0.000 claims abstract description 13
- 238000011156 evaluation Methods 0.000 claims abstract description 12
- 230000006399 behavior Effects 0.000 claims description 20
- 238000013461 design Methods 0.000 claims description 16
- 238000010801 machine learning Methods 0.000 claims description 13
- 238000012795 verification Methods 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 8
- 238000007477 logistic regression Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 230000000875 corresponding effect Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an unpaid reminding method and system for abnormal orders of online taxi appointment, wherein the method comprises the following steps: a data acquisition step, namely acquiring an abnormal order sample data set; determining a mold entering characteristic, marking an unpaid reminding classification label and a payment label based on an abnormal order sample data set, and designing and developing the mold entering characteristic; training a model, namely obtaining an unpaid reminding model through training and evaluation; and a model application step, namely inputting the abnormal order information needing to be reminded into the unpaid reminding model, outputting the unpaid reminding classification result of the abnormal order, and reminding different users of unpaid payment in a corresponding reminding mode. The system comprises a sample acquisition module, a sample marking module, a feature development module, a feature screening module, a model training module, a model evaluation module and an identification module. The method and the device can identify the unpaid reminding mode of the abnormal order so as to improve the payment rate.
Description
Technical Field
The invention relates to the field of online taxi booking in the Internet, in particular to an unpaid reminding method and system for an online taxi booking abnormal order.
Background
With the combination of the mobile communication technology and the travel service, the online car appointment travel mode on the mobile terminal greatly facilitates the travel requirements of people. The current network car booking service mostly adopts an operation mode of taking a car first and then paying, and a large number of abnormal orders which are not paid for a long time are inevitably generated in the service mode, so that high-amount capital loss of a platform is caused.
For the financial loan industry, due to the fact that the amount of the loan of the customers is large and the quantity of the loan of the customers is small, the overdue collection method is mostly used for reminding the customers in a mode of manual telephone, and the reminding mode is low in efficiency and high in cost. For unpaid orders in the network car booking industry, because the number of unpaid orders is large and the unpaid amount is small, a large amount of manpower and operation cost are consumed by adopting a traditional collection urging mode, the collection urging efficiency is low, and the platform fund loss cannot be timely and effectively recalled.
In the existing common method, an unpaid reminding strategy is generally established according to expert experience or based on user information, order information and the like, and reminding time and intervals of unpaid orders are determined. Whether the order is abnormal or not is not judged, for example, whether the driver has bad or illegal behaviors in the order service process or not is judged, and therefore the passenger does not pay for the order. However, such orders are the ones that are most difficult for passengers to pay, and therefore how to effectively increase the payment rate of abnormal unpaid orders is an urgent problem to be solved in current travel services.
Disclosure of Invention
In view of the above drawbacks of the prior art, an object of the present invention is to provide a method and a system for reminding unpaid orders of online taxi appointment, which are used to solve the problem that the payment rate of the unpaid orders cannot be increased because a proper classification reminding manner is not adopted in the prior art.
In order to achieve the above purpose and other related purposes, the invention provides an unpaid reminding method and system for abnormal orders of online taxi appointment, which take all-round consideration of data such as user behavior information, order information, driver service information and the like to describe model entering characteristics, improve the accuracy of an unpaid reminding model, and based on the classification result of the unpaid reminding model on abnormal orders, adopt corresponding reminding modes to remind different users of unpaid orders so as to improve the payment rate of the abnormal unpaid orders.
In an embodiment of the present invention, a method for reminding unpaid taxi appointment exception orders includes:
a data acquisition step, wherein an abnormal order sample data set is acquired, and the abnormal order sample data set comprises database data of a passenger side APP and a driver side APP;
a step of determining mold entering characteristics, which is to mark an unpaid reminding classification label and a payment label based on the abnormal order sample data set, design and develop the mold entering characteristics, including screening the mold entering characteristics;
training a model, namely obtaining an unpaid reminding model through training and evaluation;
a model application step, inputting the abnormal order information needing to be reminded into the unpaid reminding model, outputting the unpaid reminding classification result of the abnormal order, and adopting a corresponding reminding mode to carry out unpaid reminding on different users according to the unpaid reminding classification result of the abnormal order by the unpaid reminding model
In an embodiment of the present invention, the database data of the passenger side APP and the driver side APP includes one or more of the following information: user historical behavior information, target order attribute information, target order taker and service information.
In an embodiment of the present invention, the step of determining the mold-in characteristic includes:
labeling the order with the abnormal order sample data set according to a specific situation;
designing and developing characteristics according to the related information of the target order;
screening of the input features based on the correlation index
In an embodiment of the present invention, in the step of labeling the abnormal order sample data set according to a specific situation, the method for labeling a label includes: and if a second specific situation occurs, namely the target user has no payment behavior after the target user has no payment reminding, the order is marked as 0 in the corresponding abnormal order sample data set.
In an embodiment of the invention, the mold-entry feature includes one or more of the following: the method comprises the following steps of historical ordering number of a user before order reminding, payment amount of the user, preferential amount of a paid order, unpaid order number exceeding set time, unpaid order amount exceeding set time, unpaid order preferential amount exceeding set time, unpaid order reminding number exceeding set time, user starting number, user login number, estimated target order mileage and actual mileage, estimated target order amount and actual amount, estimated target order duration and actual target order duration, online target order taking driver duration, total driver order taking singular number, complaint order singular number of a driver and customer complaint ratio of the driver taking order.
In an embodiment of the present invention, the screening of the template-entering features is based on one or more of the following related indicators: availability index, interpretability index, information content index, relevance index, and stability index.
In an embodiment of the present invention, the training of the model includes:
training the unpaid reminding model by adopting a machine learning algorithm according to the screened features;
and evaluating the unpaid reminding model, and judging the accuracy of the unpaid reminding model in identifying abnormal orders.
In an embodiment of the present invention, the machine learning algorithm may be GBDT, support vector machine or logistic regression.
In an embodiment of the present invention, in the step of training the model, the abnormal order sample data set is divided into a training set and a verification set according to a preset proportion; training the unpaid reminding model by adopting the machine learning algorithm according to the screened model entry features for the data of the training set; and the data of the verification set is used for verifying the output result of the trained unpaid reminding model so as to judge whether the trained unpaid reminding model meets the preset requirement.
In an embodiment of the invention, in the model applying step, the classification result of the unpaid reminder at least includes:
the order fee is changed into the original estimated fee and the passenger is reminded to pay, so that the driver can detour the order, get rid of the order of the virtual mileage increment and position the order of the virtual mileage increment by the driver, and illegally add additional fees to the driver; and
and the reissued coupon reminds the passenger of paying for the condition that the mileage of the order slightly deviates.
In an embodiment of the present invention, an unpaid reminding system for abnormal orders of online taxi appointment, the system executes the method of claims 1 to 10, and the method comprises: the system comprises an abnormal order sample data set acquisition module, an abnormal order sample data set label marking module, a feature design and development module, a feature screening module, an unpaid reminding model training module, an unpaid reminding model evaluation module and an abnormal order unpaid reminding module;
the abnormal order sample data set acquisition module acquires a plurality of order sample data in a certain time period from database data of a passenger side APP and a driver side APP, the abnormal order sample data set labeling module labels the abnormal order sample data set according to specific situations, the characteristic design and development module designs and develops the model entering characteristics according to data such as user historical behavior information, target order information, driver service information and the like, the characteristic screening module screens the model entering characteristics based on relevant indexes such as availability, interpretability, information quantity, relevance, stability and the like, the unpaid reminding model training module trains the unpaid reminding model according to the screened model entering characteristics by adopting a machine learning algorithm in the characteristic screening module to obtain an unpaid reminding model, the unpaid reminding model evaluation module verifies output results of the trained unpaid reminding model, finally the abnormal order unpaid reminding payment module inputs abnormal order information to be reminded into the unpaid reminding model, outputs unpaid reminding classification results of the abnormal order, and reminds the unpaid reminding user of different payment modes of unpaid reminding the user.
As described above, the unpaid reminding method and system for abnormal orders of online taxi appointment provided by the invention have the following beneficial effects: the model entering characteristics are described in all directions by considering data such as user behavior information, order information, driver service information and the like, and the accuracy rate of the unpaid reminding model is improved. And based on the classification result of the unpaid reminding model to the abnormal order, carrying out unpaid reminding on the abnormal order by adopting a corresponding reminding mode aiming at different users so as to improve the payment rate of the abnormal unpaid order.
Drawings
Fig. 1 is a schematic diagram illustrating steps of the unpaid reminding method for abnormal orders of online taxi appointment according to the present invention.
Fig. 2 is a schematic step diagram illustrating a method for reminding unpaid taxi appointment abnormality orders according to a preferred embodiment of the present invention.
Fig. 3 is a schematic data flow diagram illustrating an unpaid reminding system for abnormal orders of online taxi appointment according to the present invention.
Fig. 4 is a schematic diagram showing a framework of an unpaid reminding system application of the online taxi appointment abnormal order according to the invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. It is also to be understood that the terminology used in the examples herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the present invention. Test methods in which specific conditions are not specified in the following examples are generally carried out under conventional conditions or under conditions recommended by the respective manufacturers.
Please refer to fig. 1 to 4. It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions under which the present invention can be implemented, so that the present invention has no technical significance, and any structural modification, ratio relationship change, or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.
Referring to fig. 1, the present invention provides a method for reminding unpaid taxi appointment exception orders, which comprises the following steps:
s1: a data acquisition step, wherein an abnormal order sample data set is acquired, and the abnormal order sample data set comprises database data of a passenger side APP and a driver side APP;
s2: determining characteristics, namely marking unpaid reminding classification labels and payment labels based on the abnormal order sample data set, and designing and developing the characteristics, including screening the characteristics;
s3: training a model, namely obtaining an unpaid reminding model through training and evaluation;
s4: and a model application step, namely inputting abnormal order information needing to be reminded into the unpaid reminding model, outputting an unpaid reminding classification result of the abnormal order, and carrying out unpaid reminding on different users in a corresponding reminding mode according to the unpaid reminding classification result of the abnormal order by the unpaid reminding model.
In the step of S1 data acquisition, a user drives a car on a platform, and data is generated by using a passenger side APP or an applet. Similarly, the driver can also generate driver data by using the driver-side App order receiving, and the characteristic design and development can be carried out based on the data to obtain the characteristic data.
In the step of determining the mold-entering characteristics in S2, the design and development of the mold-entering characteristics are carried out according to data such as user historical behavior information, target order attribute information, target order taking drivers and service information thereof, and the like, and the mold-entering characteristics are screened. The development characteristics can be designed according to available data based on expert experience, for example, with estimation and actual journey data of unpaid orders, the characteristic of 'difference between estimated journey and actual mileage' can be designed according to the deviation between estimated mileage and actual mileage.
In the step of training the model, a sample data set is divided into a training set and a verification set according to a preset proportion, wherein the training set is used for training the unpaid reminding model, the verification set is used for verifying the output result of the trained unpaid reminding model, and whether the trained unpaid reminding model meets the preset requirements or not is judged. Then, according to the screened features, a machine learning algorithm, such as GBDT (Gradient Boosting Decision Tree), a support vector machine, logistic regression and the like, is adopted to train the unpaid reminding model, and the unpaid reminding model is finally obtained after optimization.
In the step of applying the model S4, the unpaid reminding model is applied to a specific scene, firstly, orders needing unpaid reminding classification are input into the unpaid reminding model, a classification result can be obtained, and the method mainly comprises the following steps: and determining to adopt a corresponding reminding mode to remind different users of unpaid payment according to the classification result so as to improve the payment rate of the abnormal unpaid order.
The method in a preferred embodiment of the invention (as shown in fig. 2) comprises the steps of:
s21: acquiring an order sample data set which comprises user historical behavior information, target order attribute information and driver service information;
s22: marking the sample data set with labels according to a specific situation;
s23: designing and developing characteristics according to target order information, user behavior information and the like;
s24: screening the mold-entering characteristics based on the relevant indexes;
s25: training an unpaid reminding model by adopting a machine learning algorithm according to the screened features;
s26: evaluating the unpaid reminding model and judging the accuracy of the model in identifying abnormal orders;
s27: and carrying out corresponding unpaid reminding according to the classification result of the unpaid reminding model on the order.
In step S21, an order sample set (order sample set) is first obtained, where the order sample set is a number of order samples within a certain time period.
Specifically, the order sample dataset mainly includes one or more of the following information: the system comprises user historical behavior information, target order attribute information, target order taking driver and service information thereof, user behavior data such as user starting, login, order calling, order withdrawing and payment, order data such as order price, mileage and duration, on-line duration of the order taking driver, driver service customer complaint information and the like, wherein the user refers to an order taking passenger.
In step S22, labels are marked on the respective orders in the sample data set according to specific situations.
Specifically, a column of tag columns is added in the sample data set, and if a first specific situation occurs, namely after the target user performs unpaid reminding, the target user completes payment of an unpaid order, the order is marked as 1 in the corresponding sample data set; and if the second specific situation occurs, namely the target user has no payment behavior for the unpaid order after the unpaid reminder is given, marking the order as 0 in the corresponding sample data set.
In step S23, design and development of the model entering feature are performed according to data such as user historical behavior information, target order attribute information, target order taking driver and service information thereof, and the like.
In particular, the developed in-mold features include, but are not limited to: the method comprises the following steps of historical order number of a user before order reminding, order number payment of the user, payment amount of the user, order preference amount paid by the user, order number unpaid more than 30 days, order amount unpaid more than 30 days, order preference amount unpaid more than 30 days, order reminding number unpaid more than 30 days, user starting number, user login number, target order estimated mileage and actual mileage, target order estimated amount and actual amount, target order estimated duration and actual duration, target order taking driver online duration, total driver order taking singular number, driver complaint singular number, driver order taking customer complaint ratio and the like.
In step S24, screening of the template-entering features is performed based on relevant indexes such as availability, interpretability, information amount, correlation, and stability.
Specifically, the availability index: various aspects such as product flow design, user authorization protocol, compliance requirements, model application links and the like need to be comprehensively considered, and whether the characteristic data is continuously available or not is judged.
Interpretability index: the business logic of the features needs to be clear and needs to be consistent with business interpretability.
The information content index is as follows: the information quantity IV of the feature is calculated to evaluate the predictive power of the feature. Generally, the higher the IV, the stronger the predictive power. When the characteristic IV value is larger than a set threshold (generally set to be 0.02), the characteristic has prediction capability and meets the mold-entering requirement.
The correlation index is as follows: pearson's correlation coefficient of the features is calculated to evaluate the correlation between the features. The closer the correlation coefficient of the two characteristics is to 0, the weaker the linear correlation is, and the closer to 1 or-1, the stronger the linear correlation is. When the correlation coefficient between the two features is larger than a set threshold (generally set to 0.6), the features with lower IV values are rejected. The Pearson correlation coefficient is used for comparing the correlation between every two features. For example, one feature works well, and the other feature that is highly correlated with the feature works well, and the two features are very close to each other, and any feature can be used, so that the input of the feature into the model cannot work well. Therefore, model input with less relevant features is guaranteed as much as possible. The IV value is the information quantity IV, and if 1000 features are predicted by the model, they are not necessarily all needed to be applied on line, which increases the development cost and the application difficulty, so a batch of valid and available features are screened first, i.e. some availability interpretable features are screened based on the step S23 (why this is done). The information quantity IV is in particular how much this feature helps the prediction. The higher the information quantity IV, the higher the representation of the availability. For example, the use of the feature accuracy is 90%, and the use of the feature accuracy is 80%. The information quantity value indicates the degree of importance to the prediction result.
The stability index is as follows: a Population Stability Index PSI (Population Stability Index) of the feature is calculated to evaluate the Stability of the feature. When the PSI value is within the set threshold range (generally set to 0-0.1), no change or little change of the characteristics is indicated, and the stability requirement is met. Stability of a feature, the PSI value, means that the feature fluctuates over time. For example, if a feature is sometimes well-behaved and sometimes not well-behaved over a period of time, indicating that the feature is not stable, the availability of the information is not high. After the operation on the characteristic line, the PSI value is calculated to monitor the stability of the single-rate prediction result.
In the step S25, firstly, the sample data set is divided into a training set and a verification set according to a preset proportion, wherein the training set is used for training the unpaid reminder model, the verification set is used for verifying the output result of the trained unpaid reminder model, and whether the trained unpaid reminder model meets the preset requirement is judged; and then, according to the screened features, training the unpaid reminding model by adopting a machine learning algorithm, such as GBDT, a support vector machine, logistic regression and the like, so as to obtain the unpaid reminding model.
In the step S26, the verification set divided by the order sample data set is used to verify the output result of the trained unpaid reminding model, and it is determined whether the recognition accuracy of the unpaid reminding model for various risk users reaches a preset threshold; the accuracy of predicting the order of the order not formed mainly comprises two indexes: the accuracy rate of the order sample identification marked as 1, and the recall rate of the order sample identification marked as 1.
The two indexes of accuracy and recall rate are defined as follows:
precision = TP/(TP + FP), recall = TP/(TP + FN);
wherein, TP: the sample labeled 1, predicted to be 1; FP: the sample labeled 0, predicted to be 1; FN: the sample labeled 1, is predicted to be 0.
In the step S27, according to the classification result of the unpaid reminding model, a corresponding reminding mode is adopted for different users to remind unpaid, for example, a driver detour order can remind a passenger to pay only the estimated amount of the order after the order is priced differently, an additional fee order is added in violation of rules of the driver to remind the passenger to pay after the additional fee is removed from the order, and an abnormal order such as poor service and overtime of the driver can be issued a coupon reminding payment mode for the user, so that the payment rate of the abnormal unpaid order is improved.
Fig. 3 is a schematic data flow diagram of the unpaid reminding system for online taxi appointment abnormal orders according to the invention. The system of the present invention is used for executing the method of the present invention, and includes but is not limited to the following 7 modules, which are respectively:
the sample data set obtaining module 31 is responsible for collecting a plurality of order sample data within a certain time period from database data of the passenger side APP and the driver side APP, namely executing step S21.
The sample data set tagging module 32 tags each order with the sample data set according to a specific situation, that is, performs step S22.
The feature design and development module 33 performs design and development of the model-entering feature according to data such as the user historical behavior information, the target order information, and the driver service information, that is, executes step S23.
The feature screening module 34 performs screening of the entry features based on the related indexes such as availability, interpretability, information amount, correlation, and stability, that is, performs step S24.
The unpaid reminding model training module 35, according to the screened features, obtains the unpaid reminding model by training with a machine learning algorithm, namely, executes the step S25.
And the unpaid reminding model evaluation module 36 is used for verifying the output result of the trained unpaid reminding model. For the evaluation of the unpaid reminder model, step S26 may be performed using AUC (Area Under ROC Curve), KS (Kolmogorov-Smirnov), and other indexes.
And the unpaid reminding module 37 of the abnormal order takes corresponding unpaid reminding for the abnormal order according to the classification result of the unpaid reminding model, namely, step S27 is executed.
The unpaid reminding model is used for classifying different unpaid orders and then reminding different orders after classification by adopting different reminding modes. The concrete classification is as follows:
1. the driver detour order, the driver get rid of the positioning virtual mileage order and the driver illegally adds additional fees: changing the order charge into the original estimated charge and reminding the passenger to pay;
2. the mileage of the order slightly deviates: the additional coupon reminds the passenger of paying;
3. small running order form: the driver case reminds the passenger to pay (the small running order means that the actual end point of the order is inconsistent with the estimated end point, and the actual mileage is far higher than the estimated mileage);
4. passenger malicious order escape: the credit paper prompts the passenger for payment.
Fig. 4 shows an exemplary embodiment of the method of the present invention, and the idea of applying the present invention to solve the problem of the wind control algorithm specifically may include the following 4 points:
1. acquiring an original data domain, which is mainly to acquire passenger end APP and user buried point data, behavior data, equipment data, order data and the like returned by the driver end APP;
2. designing and developing the mold-entering characteristics based on the original data field, including screening the mold-entering characteristics;
3. selecting a proper algorithm, and training a model;
4. the model is applied to a specific scene.
Specifically, assume that a database is in a company's system, and original sample data can be obtained. The occurrence of each order is recorded in the database. When an unpaid reminding method development of a net appointment abnormal order needs to be carried out and an unpaid reminding model needs to be trained, the database can be derived and used as original sample data containing tens of millions of records. This sample database itself contains a large amount of order data that can be used to train the model. The key to training the model is that the more sample data input, the better. A machine (such as a computer) can analyze a large amount of sample data and find out rules by itself to design (learn) a set of unpaid reminding models which cannot be completed by human beings.
The evaluation method of the unpaid reminding model is to select a plurality of pieces of original data from ten million pieces of original data and input the pieces of original data into the model. Comparing the output result of the model with the actual situation of the original data, for example, taking 1000 orders to verify and evaluate according to the above method, it can know the accuracy of the model.
After the unpaid reminding model is determined, real-time unpaid order information is input into the unpaid reminding model for judgment, so that a reminding mode suggestion can be obtained, and several processing situations can be provided: 1. the driver detour order, the driver throwing-positioning virtual mileage order and the driver illegally add additional fees, the order fee is changed into the original estimated fee, and the passenger is reminded to pay. 2. And if the mileage of the order slightly deviates, a coupon is issued to remind the passenger of paying. 3. The actual end point of the order is inconsistent with the estimated end point, the actual mileage is far higher than the estimated mileage, and the driver documents to remind the passenger to pay. 4. The passenger maliciously escapes the order and the credit case reminds the passenger to pay.
In summary, the invention designs and develops features based on user behaviors and order data, and describes the mold entering features of the unpaid reminding model in an all-around manner by using user behavior data such as user starting, login, order calling, order withdrawing and payment, order data such as order price, mileage and duration, and related data of an order service driver, so as to improve the accuracy rate of the unpaid reminding model. And based on the developed characteristics, a machine learning algorithm is adopted to train an unpaid reminding model, an abnormal unpaid order is identified, and a corresponding unpaid reminding mode (such as issuing a coupon, order discount and the like) is adopted to remind a target user, so that the payment rate of the abnormal unpaid order is improved. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Those skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which may be made by those skilled in the art without departing from the spirit and scope of the present invention as defined in the appended claims.
Claims (11)
1. An unpaid reminding method for abnormal orders of online taxi appointment is characterized by comprising the following steps:
a data acquisition step, wherein an abnormal order sample data set is acquired, and the abnormal order sample data set comprises database data of a passenger side APP and a driver side APP; a step of determining mold entering characteristics, which is to mark an unpaid reminding classification label and a payment label based on the abnormal order sample data set, design and develop the mold entering characteristics, including screening the mold entering characteristics;
training a model, namely obtaining an unpaid reminding model through training and evaluation;
and a model application step, namely inputting abnormal order information needing to be reminded into the unpaid reminding model, outputting an unpaid reminding classification result of the abnormal order, and carrying out unpaid reminding on different users in a corresponding reminding mode according to the unpaid reminding classification result of the abnormal order by the unpaid reminding model.
2. The unpaid reminding method for abnormal online taxi appointment orders as claimed in claim 1, wherein: the database data of the passenger side APP and the driver side APP comprises one or more of the following information: user historical behavior information, target order attribute information, target order taker and service information.
3. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 1, wherein: the step of determining the mold-in characteristics comprises the following steps:
labeling the order with the abnormal order sample data set according to a specific situation;
designing and developing characteristics according to the related information of the target order;
and screening the mold-entering characteristics based on the relevant indexes.
4. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 3, wherein: in the step of labeling the abnormal order sample data set according to a specific situation, the method for labeling the label comprises the following steps: and if a second specific situation occurs, namely the target user has no payment behavior after the target user has no payment reminding, the order is marked as 0 in the corresponding abnormal order sample data set.
5. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 3, wherein: the mold-entry features include one or more of: the method comprises the following steps of historical order number of a user before order reminding, order number payment of the user, payment amount of the user, order preference amount paid after the user pays, order number unpaid after exceeding set time, order amount unpaid after exceeding set time, order preference amount unpaid after exceeding set time, order reminding number unpaid after exceeding set time, user starting number, user login number, target order estimated mileage and actual mileage, target order estimated amount and actual amount, target order estimated duration and actual duration, target order taking driver online duration, total driver order taking singular number, driver complaint singular number and driver order taking customer complaint ratio.
6. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 3, wherein: and screening the in-mold characteristics based on one or more of the following related indexes: availability index, interpretability index, information content index, relevance index, and stability index.
7. The unpaid reminding method for abnormal online taxi appointment orders as claimed in claim 1, wherein: the training model step comprises:
training the unpaid reminding model by adopting a machine learning algorithm according to the screened features;
evaluating the unpaid reminding model, and judging the accuracy of the unpaid reminding model in identifying abnormal orders.
8. The unpaid reminding method for abnormal online taxi appointment orders as claimed in claim 7, wherein: the machine learning algorithm may be GBDT, support vector machine or logistic regression.
9. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 7, wherein: in the step of training the model, dividing the abnormal order sample data set into a training set and a verification set according to a preset proportion; training the unpaid reminding model by adopting the machine learning algorithm according to the screened model entry features for the data of the training set; and the data of the verification set is used for verifying the output result of the trained unpaid reminding model so as to judge whether the trained unpaid reminding model meets the preset requirements.
10. The unpaid reminding method for abnormal orders of online taxi appointment as claimed in claim 1, wherein: in the model applying step, the unpaid reminder classification result at least includes:
changing the order fee into the original estimated fee and reminding passengers to pay for the situations of a driver detour order, a driver throwing and positioning virtual mileage increase order and a driver illegally adding additional fees; and
and the reissued coupon reminds the passenger of paying for the condition that the mileage of the order slightly deviates.
11. The utility model provides an unpaid reminder system of unusual order of net car appointment which characterized in that: the system performs the method of claims 1 to 10, comprising: the system comprises an abnormal order sample data set acquisition module, an abnormal order sample data set label marking module, a feature design and development module, a feature screening module, an unpaid reminding model training module, an unpaid reminding model evaluation module and an abnormal order unpaid reminding module;
the abnormal order sample data set acquisition module acquires a plurality of order sample data in a certain time period from database data of a passenger side APP and a driver side APP, the abnormal order sample data set labeling module labels the abnormal order sample data set according to specific situations, the characteristic design and development module designs and develops the model entering characteristics according to data such as user historical behavior information, target order information, driver service information and the like, the characteristic screening module screens the model entering characteristics based on relevant indexes such as availability, interpretability, information quantity, relevance, stability and the like, the unpaid reminding model training module trains the model entering characteristics according to the screened model entering characteristics by adopting a machine learning algorithm to obtain an unpaid reminding model, the unpaid reminding model evaluation module verifies output results of the trained unpaid reminding model, and finally the abnormal order unpaid reminding module inputs abnormal order information to be reminded into the unpaid reminding model to output unpaid reminding classification results of the abnormal orders and reminds the user of unpaid orders in different unpaid reminding modes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211071433.7A CN115456713A (en) | 2022-09-02 | 2022-09-02 | Unpaid reminding method and system for abnormal order of online taxi appointment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211071433.7A CN115456713A (en) | 2022-09-02 | 2022-09-02 | Unpaid reminding method and system for abnormal order of online taxi appointment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115456713A true CN115456713A (en) | 2022-12-09 |
Family
ID=84301212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211071433.7A Pending CN115456713A (en) | 2022-09-02 | 2022-09-02 | Unpaid reminding method and system for abnormal order of online taxi appointment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115456713A (en) |
-
2022
- 2022-09-02 CN CN202211071433.7A patent/CN115456713A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210312567A1 (en) | Automobile Monitoring Systems and Methods for Loss Reserving and Financial Reporting | |
| US11887141B2 (en) | Telematics service detection and action messaging based on machine learning for assisting car sharing platform | |
| US10360601B1 (en) | Method for generating a repair estimate through predictive analytics | |
| CN105894359A (en) | Order pushing method, device and system | |
| Welde et al. | The efficiency of Norwegian road toll companies | |
| CN110827088A (en) | Vehicle cost prediction method and device based on big data and storage medium | |
| CN113674015A (en) | Vehicle rapid damage assessment and residual value assessment method | |
| Meinzer et al. | Can machine learning techniques predict customer dissatisfaction? A feasibility study for the automotive industry. | |
| CN107798522A (en) | A kind of fare pays out method and device | |
| CN110188255B (en) | Power consumer behavior mining method and system based on business data sharing fusion | |
| CN114418748A (en) | Vehicle credit evaluation method, device, equipment and storage medium | |
| CN107463941A (en) | A kind of vehicle owner identification method and device | |
| KR102523050B1 (en) | Method for providing subscription economy based rental price comparison service | |
| Samad et al. | Towards a framework for development of Crowd Logistics: Paving the way for sustainable logistics | |
| Navandar et al. | Users perception based service quality analysis at toll plazas using structural equation modeling | |
| US11710185B1 (en) | Smart estimatics methods and systems | |
| CN107507413A (en) | Vehicle location Forecasting Methodology and device | |
| CN115545276A (en) | Order receiving rate prediction method and system for abnormal orders of online taxi appointment | |
| CN115456713A (en) | Unpaid reminding method and system for abnormal order of online taxi appointment | |
| CN115797084A (en) | Insurance pricing guidance method based on driving behavior of vehicle owner and related equipment thereof | |
| CN113870020A (en) | Overdue risk control method and device | |
| Bilodeau | Intelligent parking technology adoption | |
| CN113570469B (en) | Intelligent vehicle change prediction method for vehicle insurance user | |
| CN116416056B (en) | Credit data processing method and system based on machine learning | |
| Ahmed et al. | Demand prediction for urban air mobility using deep learning |
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 |