CN111861497A

CN111861497A - Data processing method and device, computer equipment and storage medium

Info

Publication number: CN111861497A
Application number: CN201910301380.5A
Authority: CN
Inventors: 曹利锋; 李奘; 杜龙志; 常智华
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-04-15
Filing date: 2019-04-15
Publication date: 2020-10-30

Abstract

The embodiment of the application provides a data processing method, a data processing device, computer equipment and a storage medium, and relates to the technical field of data processing, wherein the method comprises the following steps: when a target event is detected to occur, acquiring real-time characteristic information associated with the target event; acquiring offline feature information associated with the target event in a latest offline feature updating period; generating a feature snapshot of the target event based on the acquired real-time feature information and the offline feature information; and processing the associated matters of the target event based on the characteristic snapshot of the target event. In the embodiment of the application, the feature information is extracted based on the feature snapshot, so that rich feature information can be extracted from the feature snapshot, the collection cost of the feature information in the processing process is reduced, the collection period of the feature information is shortened, and the data processing efficiency is improved.

Description

Data processing method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a data processing method and apparatus, a computer device, and a storage medium.

Background

With the rapid development of the internet, more and more internet products are used by people, such as a network appointment platform. The passenger can request the car booking service through the operation network car booking platform, and the driver can provide the car booking service through the network car booking platform.

In practice, when a passenger operates the network car booking platform to request car booking service or when a driver provides car booking service through the network car booking platform, the operation of the network car booking platform generally generates corresponding events, and the network car booking platform needs to process the generated events so as to provide better service for the passenger or the driver. At present, a method for processing an event by a network appointment platform is generally as follows, collecting characteristic information corresponding to the same event within a period of time, and then processing the corresponding event based on the collected characteristic information corresponding to the same event.

However, the above processing method has a limited amount of collected feature information, a high cost for collecting feature information, a long collection period, and a low event processing efficiency.

Disclosure of Invention

In view of this, an object of the embodiments of the present application is to provide a data processing method, an apparatus, a computer device, and a storage medium, which can obtain rich feature information, reduce the collection cost of the feature information, shorten the collection period of the feature information, and improve the data processing efficiency.

In a first aspect, an embodiment of the present application provides a data processing method, including:

when a target event is detected to occur, acquiring real-time characteristic information associated with the target event;

acquiring offline feature information associated with the target event in a latest offline feature updating period;

generating a feature snapshot of the target event based on the acquired real-time feature information and the offline feature information;

and processing the associated matters of the target event based on the characteristic snapshot of the target event.

With reference to the first aspect, an embodiment of the present application provides a first possible implementation manner of the first aspect, where the processing, based on the feature snapshot of the target event, the associated item of the target event includes:

training or updating a target model for processing the target event based on the feature snapshot of the target event.

With reference to the first possible implementation manner of the first aspect, an embodiment of the present application provides a second possible implementation manner of the first aspect, where the training or updating a target model for processing the target event based on the feature snapshot of the target event includes:

After target characteristic information influencing a target model for processing the target event is predicted, selecting historical target characteristic information corresponding to the target event in a first historical time period from the characteristic snapshot corresponding to the target event;

and training or updating a target model for processing the target event based on the selected historical target characteristic information to obtain a new target model.

With reference to the first aspect, an embodiment of the present application provides a third possible implementation manner of the first aspect, where the processing, based on the feature snapshot of the target event, the associated item of the target event includes:

acquiring real-time characteristic information and offline characteristic information of the target event in a second historical time period from the characteristic snapshot of the target event;

and determining abnormal information corresponding to the target event based on the acquired real-time characteristic information and the acquired offline characteristic information.

With reference to the first aspect, an embodiment of the present application provides a fourth possible implementation manner of the first aspect, where the processing, based on the feature snapshot of the target event, the associated item of the target event includes:

Acquiring real-time characteristic information and offline characteristic information of the target event in a third history time period from the characteristic snapshot of the target event;

counting the acquired real-time characteristic information and the acquired off-line characteristic information according to a preset counting period aiming at each index associated with the target event, and acquiring the statistical data of the index in each counting period;

and generating a variation trend graph of the target event under the index according to the statistical data of the index under each statistical period.

With reference to the second possible implementation manner of the first aspect, an embodiment of the present application provides a fifth possible implementation manner of the first aspect, where after obtaining a new target model, the method further includes:

evaluating the performance index of the new target model;

and if the performance index of the new target model meets the preset condition, processing the new target event by using the new target model when detecting that the new target event occurs.

With reference to the fifth possible implementation manner of the first aspect, an embodiment of the present application provides a sixth possible implementation manner of the first aspect, where the target event includes an bubbling event; the target model for processing the bubble event includes one or more of the following models: a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model;

The processing the new target event by using the new target model includes:

the method comprises the steps of predicting queuing information of a new bubbling event by using a new queuing model, predicting the terminal arrival time of the new bubbling event by using a new time prediction model, predicting price information of the new bubbling event by using a new dynamic price adjusting model, and predicting discount information of the new bubbling event by using a new dynamic discount model.

With reference to the fifth possible implementation manner of the first aspect, an embodiment of the present application provides a seventh possible implementation manner of the first aspect, where the processing the new target event by using the new target model includes:

and carrying out accountability processing on the new complaint event by using the new accountability model.

With reference to the fifth possible implementation manner of the first aspect, an embodiment of the present application provides an eighth possible implementation manner of the first aspect, where the processing the new target event by using the new target model includes:

and carrying out order separation processing on the new order separation event by using the new order separation model.

With reference to the first aspect, an embodiment of the present application provides a ninth possible implementation manner of the first aspect, where the generating a feature snapshot of the target event based on the obtained real-time feature information and the offline feature information specifically includes:

Acquiring event basic information corresponding to the target event; the event basic information comprises target event identification information, area identification information corresponding to a target event and target event occurrence time information; and, at least one of the following information: service requester identification information, service provider identification information;

acquiring real-time characteristic information corresponding to each identification information in the event basic information and offline characteristic information in a latest offline characteristic updating period;

and generating a feature snapshot of the event based on the target event occurrence time information, the acquired real-time feature information and the acquired offline feature information.

With reference to the sixth possible implementation manner of the first aspect, an embodiment of the present application provides a tenth possible implementation manner of the first aspect, where the bubbling event includes: an event start point and an event end point;

the real-time characteristic information corresponding to the bubbling event comprises one or more of the following characteristic information: current region information of a bubbling event, current region information and current environment information of the event starting point, current region information and current environment information of the event ending point, service provider information around the region where the event starting point is located, service requester information around the region where the event starting point is located, service provider information around the region where the event ending point is located, and service requester information around the region where the event ending point is located; the environment information comprises weather information and road condition information.

With reference to the seventh possible implementation manner of the first aspect, an example of the present application provides an eleventh possible implementation manner of the first aspect, where the real-time feature information corresponding to the complaint event includes one or more of the following feature information: complaint work order information, complaint order information, service provider information, service requester information, detour information, drunkenness information, abnormal stay information, abnormal yaw information, and high-risk driving information.

In a second aspect, an embodiment of the present application further provides a data processing apparatus, including:

the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring real-time characteristic information associated with a target event when the target event is detected to occur;

a second obtaining module, configured to obtain offline feature information associated with the target event in a latest offline feature update period;

the generating module is used for generating a feature snapshot of the target event based on the acquired real-time feature information and the acquired offline feature information;

and the processing module is used for processing the associated matters of the target event based on the characteristic snapshot of the target event.

With reference to the second aspect, an embodiment of the present application provides a first possible implementation manner of the second aspect, where the processing module is specifically configured to:

With reference to the first possible implementation manner of the second aspect, an embodiment of the present application provides a second possible implementation manner of the second aspect, where the processing module is specifically configured to:

With reference to the second aspect, an embodiment of the present application provides a third possible implementation manner of the second aspect, where the processing module is specifically configured to:

With reference to the second aspect, an embodiment of the present application provides a fourth possible implementation manner of the second aspect, where the processing module is specifically configured to:

With reference to the second possible implementation manner of the second aspect, an embodiment of the present application provides a fifth possible implementation manner of the second aspect, where the data processing apparatus further includes:

the evaluation module is used for evaluating the performance index of the new target model;

and the processing module is further used for processing a new target event by using the new target model when the new target event is detected to occur if the performance index of the new target model meets a preset condition.

With reference to the fifth possible implementation manner of the second aspect, the present application provides a sixth possible implementation manner of the second aspect, wherein the target event includes an bubbling event; the target model for processing the bubble event includes one or more of the following models: a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model;

the processing module is specifically used for predicting queuing information of the new bubbling event by using the new queuing model, predicting the end point arrival time of the new bubbling event by using the new time prediction model, predicting price information of the new bubbling event by using the new dynamic price adjusting model, and predicting discount information of the new bubbling event by using the new dynamic discount model.

In combination with the fifth possible implementation manner of the second aspect, the present application provides a seventh possible implementation manner of the second aspect, wherein,

and the processing module is specifically used for performing responsibility judgment processing on the new complaint event by using the new responsibility judgment model.

In combination with the fifth possible implementation manner of the second aspect, the present application provides an eighth possible implementation manner of the second aspect, wherein,

And the processing module is specifically used for carrying out order distribution processing on the new order distribution event by utilizing the new order distribution model.

With reference to the second aspect, an embodiment of the present application provides a ninth possible implementation manner of the second aspect, where the generating module is specifically configured to:

With reference to the sixth possible implementation manner of the second aspect, this application example provides a tenth possible implementation manner of the second aspect, where the bubbling event includes: an event start point and an event end point;

With reference to the seventh possible implementation manner of the second aspect, in this example, an eleventh possible implementation manner of the second aspect is provided, where the real-time feature information corresponding to the complaint event includes one or more of the following feature information: complaint work order information, complaint order information, service provider information, service requester information, detour information, drunkenness information, abnormal stay information, abnormal yaw information, and high-risk driving information.

In a third aspect, an embodiment of the present application further provides a computer device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the computer device is running, the processor executing the machine-readable instructions to perform the steps of the data processing method according to any one of the first aspect.

In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the data processing method according to any one of the first aspect.

According to the data processing method, the data processing device, the computer equipment and the storage medium, when a target event occurs, a feature snapshot of the target event is made by acquiring real-time feature information of the target event and offline feature information associated with the target event in a latest offline feature updating period, and associated matters of the target event are processed based on the made feature snapshot. In the embodiment of the application, the feature information is extracted based on the feature snapshot, so that rich feature information can be extracted from the feature snapshot, the collection cost of the feature information in the processing process is reduced, the collection period of the feature information is shortened, and the data processing efficiency is improved.

Further, when the relevant item of the target event is a model, if the model is to be updated, the prior art needs to collect relevant features based on the running model, so that the model needs to be frequently online and offline, and the stability of the model is poor; in the embodiment of the application, the characteristic snapshot mode is adopted, the characteristics required by the model updating can be directly obtained from the characteristic snapshot of the target event, and the model is updated based on the obtained characteristics, so that the model only needs to be online and offline when being updated, the online and offline times of the model are reduced, and the stability of the model is ensured.

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 shows a flowchart of a data processing method provided in an embodiment of the present application.

Fig. 2 is a flowchart illustrating another data processing method provided in an embodiment of the present application.

Fig. 3 shows a schematic structural diagram of a snapshot platform provided in an embodiment of the present application.

Fig. 4 shows a flowchart of another data processing method provided in the embodiment of the present application.

Fig. 5 is a flowchart illustrating another data processing method provided in an embodiment of the present application.

Fig. 6 shows a flowchart of another data processing method provided in the embodiment of the present application.

Fig. 7 is a flowchart illustrating another data processing method provided in an embodiment of the present application.

Fig. 8 shows a schematic structural diagram of a data processing apparatus according to an embodiment of the present application.

Fig. 9 shows a schematic structural diagram of a computer device provided in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

In consideration of the problems that collected feature information is limited, the collection cost of the feature information is high, the collection period is long, and the processing efficiency of an event is low in a method for processing the event by a network appointment platform, embodiments of the present application provide a data processing method, apparatus, computer device, and storage medium, which can obtain abundant feature information, reduce the collection cost of the feature information, shorten the collection period of the feature information, and improve the processing efficiency of data.

The following describes a data processing method and apparatus provided in the embodiments of the present application with reference to the field of network appointment.

As shown in fig. 1, a data processing method is provided for an embodiment of the present application, where the method may be applied to a server, and the method specifically includes:

s101, when a target event is detected to occur, acquiring real-time characteristic information associated with the target event.

In the embodiment of the application, when a service requester triggers a target event at a first user, the first user sends target event information to a server, and the server detects that the corresponding target event occurs after receiving the target event information from the first user. Or, after the service provider triggers an event at the second user, the second user sends target event information to the server, and the server detects that a corresponding target event occurs after receiving the target event information from the second user. The service request party is a passenger, and the corresponding first user side is a passenger side; the service provider is a driver, and the corresponding second user side is a driver side.

As an embodiment, the target event may be an bubbling event, a sorting event, a complaint event, or a safety event; here, the security event includes, but is not limited to: abnormal stay events, high-risk order events, abnormal relation events and abnormal preference events.

In the field of online booking, a bubbling event: when a passenger uses the network car booking, the passenger needs to input the starting position and the ending position of the network car booking, and after the passenger inputs the ending position of the network car booking at the passenger end, the passenger end sends the ending position of the network car booking to the server, and at the moment, a bubbling event is triggered. Event sorting: when the passenger confirms the order at the passenger terminal, the passenger terminal generates corresponding order information and sends the order information to the server, and at the moment, an order distribution event is triggered. Complaint events: when a passenger triggers complaints on a target order at a passenger end, the passenger end generates complaint information of the target order and sends the complaint information to a server, and at the moment, a complaint event is triggered; or after the driver triggers complaint of the target order at the driver end, the driver end generates complaint information of the target order and sends the complaint information to the server, and at the moment, a complaint event is triggered.

In the embodiment of the application, the target event has real-time characteristic information; the real-time characteristic information corresponding to different target events is different.

S102, obtaining offline feature information associated with the target event in a latest offline feature updating period.

In the embodiment of the application, the storage system corresponding to the server stores the offline feature information of the target event in advance. As an embodiment, the storage system may be a storage medium in a server; or a separate storage system, and the server can access the separate storage system and obtain the offline feature information in the storage system.

The server can periodically update offline feature information associated with the target event in the storage system; for example, the server updates offline feature information associated with the target event every day, or the server updates offline feature information associated with the target event every week. Correspondingly, when a target event occurs, the server acquires offline feature information associated with the target event in a latest offline feature updating period.

For example, the server updates the offline feature information associated with the target event at 24 points per day, and assuming that the target event occurs at 10 am on day 1/2 in 2019, the server obtains the offline feature information associated with the target event, which is updated at 24 points 1/1 in 2019.

In the embodiment of the present application, the offline feature information associated with the target event includes one or more of the following information associated with the target event: the method comprises the steps of service requester offline characteristic information, service provider offline characteristic information, economic status of a target area, Point of Interest (Point) type of the target area and blooming degree of the target area. Here, the service requester offline feature information includes: passenger's name, passenger's age family condition, etc.; the service provider offline feature information includes: the name of the driver, the age of the driver, the family condition of the driver, the work and rest time of the driver, the work age information of the driver, and the like.

S103, generating a feature snapshot of the target event based on the acquired real-time feature information and the acquired offline feature information.

In the embodiment of the application, after acquiring the real-time characteristic information and the offline characteristic information associated with the target event, the server writes the real-time characteristic information and the offline characteristic information associated with the target event into the corresponding storage system together to serve as the characteristic snapshot of the target event.

And S104, processing the related matters of the target event based on the characteristic snapshot of the target event.

In this embodiment of the present application, specifically, processing the associated item of the target event may include: training or updating a target model for processing the target event, playing back scenes of the target event in any historical time period, building a simulation environment based on specific characteristic information corresponding to the target event, and changing trend of indexes related to the target event.

The data processing method provided by the embodiment of the application can extract rich feature information from the feature snapshot based on the feature snapshot, so that the collection cost of the feature information in the processing process is reduced, the collection period of the feature information is shortened, and the data processing efficiency is improved.

In the embodiment of the present application, the real-time characteristic information corresponding to different target events is different, and the following describes the real-time characteristic information corresponding to the target event with reference to some specific target events:

1. for bubble events:

the bubble events include: an event start point and an event end point; the real-time characteristic information corresponding to the bubbling event comprises one or more of the following characteristic information: current region information of a bubbling event, current region information and current environment information of the event starting point, current region information and current environment information of the event ending point, service provider information around the region where the event starting point is located, service requester information around the region where the event starting point is located, service provider information around the region where the event ending point is located, and service requester information around the region where the event ending point is located; the environment information comprises weather information and road condition information.

Wherein, the current area information of the bubbling event comprises: identification (ID) numbers of current areas where bubbling events occur, whether large-scale activities exist in the current areas, road condition information of the current areas, weather information of the current areas, drivers who are in a state of listening to orders in 500 meters around, passengers who are in a state of issuing orders in 500 meters around, and the like; the current area information and the current environment information of the event starting point are as follows: whether large-scale activities exist in the current area, road condition information of the current area, weather information of the current area and the like; the current area information and the current environment information of the event endpoint are as follows: whether large-scale activities exist in the current area, road condition information of the current area, weather information of the current area and the like. The service provider information around the area where the event origin is located includes: the number of drivers in the order listening state 500 meters around the area where the event starting point is located, the total number of drivers 500 meters around the area where the event starting point is located and the like; the service requester information around the area where the event start point is located is: the number of passengers in the order issuing state is 500 meters around the area where the event starting point is located, and the total number of passengers is 500 meters around the area where the event starting point is located; the service provider information around the area where the event endpoint is located is: the number of drivers in the order listening state is 500 meters around the area where the event terminal is located, and the total number of drivers is 500 meters around the area where the event terminal is located; the service requester information around the area where the event endpoint is located is: the number of passengers in the order issuing state is 500 meters around the area of the event terminal, and the total number of passengers is 500 meters around the area of the event terminal.

2. For a billing event:

the real-time characteristic information corresponding to the order issuing event comprises one or more of the following characteristic information: the method comprises the steps of obtaining current area information of an order issuing event, order information, environment information of an order starting point, environment information of an order ending point, service provider information around the area where the order starting point is located, service requester information around the area where the order starting point is located, service provider information around the area where the order ending point is located, service requester information around the area where the order ending point is located, and order issuing reason information.

3. For a complaint event:

the real-time characteristic information corresponding to the complaint event comprises one or more of the following characteristic information: complaint work order information, complaint order information, service provider information, service requester information, detour information, drunkenness information, abnormal stay information, abnormal yaw information, and high-risk driving information.

Further, as shown in fig. 2, in the data processing method provided in the embodiment of the present application, the generating a feature snapshot of the target event based on the obtained real-time feature information and the offline feature information specifically includes:

s201, acquiring event basic information corresponding to the target event; the event basic information comprises target event identification information, area identification information corresponding to a target event and target event occurrence time information; and, at least one of the following information: service requester identification information, service provider identification information.

In an embodiment of the present application, the target event includes: target event identification information, area identification information corresponding to a target event, target event occurrence time information and service requester identification information; alternatively, the target event includes: target event identification information, area identification information corresponding to a target event, target event occurrence time information and service provider identification information; alternatively, the target event includes: target event identification information, area identification information corresponding to a target event, target event occurrence time information, service requester identification information, and service provider identification information.

In one embodiment, the target event is a bubble event, and the bubble event includes a start position and an end position. Accordingly, the above bubbling events include: bubble event ID, start point position ID, end point position, occurrence time information of bubble event, passenger identification information.

S202, acquiring real-time characteristic information corresponding to each identification information in the event basic information and offline characteristic information in a latest offline characteristic updating period.

In this embodiment of the present application, taking a bubbling event as an example, the real-time feature information corresponding to the starting position ID includes: the current road condition information and weather information of the starting point position, whether large-scale activities exist, the number of passengers in the order issuing state around the starting point position, the total number of passengers around the starting point position, the number of drivers in the order listening state around the starting point position and the total number of drivers around the starting point position; passenger information in a departure state around the end position, the total number of passengers around the end position, driver information in an order listening state around the end position, and the total number of drivers around the end position. The surrounding of the starting point refers to a preset range of the starting point, for example, within 500 meters around the starting point; accordingly, the periphery of the end position refers to a predetermined range of the end position, for example, within 500 meters of the end position.

S203, generating a feature snapshot of the event based on the target event occurrence time information, the acquired real-time feature information and the acquired offline feature information.

In the embodiment of the application, the server establishes an association relationship among the target event identification information, the target event occurrence time information and the area identification information corresponding to the target event in advance, and then writes the real-time characteristic information and the offline characteristic information corresponding to the acquired identification information into the corresponding storage system to obtain the characteristic snapshot of the target event. Fig. 3 shows a schematic structural diagram of a snapshot platform provided in an embodiment of the present application.

In the embodiment of the present application, a feature snapshot of a target event generated in the embodiment is described below with reference to a specific application scenario.

Application scenario one

With the rapid development of the internet, the artificial intelligence technology has also been rapidly developed, and machine learning has also been rapidly developed as an important research field in artificial intelligence. The model obtained by machine learning training is widely applied to various scenes, such as a chat robot, a reading understanding model, a voice recognition model, a text recognition model, a dynamic price adjustment model and the like; and the various scenes are also widely applied to the network car booking platform. In this embodiment, the server may train or update a target model for processing the target event based on the feature snapshot of the target event.

As shown in fig. 4, training or updating a target model for processing the target event based on the feature snapshot of the target event includes:

s401, after the target characteristic information affecting a target model for processing the target event is predicted, selecting historical target characteristic information corresponding to the target event in a first historical time period from the characteristic snapshot corresponding to the target event.

As an implementation manner, after receiving target feature information which is input by a worker and can affect a target model for processing a target event, a server obtains a feature snapshot corresponding to the target event, and selects historical target feature information corresponding to the target event in a first historical time period from the feature snapshot.

For example, for an bubbling event, the server predicts that a target feature of 'the number of drivers in a single listening state within 500 meters around the initial position of the bubbling event' is beneficial to a dynamic price adjustment model for processing the bubbling event, and correspondingly, the server obtains feature snapshots corresponding to the bubbling event and selects a plurality of historical target feature information in a half year before the current time from the obtained feature snapshots; the plurality of pieces of historical target feature information are: the number of drivers in the order listening state within 500 meters around the starting position of each bubbling event.

S402, training or updating a target model for processing the target event based on the selected historical target characteristic information to obtain a new target model.

After the plurality of historical target characteristic information is selected, training an original model for processing the bubbling events based on the selected historical target characteristic information to obtain a trained dynamic price-adjusting model, wherein the number of drivers in the order listening state within 500 meters around the initial position of each bubbling event in a half year before the current time is selected by the server; or the server updates the dynamic price adjustment model for processing the bubbling event based on the selected historical target characteristic information to obtain the updated dynamic price adjustment model. The new target model is a trained dynamic price adjustment model or an updated dynamic price adjustment model.

At present, when a model is updated, an online model is needed, feature information corresponding to a target event in a period of time is collected based on the online model, after the feature information is collected, the model is offline and updated based on the collected features, and if the performance index of the updated model meets a preset condition, the updated model is online; and if the performance index of the updated model does not meet the preset condition, the original model is online, and the characteristic or other characteristics are continuously collected to update the model. However, the model in this mode needs to be continuously on and off, so that the stability of the model is poor.

In the embodiment of the application, the characteristic information required by the model updating can be directly acquired from the characteristic snapshot of the target event by adopting the characteristic snapshot mode, and the model is updated based on the acquired characteristic information, so that the model only needs to be online and offline when being updated, the online and offline times of the model are reduced, and the stability of the model is ensured.

As shown in fig. 5, in the embodiment of the present application, after obtaining a new target model, the method further includes:

s501, evaluating the performance index of the new target model.

In the embodiment of the present application, the target model corresponding to the target event is usually a regression model, for example, the dynamic price adjustment model corresponding to the bubbling event is the regression model; wherein, the performance indexes of the regression model comprise: mean Absolute Error (MAE), mean variance (MSE), etc.

S502, if the performance index of the new target model meets the preset condition, when a new target event is detected, the new target event is processed by using the new target model.

As an implementation manner, when the server determines that the average absolute error (MAE) of the trained dynamic price adjustment model is smaller than an error threshold and the average variance (MSE) is smaller than a variance threshold, it determines that the performance index of the trained dynamic price adjustment model meets a preset condition; correspondingly, the server carries out online training on the dynamic price adjusting model so as to process the new target event through the dynamic price adjusting model.

As another embodiment, when the server determines that the average absolute error (MAE) of the updated dynamic price adjustment model is smaller than the error threshold and the average variance (MSE) is smaller than the variance threshold, it determines that the performance index of the updated dynamic price adjustment model meets the preset condition; correspondingly, the server downloads the dynamic price adjusting model before updating and uploads the updated dynamic price adjusting model. And after the server is online with the updated dynamic price adjusting model, when detecting that a new bubbling event occurs, predicting the price information of the new bubbling event by using the online updated dynamic price adjusting model.

Further, in the data processing method provided in the embodiment of the present application, the target event includes an bubbling event; the target model for processing the bubble event includes one or more of the following models: a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model;

processing the new target event using the new target model, including:

In the embodiment of the application, after a passenger triggers a new bubbling event, the passenger generates corresponding bubbling event information and inputs the bubbling event information into a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model respectively, and correspondingly, the queuing information, the terminal arrival time, the price information and the discount information of the bubbling event are output by the models in sequence.

The queuing information is the current queuing position of the new bubble event, for example, the current queuing position of the new bubble event is the 16 th position. The end point arrival time is the estimated end point arrival time of the new bubbling event, for example, the occurrence time of the new bubbling event is 10 am on 1/2/2019, and the corresponding end point arrival time is 11 am on 1/2/2019. The price information is an estimated price corresponding to the new bubble event, for example, 52 yuan. The discount information is an estimated discount price corresponding to the new bubbling event, for example, 5 yuan.

Further, in the data processing method provided in the embodiment of the present application, processing the new target event by using the new target model includes:

In an embodiment of the present application, the target event includes a complaint event, and the target model for processing the complaint event includes a disclaimer model. In one embodiment, when a passenger complains a target order at a passenger terminal, the passenger terminal generates and transmits complaint information of the target order to a server, and the server inputs the complaint information of the target order into a disclaimer model and outputs a disclaimer result.

In this embodiment of the present application, the target event includes a billing event, and the target model for processing the billing event includes a billing model. As an implementation mode, if the passenger confirms the order at the passenger terminal, the passenger terminal generates corresponding order information and sends the corresponding order information to the server, and the server inputs the received order information into the order distribution model and outputs the order distribution result.

Application scenario two

As shown in fig. 6, in the data processing method provided in the embodiment of the present application, processing the related items of the target event based on the feature snapshot of the target event includes:

S601, acquiring real-time characteristic information and offline characteristic information of the target event in a second historical time period from the characteristic snapshot of the target event.

S602, determining abnormal information corresponding to the target event based on the acquired real-time characteristic information and the acquired offline characteristic information.

In step 601 and step 602, as an implementation manner, if a passenger complains about a certain target event, the server obtains a feature snapshot of the target event, obtains real-time feature information and offline feature information of the feature snapshot, and determines abnormal information from the real-time feature information and the offline feature information, so as to determine a solution for the current complaint based on the abnormal information.

The first history time period and the second history time period may be the same or different.

Application scenario three

As shown in fig. 7, in the data processing method provided in the embodiment of the present application, processing the related items of the target event based on the feature snapshot of the target event includes:

s701, acquiring real-time characteristic information and offline characteristic information of the target event in a third history time period from the characteristic snapshot of the target event.

In the embodiment of the present application, the first history time period, the second history time period, and the third history time period may be the same or different.

S702, counting the acquired real-time characteristic information and the acquired off-line characteristic information according to a preset counting period aiming at each index associated with the target event, and acquiring the statistical data of the index in each counting period.

As an embodiment, the server counts the number of drivers in the order listening state around the area a per hour. And the server counts the real-time characteristic information and the off-line characteristic information in the half year before the current time to obtain the number of drivers in the state of listening to the list around the area A in each hour in the half year before the current time.

And S703, generating a change trend graph of the target event under the index according to the statistical data of the index under each statistical period.

In one embodiment, the server generates a change trend graph of the target event under the index according to the number of drivers in the state of listening to the list around the area a in a half year before the current time and corresponding time information, so that the staff can observe the change trend of the bubbling event under the index.

The target event-related index can be used for composition analysis, trend analysis, construction and optimization of a target model for processing the target event, index evaluation, root cause analysis and the like.

According to the data processing method provided by the embodiment of the application, the feature information is extracted based on the feature snapshot, rich feature information can be extracted from the feature snapshot, the collection cost of the feature information in the processing process is reduced, the collection period of the feature information is shortened, and the data processing efficiency is improved.

As shown in fig. 8, a data processing apparatus is provided for an embodiment of the present application, and is configured to process the data processing method, where the apparatus includes:

A first obtaining module 801, configured to obtain real-time feature information associated with a target event when the target event is detected to occur;

a second obtaining module 802, configured to obtain offline feature information associated with the target event in a latest offline feature updating period;

a generating module 803, configured to generate a feature snapshot of the target event based on the obtained real-time feature information and the offline feature information;

a processing module 804, configured to process the associated item of the target event based on the feature snapshot of the target event.

Further, in the data processing apparatus provided in the embodiment of the present application, the processing module 804 is specifically configured to:

Further, the data processing apparatus provided in the embodiment of the present application further includes an evaluation module; wherein the content of the first and second substances,

the processing module 804 is further configured to, if the performance index of the new target model meets a preset condition, process a new target event by using the new target model when the new target event is detected.

Further, in the data processing apparatus provided in the embodiment of the present application, the target event includes an bubbling event; the target model for processing the bubble event includes one or more of the following models: a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model;

the processing module 804 is specifically configured to predict queuing information of a new bubbling event by using the new queuing model, predict end point arrival time of the new bubbling event by using the new time prediction model, predict price information of the new bubbling event by using the new dynamic price adjustment model, and predict discount information of the new bubbling event by using the new dynamic discount model.

Further, in the data processing apparatus provided in the embodiment of the present application, the processing module 804 is specifically configured to perform responsibility judgment processing on a new complaint event by using a new responsibility judgment model.

Further, in the data processing apparatus provided in the embodiment of the present application, the processing module 804 is specifically configured to perform order splitting processing on a new order splitting event by using a new order splitting model.

Further, in the data processing apparatus provided in the embodiment of the present application, the generating module 803 is specifically configured to:

Further, in the data processing apparatus provided in the embodiment of the present application, the bubbling event includes: an event start point and an event end point;

Further, in the data processing apparatus provided in the embodiment of the present application, the real-time characteristic information corresponding to the complaint event includes one or more of the following characteristic information: complaint work order information, complaint order information, service provider information, service requester information, detour information, drunkenness information, abnormal stay information, abnormal yaw information, and high-risk driving information.

According to the data processing device provided by the embodiment of the application, when a target event occurs, a feature snapshot of the target event is made by acquiring real-time feature information of the target event and offline feature information associated with the target event in a latest offline feature updating period, and related matters of the target event are processed based on the made feature snapshot. In the embodiment of the application, the feature information is extracted based on the feature snapshot, so that rich feature information can be extracted from the feature snapshot, the collection cost of the feature information in the processing process is reduced, the collection period of the feature information is shortened, and the data processing efficiency is improved.

As shown in fig. 9, an embodiment of the present application provides a computer device 90, including: a processor 901, a memory 902 and a bus, wherein the memory 902 stores machine-readable instructions executable by the processor 901, when the computer device runs, the processor 901 communicates with the memory 902 through the bus, and the processor 901 executes the machine-readable instructions to execute the steps of the data processing method.

Specifically, the memory 902 and the processor 901 can be general-purpose memories and processors, which are not specifically limited herein, and the data processing method can be executed when the processor 901 executes a computer program stored in the memory 902.

Corresponding to the data processing method, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the data processing method.

It should be noted that in the embodiments of the present application, the term "comprising" is used to indicate the presence of the features stated hereinafter, but does not exclude the addition of further features.

In the embodiments of the present application, the terms "passenger" and "service requestor" are used interchangeably to refer to an individual, entity, or tool that may request or order a service. In the embodiments of the present application, the terms "driver" and "service provider" are used interchangeably to refer to an individual, entity or tool that may provide a service. In the embodiments of the present application, "first client" and "passenger client" may be used interchangeably, and "second client" and "driver client" may be used interchangeably.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this application. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various storage media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

Claims

1. A data processing method, comprising:

2. The data processing method according to claim 1, wherein the processing the related matters of the target event based on the feature snapshot of the target event comprises:

3. The data processing method of claim 2, wherein the training or updating of the target model for processing the target event based on the feature snapshot of the target event comprises:

4. The data processing method according to claim 1, wherein the processing the related item of the target event based on the feature snapshot of the target event further comprises:

5. The data processing method according to claim 1, wherein the processing the related item of the target event based on the feature snapshot of the target event further comprises:

6. The data processing method of claim 3, wherein after obtaining a new object model, the method further comprises:

evaluating the performance index of the new target model;

7. The data processing method of claim 6, wherein the target events comprise one or more of the following events: bubble events, complaint events, triage events;

the target model corresponding to the bubbling event comprises one or more of the following models: a queuing model, a time estimation model, a dynamic price adjustment model and a dynamic discount model;

the target model corresponding to the complaint event comprises a disclaimer model;

the target model corresponding to the order splitting event comprises an order splitting model.

8. The data processing method according to claim 1, wherein the generating a feature snapshot of the target event based on the acquired real-time feature information and the offline feature information specifically includes:

9. The data processing method of claim 7, wherein the bubble event comprises: an event start point and an event end point;

10. A data processing apparatus, comprising:

11. A computer device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the computer apparatus is run, the processor executing the machine-readable instructions to perform the steps of the data processing method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, performs the steps of the data processing method according to any one of claims 1 to 9.