CN111292112A

CN111292112A - Service data prediction method and device, electronic equipment and storage medium

Info

Publication number: CN111292112A
Application number: CN201811498153.8A
Authority: CN
Inventors: 路劲; 郄小虎; 李奘
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2018-12-07
Filing date: 2018-12-07
Publication date: 2020-06-16

Abstract

The application provides a service data prediction method, a service data prediction device, an electronic device and a storage medium, wherein the method comprises the following steps: acquiring characteristic information of a service provider; obtaining the prediction probability of the service provider according to the characteristic information and the prediction model of the service provider; and determining the predicted service times of the service provider in N days according to the prediction probability and the mapping relation between the prediction probability and the predicted service times. The method comprises the steps of training a prediction model according to characteristic information of a service provider sample and actual service times of N days, calculating prediction probability corresponding to the characteristic information of the service provider according to the characteristic information of the service provider and the trained prediction model, and determining the predicted service times of the service provider for N days according to the mapping relation between the prediction probability and the predicted service times. The method integrates various characteristic information of the service provider sample, and performs service data prediction by combining the actual service times of the service provider for N days, thereby effectively improving the accuracy of service data prediction.

Description

Service data prediction method and device, electronic equipment and storage medium

Technical Field

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

Background

With the rapid development of the internet, various service applications are widely popularized due to the capability of better meeting the requirements of users, such as: network car booking service, take-out service and the like. For the service platform, the service times of the service provider in a period of time after the service provider completes the first service can be accurately predicted, and the service platform is favorable for better classified management of the service provider.

In the prior art, the prediction of the number of services of a service provider within a period of time after the service provider completes the first service is roughly evaluated by some apparent behavior characteristics of the service provider on the day of completing the first service.

However, because the prediction methods in the prior art have relatively few bases for being used as judgment references, some potential behavior attributes and crowd attributes of drivers cannot be effectively utilized, and the prediction accuracy is relatively low.

Disclosure of Invention

In view of this, an embodiment of the present application provides a service data prediction method, a service data prediction apparatus, an electronic device, and a storage medium, which are used to solve the problem of less prediction evaluation basis in the prior art, and achieve the effect of improving the accuracy of service data prediction.

In a first aspect, an embodiment of the present application provides a service data prediction method, where the method includes:

acquiring characteristic information of a service provider; obtaining the prediction probability of the service provider according to the feature information and the prediction model of the service provider, wherein the prediction model is obtained by training according to the feature information of a plurality of service provider samples and the N-day actual service times of the service provider samples and is used for indicating the prediction probability corresponding to the feature information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are greater than a first preset threshold, and N is an integer greater than 0; and determining the predicted service times of the service provider in N days according to the predicted probability and the mapping relation between the predicted probability and the predicted service times.

Optionally, before obtaining the prediction probability of the service provider according to the feature information and the prediction model of the service provider, the method further includes: acquiring characteristic information of a plurality of service provider samples and N-day actual service times of the service provider samples; establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the plurality of service provider samples and the N-day actual service times of the service provider samples; and training to obtain a prediction model according to the mapping relation between the characteristic information and the actual service times of N days.

Optionally, training to obtain the prediction model according to the mapping relationship between the feature information and the actual service times of N days includes: dividing the characteristic information of a plurality of service provider samples into first type sample characteristic information and second type sample characteristic information according to the mapping relation between the characteristic information and the N-day actual service times, wherein the N-day actual service times corresponding to the characteristic information in the first type sample characteristic information are greater than or equal to a second preset threshold value, and the N-day actual service times corresponding to the characteristic information in the second type sample characteristic information are less than the second preset threshold value; and training to obtain the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

Optionally, after training and obtaining the prediction model according to the mapping relationship between the feature information and the actual service times of N days, the method further includes: acquiring the prediction probability of each service provider sample by adopting a prediction model; sequencing the plurality of service provider samples according to the prediction probability of each service provider sample; dividing a plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples; acquiring a predicted service frequency corresponding to each gear according to the N-day actual service frequency of the service provider sample in each gear and a preset rule; and acquiring the mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of the service provider sample contained in each gear.

Optionally, the actual service times of the service provider sample in N days are the actual service times of the service provider sample in N days after the service provider sample provides service for the first time; the predicted service times of the service provider in N days are the predicted service times of the service provider in N days after the service provider provides the service for the first time.

Optionally, obtaining the predicted service times corresponding to each gear according to the N-day actual service times of the service provider sample in each gear and a preset rule, includes: obtaining preset ordered target service provider samples from the plurality of service provider samples of each gear; and taking the actual service times of the target service provider samples as the predicted service times corresponding to the gear.

Optionally, before dividing the feature information of the multiple service provider samples into first type sample feature information and second type sample feature information according to a mapping relationship between the feature information and the actual service times of N days, the method further includes: sequencing the N-day actual service times of the plurality of service provider samples to obtain preset sequenced target service provider samples; and taking the actual service times of the target service provider sample in N days as a second preset threshold value.

Optionally, after determining the predicted service times of the service provider in N days according to the prediction probability and the mapping relationship between the prediction probability and the predicted service times, the method further includes: counting and acquiring the actual service times of the service provider in N days; and calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

Optionally, the characteristic information includes one or more of the following items: identity characteristic information, registration characteristic information, service behavior characteristic information, terminal characteristic information and service requester characteristic information in historical service of the service provider.

In a second aspect, an embodiment of the present application provides a service data prediction apparatus, including: the device comprises an acquisition module, a calculation module and a determination module.

The acquisition module is used for acquiring the characteristic information of the service provider; the computing module is used for obtaining the prediction probability of the service provider according to the characteristic information and the prediction model of the service provider, wherein the prediction model is obtained according to the characteristic information of the multiple service provider samples and the N-day actual service times of the service provider samples and used for indicating the prediction probability corresponding to the characteristic information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are greater than a first preset threshold, and N is an integer greater than 0; and the determining module is used for determining the predicted service times of the service provider in N days according to the prediction probability and the mapping relation between the prediction probability and the predicted service times.

Optionally, the apparatus further comprises an establishing module and a training module;

the acquiring module is further configured to acquire feature information of a plurality of the service provider samples and N-day actual service times of the service provider samples;

the establishing module is used for establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the multiple service provider samples and the N-day actual service times of the service provider samples;

and the training module is used for training and obtaining the prediction model according to the mapping relation between the characteristic information and the actual service times of N days.

Optionally, the training module is specifically configured to divide the feature information of the multiple service provider samples into first type sample feature information and second type sample feature information according to a mapping relationship between the feature information and the N-day actual service times, where the N-day actual service times corresponding to the feature information in the first type sample feature information are greater than or equal to a second preset threshold, and the N-day actual service times corresponding to the feature information in the second type sample feature information are smaller than the second preset threshold; and training to obtain the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

Optionally, the calculation module is further configured to obtain a prediction probability of each service provider sample by using a prediction model; sequencing the plurality of service provider samples according to the prediction probability of each service provider sample; dividing the plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples; acquiring a predicted service frequency corresponding to each gear according to the N-day actual service frequency of the service provider sample in each gear and a preset rule; and acquiring the mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of the service provider sample contained in each gear.

Optionally, the computing module is specifically configured to obtain preset ordered target service provider samples from the multiple service provider samples in each gear; and taking the actual service times of the target service provider samples as the predicted service times corresponding to the gear.

Optionally, the apparatus further comprises a determining module;

the obtaining module is further configured to sequence the N-day actual service times of the plurality of service provider samples, and obtain a preset sequenced target service provider sample;

and the determining module is used for taking the actual service times of the target service provider sample in N days as a second preset threshold.

Optionally, the calculation module is further configured to count and obtain actual service times of the service provider in N days; and calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the service data prediction method as provided in the first or second aspect.

In a fourth aspect, the present application provides a storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the service data prediction method as provided in the first aspect or the second aspect.

The service data prediction method provided by the embodiment of the application can adopt the characteristic information of the service provider sample and the actual service times of N days to train the prediction model, calculate the prediction probability corresponding to the characteristic information of the service provider according to the characteristic information of the service provider and the trained prediction model, and determine the predicted service times of the service provider in N days according to the mapping relation between the prediction probability and the predicted service times. The service data prediction is realized by comprehensively evaluating various characteristic information of the service provider sample and combining the actual service times of the service provider for N days, and the accuracy of the service data prediction is effectively improved.

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 is a block diagram of a service data prediction system of some embodiments of the present application;

FIG. 2 illustrates a schematic diagram of exemplary hardware and software components of an electronic device of some embodiments of the present application;

FIG. 3 is a flow chart illustrating a service data prediction method according to an embodiment of the present application;

FIG. 4 is a flow chart illustrating another service data prediction method provided by an embodiment of the present application;

FIG. 5 is a flow chart illustrating a further service data prediction method provided by an embodiment of the present application;

FIG. 6 is a flow chart illustrating another service data prediction method provided by an embodiment of the present application;

FIG. 7 is a flow chart illustrating another service data prediction method provided by an embodiment of the present application;

FIG. 8 is a flow chart illustrating another service data prediction method provided by an embodiment of the present application;

FIG. 9 is a flow chart illustrating another service data prediction method provided by an embodiment of the present application;

fig. 10 is a schematic structural diagram illustrating a service data prediction apparatus according to an embodiment of the present application;

fig. 11 is a schematic structural diagram illustrating another service data prediction apparatus provided in an embodiment of the present application;

fig. 12 is a schematic structural diagram illustrating a further service data prediction apparatus according to an embodiment of the present application.

Detailed Description

In order to make the purpose, 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 should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

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.

To enable those skilled in the art to utilize the present disclosure, the following embodiments are presented in conjunction with a specific application scenario, "network appointment service. It will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the application. Although the present application is described primarily in the context of a network appointment service, it should be understood that this is merely one exemplary embodiment. The present application may be applied to any other scenario. For example, the method and the system can be applied to services such as house co-rental, takeout ordering, express delivery and the like.

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.

FIG. 1 is a block diagram of a service data prediction system of some embodiments of the present application. For example, the service data prediction system may be an internet service platform or the like for providing services such as network appointment, take-out, delivery of express, and the like.

The service data prediction system may include one or more of a server 110, a network 120, a service requester terminal 130, a service provider terminal 140, and a database 150, and the server 110 may include a processor therein that performs instruction operations.

In some embodiments, the server 110 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system). In some embodiments, the server 110 may be local or remote to the terminal. For example, the server 110 may access information and/or data stored in the service requester terminal 130, the service provider terminal 140, or the database 150, or any combination thereof, via the network 120. As another example, the server 110 may be directly connected to at least one of the service requester terminal 130, the service provider terminal 140, and the database 150 to access stored information and/or data. In some embodiments, the server 110 may be implemented on a cloud platform; by way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud (community cloud), a distributed cloud, an inter-cloud, a multi-cloud, and the like, or any combination thereof. In some embodiments, the server 110 may be implemented on an electronic device 200 having one or more of the components shown in FIG. 2 in the present application.

In some embodiments, the server 110 may include a processor. In some embodiments, a processor may include one or more processing cores (e.g., a single-core processor (S) or a multi-core processor (S)). Merely by way of example, a Processor may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an Application Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a reduced Instruction Set computer (reduced Instruction Set computer), a microprocessor, or the like, or any combination thereof.

Network 120 may be used for the exchange of information and/or data. In some embodiments, one or more components in the service data prediction system (e.g., server 110, service requestor terminal 130, service provider terminal 140, and database 150) may send information and/or data to other components. For example, the server 110 may obtain a service request from the service requester terminal 130 via the network 120. In some embodiments, the network 120 may be any type of wired or wireless network, or combination thereof. Merely by way of example, the Network 130 may include a wired Network, a Wireless Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a ZigBee Network, a Near Field Communication (NFC) Network, or the like, or any combination thereof. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired or wireless network access points, such as base stations and/or network switching nodes, through which one or more components of the service data prediction system may connect to network 120 to exchange data and/or information.

Database 150 may store data and/or instructions. In some embodiments, the database 150 may store data obtained from the service requester terminal 130 and/or the service provider terminal 140. In some embodiments, database 150 may store data and/or instructions for the exemplary methods described herein. In some embodiments, database 150 may include mass storage, removable storage, volatile Read-write Memory, or Read-Only Memory (ROM), among others, or any combination thereof. By way of example, mass storage may include magnetic disks, optical disks, solid state drives, and the like; removable memory may include flash drives, floppy disks, optical disks, memory cards, zip disks, tapes, and the like; volatile read-write Memory may include Random Access Memory (RAM); the RAM may include Dynamic RAM (DRAM), Double data Rate Synchronous Dynamic RAM (DDR SDRAM); static RAM (SRAM), Thyristor-Based Random Access Memory (T-RAM), Zero-capacitor RAM (Zero-RAM), and the like. By way of example, ROMs may include Mask Read-Only memories (MROMs), Programmable ROMs (PROMs), Erasable Programmable ROMs (PERROMs), Electrically Erasable Programmable ROMs (EEPROMs), compact disk ROMs (CD-ROMs), digital versatile disks (ROMs), and the like. In some embodiments, database 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, across clouds, multiple clouds, or the like, or any combination thereof.

In some embodiments, a database 150 may be connected to network 120 to communicate with one or more components in a service data prediction system (e.g., server 110, service requestor terminal 130, service provider terminal 140, etc.). One or more components in the service data prediction system may access data or instructions stored in database 150 via network 120. In some embodiments, the database 150 may be directly connected to one or more components in the service data prediction system (e.g., the server 110, the service requestor terminal 130, the service provider terminal 140, etc.); alternatively, in some embodiments, database 150 may also be part of server 110.

Fig. 2 illustrates a schematic diagram of exemplary hardware and software components of an electronic device of some embodiments of the present application.

For example, a processor may be used on the electronic device 200 and to perform the functions herein.

The electronic device 200 may be a general purpose computer or a special purpose computer, both of which may be used to implement the service location acquisition method of the present application. Although only a single computer is shown, for convenience, the functions described herein may be implemented in a distributed fashion across multiple similar platforms to balance processing loads.

For example, the electronic device 200 may include a network port 210 connected to a network, one or more processors 220 for executing program instructions, a communication bus 230, and a different form of storage medium 240, such as a disk, ROM, or RAM, or any combination thereof. Illustratively, the computer platform may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented in accordance with these program instructions. The electronic device 200 also includes an Input/Output (I/O) interface 250 between the computer and other Input/Output devices (e.g., keyboard, display screen).

For ease of illustration, only one processor is depicted in the electronic device 200. However, it should be noted that the electronic device 200 in the present application may also comprise a plurality of processors, and thus the steps performed by one processor described in the present application may also be performed by a plurality of processors in combination or individually. For example, if the processor of the electronic device 200 executes steps a and B, it should be understood that steps a and B may also be executed by two different processors together or separately in one processor. For example, a first processor performs step a and a second processor performs step B, or the first processor and the second processor perform steps a and B together.

For simplicity, the embodiment of the present application only takes an application program of the network appointment service as an example, and the service data prediction method provided by the present invention is exemplified, but in practical applications, the embodiment of the present application does not limit the method.

Fig. 3 shows a schematic flow diagram of a service data prediction method provided by an embodiment of the present application, where an execution subject of the embodiment may be a computer, a server, and the like, and as shown in fig. 3, the service data prediction method provided by the present application includes:

s101, acquiring characteristic information of a service provider.

In the present application, a service provider is an application user that can provide services for a service requester, for example, when a taxi is taken as an example, and the service requester is a passenger, the corresponding service provider can be a driver; taking the order for takeout as an example, when the service requester is a buyer, the corresponding service provider may be a takeout rider, a store merchant, etc.

It should be noted that predicting the number of services provided by the service provider can help the service platform to perform development and planning, thereby improving the service quality for the user to a certain extent. For example, the service platform may issue a corresponding reward to a service provider with a greater number of services to encourage it to complete more services. In addition, the service platform can also distribute more service tasks to service providers which complete services more actively so as to better provide services for service requesters.

Alternatively, when predicting the number of services completed by a service provider over a period of time, the prediction may be made according to the behavior characteristics of the service provider.

Optionally, the server background database may store a user identifier of each service provider and corresponding historical service data (feature information), where the historical service data includes a plurality of data. For example: service data of a service provider and request data of a serviced service requester. The server can search and match historical service data corresponding to the service provider in the database according to the user identification of the service provider, further analyze and process the obtained historical service data of the service provider, and predict the service quantity of the service provider in a period of time. Taking the network car booking service as an example, the prediction can be carried out according to various information such as the number of times that a service provider (a driver) provides services recently, the enthusiasm of receiving the services, the evaluation of the service requester (a passenger) on the services, and the like.

The user identifier may be an account, a name, or a terminal identifier registered by the service provider on the application platform, which is not limited herein.

And S102, obtaining the prediction probability of the service provider according to the characteristic information and the prediction model of the service provider.

The prediction model is obtained by training according to the characteristic information of a plurality of service provider samples and the actual service times of N days of the service provider samples and is used for indicating the prediction probability corresponding to the characteristic information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are larger than a first preset threshold, and N is an integer larger than 0.

After the feature information of the service provider is obtained, the server can calculate the probability that the service times within N days corresponding to the feature information of the service provider are greater than a first preset threshold according to the trained prediction model.

It should be noted that the plurality of service provider samples may be randomly extracted by the server, the number of the samples is not limited, and when there are many samples, the calculation of the obtained prediction probability is relatively more accurate. And training a prediction model by taking the characteristic information of the plurality of extracted service provider samples and the N-day actual service times of the service provider samples as input parameters, and calculating the probability that the service times of the service provider in N days are greater than a first preset threshold value by using the trained model and the characteristic information of the service provider.

S103, determining the predicted service times of the service provider in N days according to the prediction probability and the mapping relation between the prediction probability and the predicted service times.

After the probability (prediction probability) that the service frequency of the service provider is greater than the first preset threshold value within N days is obtained through calculation, the service frequency of the service provider within N days can be predicted according to the mapping relation between the prediction probability and the predicted service frequency.

Alternatively, as shown in table 1 below, the mapping relationship between the prediction probability and the prediction service number may be constructed by creating a table. The server background database stores a mapping relation table of the prediction probability and the prediction service times, when the prediction probability is in the range of A-B, the corresponding prediction service times are all a, when the prediction probability is in the range of B-C, the corresponding prediction service times are all B, when the prediction probability is in the range of C-D, the corresponding prediction service times are all C, and the like. Therefore, after the prediction probability of the service provider is obtained through calculation, the server calls the mapping relation table stored in the background of the server, and searches and obtains the prediction service times corresponding to the service provider.

TABLE 1

Prediction probability	Predicting number of services
		A～B	a
B～C	b
		C～D	c

Alternatively, the mapping relationship between the prediction probability and the prediction service times can be constructed in a mode of setting a label. The mapping relation is that a plurality of prediction probabilities in a prediction probability interval correspond to the same prediction service times, and in addition, one prediction probability corresponds to one prediction service time, and specifically, the mapping relation is not specifically limited so as to be capable of meeting the requirement of relatively accurately predicting the service times.

The service data prediction method provided by the embodiment of the application can adopt the characteristic information of the service provider sample and the actual service times of N days to train the prediction model, calculate the prediction probability corresponding to the characteristic information of the service provider according to the characteristic information of the service provider and the trained prediction model, and determine the predicted service times of the service provider in N days according to the mapping relation between the prediction probability and the predicted service times. By comprehensively evaluating the various characteristic information of the service provider sample and combining the actual service times of the service provider for N days, the service data is predicted, and the accuracy of service data prediction is effectively improved.

Fig. 4 is a schematic flow chart of another service data prediction method provided in an embodiment of the present application, and as shown in fig. 4, before obtaining the prediction probability of the service provider according to the feature information and the prediction model of the service provider, the method further includes:

s201, obtaining characteristic information of a plurality of service provider samples and N-day actual service times of the service provider samples.

Before service data prediction is performed by using the prediction model, the model needs to be trained by using a plurality of sample data so as to meet the condition for performing service data prediction.

Optionally, the server background database stores historical service data of each service provider, the server first randomly extracts a plurality of service provider samples, and according to the terminal identifiers of the plurality of service providers, the server searches the corresponding historical service data in the database, where the historical service data includes the feature information of the plurality of service provider samples and the actual service times of N days.

S202, establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the plurality of service provider samples and the N-day actual service times of the service provider samples.

And after the characteristic information and the N-day actual service times of the plurality of service provider samples are obtained, establishing a mapping relation between the characteristic information and the N-day actual service times through a preset rule.

Optionally, the feature information may include multiple types of feature information, and the obtained feature information of each service provider may include multiple types, for example, the feature information of the service provider includes: A. b, C, A, B, D, D, E, F, etc. When the feature information is A, B, C, the corresponding number of actual services may be 10, when the feature information is A, B, D, the corresponding number of actual services may be 13, and when the feature information is D, E, F, the corresponding number of actual services may be 15. The feature information corresponding to different actual service times may have the same type of feature information, for example, when the actual service times are 10 and 13, the corresponding feature information includes features a and B. The feature types in the feature information are different, and the corresponding actual service times are also different, for example, the feature information is respectively: A. b, C, and D, E, F, the corresponding actual service times are 10 and 15 hours, respectively.

And S203, training and obtaining the prediction model according to the mapping relation between the characteristic information and the actual service times of N days.

After the mapping relation between the characteristic information and the N-day actual service times is established through the preset rule, the prediction model is trained according to the characteristic information of the plurality of service provider samples, the N-day actual service times and the mapping relation between the characteristic information and the N-day actual service times, so that the prediction model has the function of service data prediction.

It should be noted that, when the prediction model is trained, some samples may be extracted from multiple service providers to obtain feature information of multiple service provider samples, and the actual service times of the multiple service provider samples in N days may be obtained by counting the actual service times of the multiple service provider samples in the next N days, or may be obtained by counting the service times of the multiple service providers in N days that have completed service, which is not limited herein.

Fig. 5 is a schematic flow chart of another service data prediction method provided in an embodiment of the present application, and as shown in fig. 5, further, the training of the obtained prediction model according to the mapping relationship between the feature information and the actual service times of N days may include:

s301, dividing the characteristic information of the multiple service provider samples into first type sample characteristic information and second type sample characteristic information according to the mapping relation between the characteristic information and the actual service times of N days.

The N-day actual service times corresponding to the characteristic information in the first type of sample characteristic information are greater than or equal to a second preset threshold, and the N-day actual service times corresponding to the characteristic information in the second type of sample characteristic information are less than the second preset threshold.

First, feature information of the plurality of service provider samples and N-day actual service times are acquired, and the feature information can be classified according to the mapping relationship. The classification aims to match service providers with different characteristic information with the actual service times of N days, so that the characteristic information can be conveniently classified through the actual service times of N days of the service providers.

Optionally, in some embodiments, a second preset threshold may be set, and it is determined that the feature information corresponding to the service provider whose actual service frequency in N days is greater than or equal to the second preset threshold is divided into first type sample feature information (that is, high-value sample feature information), and the feature information corresponding to the service provider whose actual service frequency in N days is less than the second preset threshold is divided into second type sample feature information (that is, low-value sample feature information). I.e. training the first type of samples as positive samples and the second type of samples as negative samples.

Optionally, the N-day actual service times of the obtained multiple service provider samples may be sorted according to the number of times, and the N-day actual service times of one service provider sample may be selected as the second preset threshold according to the distribution of the N-day actual service times of all the service provider samples. For example: the minimum completion criterion of the N-day actual service times may be set, and assuming that the minimum completion criterion is 100, the N-day actual service times of the service providers at the critical point on the minimum completion criterion 100 in the multiple service provider samples in which the N-day actual service times are sorted according to the times are used as a second preset threshold. Such as: the actual service times of the sequenced multiple service provider samples in N days are as follows: 300. 260, 240, 220, 180, 150, 120, 105, 80, 60, 40, etc., the N-day actual service times of the service provider sample with the N-day actual service times of 105 may be used as the second preset threshold, that is, 105 is selected as the second preset threshold. Then, the feature information corresponding to a plurality of service provider samples with the actual service times of N days of the service provider samples being 300, 260, 240, 220, 180, 150, 120, and 105 respectively is divided into the first type sample feature information, that is, the high-value sample feature information.

In addition, the N-day actual service times of the plurality of acquired service provider samples may be directly calculated without sorting, and the average value of the N-day actual service times of the plurality of service provider samples may be used as the second preset threshold. The selection mode of the second preset threshold is not particularly limited, and the feature information can be accurately divided into the preconditions.

S302, training and obtaining the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

After the feature information of the plurality of service provider samples is classified, the classification result is used as a training parameter to train the prediction model, so that the trained prediction model can be used for accurately classifying the input new feature information according to the division rule.

Optionally, in the present application, prediction is mainly performed on future service data of a service provider having high-value feature information. The prediction model can divide input new characteristic information into high-value characteristic information or low-value characteristic information, the service providers corresponding to the high-value characteristic information are relatively high in service completion prediction times, the service platform can adopt corresponding reward policies for the service providers to encourage the service providers to complete the enthusiasm of the service, better service is provided for service requesters, better commercial value is achieved for predicting service data of the service providers corresponding to the high-value characteristic information, and the prediction model has guiding significance on how the service platform is better developed and planned.

Fig. 6 is a schematic flow chart of another service data prediction method provided in the embodiment of the present application, and as shown in fig. 6, further after training and obtaining a prediction model according to a mapping relationship between feature information and actual service times of N days, the method further includes:

s401, obtaining the prediction probability of each service provider sample by adopting a prediction model.

After the prediction model is trained through the mapping relation between the feature information and the actual service times of N days, the probability that the feature information of the service provider sample belongs to the first type sample characteristic information (high-value sample feature information) is calculated by using the trained prediction model.

In addition, the feature information is divided into two types according to the mapping relationship between the feature information and the actual service times of N days, and in this embodiment, the probability that the feature information of the service provider sample is the first type sample feature information is mainly calculated.

Alternatively, the prediction probability may be calculated by calculating a matching degree of the acquired service provider sample feature information and the first type sample feature information. For example, the feature information corresponding to the feature information of the service provider sample for N days is a, and the feature information corresponding to the feature information for N days for a is divided into the first type sample feature information, so that the feature information of the service provider sample matches with the first type sample feature information, and the greater the N days for the service provider sample, the greater the prediction probability that the service provider sample matches with the first type sample feature information is.

S402, sequencing the plurality of service provider samples according to the prediction probability of each service provider sample.

And S403, dividing the plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples.

The prediction probabilities of the plurality of service provider samples obtained by the calculation are different, and the plurality of service provider samples may be sorted according to the magnitude of the prediction probability, or alternatively, sorted according to the order of the prediction probabilities from large to small.

For example: the prediction probability is 1-0.8, the corresponding gear is A gear, the prediction probability is 0.8-0.6, the corresponding gear is B gear, the prediction probability is 0.6-0.4, the corresponding gear is C gear and the like. Assuming that the calculated prediction probability of the service provider sample is 0.9, the service provider sample is divided into a class A, and the calculated prediction probability of the service provider sample is 0.5, the service provider sample is divided into a class C, and so on.

Optionally, the plurality of service provider samples may not be sequenced, and each different prediction probability may be corresponding to a different gear according to a preset algorithm, that is, the n prediction probabilities correspond to the n gears.

It should be noted that the sequencing of the plurality of service provider samples is to divide the plurality of service provider samples into a plurality of gears, and the gears are used as prediction units, so that the prediction efficiency of the service data is higher.

S404, acquiring the predicted service times corresponding to each gear according to the N-day actual service times of the service provider samples in each gear and a preset rule.

After the plurality of service provider samples are divided into the plurality of gears, each gear comprises the plurality of service provider samples, and the predicted service times corresponding to each gear can be calculated according to the actual service times of the plurality of service provider samples in each gear in N days.

Alternatively, the calculation result may be used as the predicted service times of each gear by calculating an average value of the N-day actual service times of the plurality of service provider samples in each gear, or the N-day actual service times of a specific service provider in the plurality of sequenced service provider samples in each gear may be used as the preset service times of the gear. The predicted service times corresponding to each gear are determined, and are not particularly limited herein.

Optionally, the predicted number of services for the plurality of service provider samples in each gear is the same, for example: the predicted service times corresponding to the a-file are 100, and then the predicted service times of each service provider sample in the a-file are all 100.

Optionally, the predicted number of services for each gear shift sample is different when n service provider samples correspond to n gear shifts.

S405, obtaining a mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of the service provider sample contained in each gear.

According to the mapping relation between the prediction probabilities of the multiple service provider samples and the gear positions and the mapping relation between the prediction service times of the gear positions and the gear positions, the mapping relation between the prediction probabilities and the prediction service times can be obtained.

For example: in connection with the above example, the prediction probability of the service provider sample is calculated to be 0.9 by the prediction model, the service provider sample can be divided into a class a according to the mapping relationship between the prediction probability and the class, and the predicted service frequency corresponding to the class a is 100 according to the mapping relationship between the class and the predicted service frequency, so that the service provider sample with the prediction probability of 0.9 can be calculated, and the predicted value of the service frequency is 100.

Further, the actual service number of the service provider sample in N days is the actual service number of the service provider sample in N days after the service is provided for the first time.

The number of actual services of the service provider sample in N days may be the number of actual services of the service provider in the next N days after the service provider completes the first service after the service provider performs registration with the terminal of the service provider, for example, the number of actual services of the service provider sample in N days after the service provider completes the first service is counted. For example, after the first service is completed by the service provider sample, the actual service times for N days may be counted from the second service, and may be counted from any one of the fifth service, the eighth service, and the like, and this is not particularly limited herein.

The predicted service times of the service provider in N days are the predicted service times of the service provider in N days after the service provider provides the service for the first time.

It should be noted that, when the predicted service times of the service provider in N days correspond to the actual service times of the service provider in N days of the sample, and the actual service times of the service provider in N days of the sample is the actual service times in the next N days after the first service is completed, the predicted service times in N days corresponding to the predicted service times in the next N days after the first service is completed are also the predicted service times in the next N days after the first service is completed, and details are not described here again.

Fig. 7 is a schematic flow chart of another service data prediction method provided in an embodiment of the present application, and as shown in fig. 7, further, the obtaining the predicted service times corresponding to each gear according to the N-day actual service times of the service provider sample in each gear and a preset rule includes:

s501, obtaining preset ordered target service provider samples from the multiple service provider samples of each gear.

And S502, taking the actual service times of the target service provider samples as predicted service times corresponding to the gears.

For example, the service provider sample ranked sixty-tenth of the percentile in each gear may be selected as the target service provider sample, and the corresponding actual service number within N days may be used as the predicted service number of the gear.

For example: if the a-file contains 100 service provider samples, the service provider sample ranked sixteenth is the service provider sample ranked sixty percent in the file, and the actual service times within N days of the sixteenth service provider sample are used as the predicted service times of the a-file. Or, if the a-file contains 10 service provider sample samples, the service provider sample ranked as the sixth service provider sample is the service provider sample ranked as sixty-tenth of the file, and the actual service times of the sixth service provider sample within N days are used as the predicted service times of the a-file.

Optionally, in this embodiment, a sixty-tenth percentile sample of the service provider is selected as the target service provider sample, so that the prediction accuracy of the service data prediction model provided by the present solution is relatively accurate. But not limited to this selection manner, a service provider sample close to sixty-tenths of a percentile may be selected as the target service provider sample, or the selection may be performed according to an empirical value, or may be performed according to multiple experiments, or may be performed by a preset algorithm, so as to be conditioned on more accurate service data prediction.

Fig. 8 is a schematic flow chart of another service data prediction method provided in an embodiment of the present application, and as shown in fig. 8, before dividing the feature information of the multiple service provider samples into first-type sample feature information and second-type sample feature information according to a mapping relationship between the feature information and the actual service times of N days, the method further includes:

s601, sequencing the N-day actual service times of the multiple service provider samples, and acquiring preset sequenced target service provider samples.

And S602, taking the actual service times of the target service provider sample in N days as a second preset threshold value.

Optionally, in this embodiment, a service provider sample ranked seventy-fifth of the ranked multiple service provider samples may be selected as a target service provider sample, and an actual service frequency within N days corresponding to the target service provider sample may be used as a second preset threshold.

It should be noted that, in this embodiment, a determination manner of the second preset threshold is the same as that in steps S501 and S502, and a principle of a determination manner of the predicted service times corresponding to each gear is the same, which is not described herein again.

Fig. 9 is a schematic flowchart of another service data prediction method provided in an embodiment of the present application, and as shown in fig. 9, further, after determining the predicted service times of the service provider in N days according to the prediction probability and the mapping relationship between the prediction probability and the predicted service times, the method further includes:

s701, counting the actual service times of the service provider in N days.

S702, calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

Through the embodiment, the number of predicted services of the service provider in N days can be calculated and obtained, in order to verify the accuracy of the service data prediction method provided by the application, in addition, the actual service times of the service provider in N days can be counted, the predicted service times of the N days and the actual service times are compared, and the prediction accuracy of the service data prediction method is calculated.

Optionally, a threshold may be set, and when the predicted service times and the actual service times of the service provider N days are greater than the threshold, the error between the predicted service times and the actual service times of the service provider N days is larger, and the corresponding prediction accuracy is reduced; when the predicted service times and the actual service times of the service provider for N days are smaller than the threshold value, the error between the predicted service times and the actual service times of the service provider for N days is in a normal range, and the corresponding prediction accuracy is increased. The method is used for calculating the accuracy of the service data prediction method for the N days of a plurality of service providers, so that the prediction accuracy of the service data prediction method is obtained.

In addition, a preset error calculation method can be adopted to calculate the prediction accuracy of the service data prediction method. The method for calculating the prediction accuracy is not limited.

It should be noted that after the prediction accuracy of the service data prediction method is calculated and obtained, the accuracy can be analyzed, so that the prediction model in the service data prediction method provided by the application can be optimized and updated better, the prediction model is more perfect, and the accuracy of service data prediction is improved. For example: when the prediction accuracy is low, the prediction model can be continuously updated by adjusting each threshold parameter, the updated model is used for service data prediction, and the updating is repeated and perfected so as to improve the prediction performance of the prediction model.

Further, the characteristic information includes one or more of: identity characteristic information, registration characteristic information, service behavior characteristic information, terminal characteristic information and service requester characteristic information in historical service of the service provider.

It should be noted that the feature information in the foregoing embodiments of the present application includes multiple types, and the prediction accuracy of the service data prediction method can be effectively improved by integrating multiple types of feature information of the service provider. Alternatively, the kind of feature information owned by each service provider may be different, and may include only one or more of a plurality of feature information.

Taking the network car booking service as an example, the service provider may be a "driver", the service may refer to "driver order taking", and the service requester may be a "passenger". The driver's identity information may include: the age, sex, city class, city level, and vehicle supply and demand level. Depending on the age and sex of the driver, the driver's demand for service can be analyzed, for example, men, who are typically between 30-40 years of age, require more service due to family burden issues. According to the city where the driver is located, the city level and the online car appointment supply and demand level of the city, the demand of the passenger for requesting service can be analyzed, and the times that the driver provides service are correspondingly more when the demand is larger.

The registration feature information may include: the driver registers whether the city is at a senior home, the days spent from the filling of the registration data to the completion of the filling, the verification failure times of the registration data, the times of the modification of the registration data, the source of the registration channel of the driver, the days of the activation interval of the driver, whether the driver has no vehicle and is joined, whether the local license plate is available, and the like. When the registered city is consistent with the old city, the driver can be a local driver, the chance of providing service locally for a long time is higher, and if the driver is a foreign driver, the driver can just go on the way, the service is provided for the local driver, and the driver does not move locally for a long time. In addition, for drivers who register more actively filling data, the drivers are more expected to provide services as soon as possible, the demands on the services are more urgent, and the service times are relatively more. Whereas unaffiliated drivers, which may be interest-driven, do not offer much service relatively often.

The service behavior feature information may include: the number of days when a driver receives a service request online, the number of hours when the driver receives the service request online, the mode of the driver receiving the service request online, the longest service request receiving time continuously on a single day, the distribution of service request receiving time periods on the day when the driver completes the first service, the distribution of various service request modes on the day when the driver completes the first service, the accumulated service request receiving time period on the day when the driver completes the first service, the accumulated service completion number on the day when the driver completes the first service, the average service price on the day when the driver completes the first service, the number of times the driver is complained by passengers on the day when the driver completes the first service, the ratio of receiving real-time service requests on the day when the driver completes the first service, the ratio of receiving reservation service requests on the day when the driver completes the first service, whether a change exists on the day when the driver completes the first, The waiting time from the beginning of receiving the service request to the successful receiving of the service request after the driver finishes the first service, the duty ratio of early peak receiving service request on the day when the driver finishes the first service, the duty ratio of late peak receiving service request on the day when the driver finishes the first service, the duty ratio of average peak receiving service request on the day when the driver finishes the first service, the duty ratio of night receiving service request on the day when the driver finishes the first service, and the like. According to the various service behavior characteristic information, the dynamic and the positive of the driver who wants to provide the service can be analyzed systematically, and the service data prediction has a good reference value.

And the terminal characteristic information may include: whether the driver is also provided with other online car appointment clients, the number of times the driver is active in receiving service requests at a plurality of service platforms, and the like. Optionally, the server may obtain information of a plurality of clients installed on the driver terminal according to the driver terminal identifier, so as to obtain whether the driver terminal is further installed with other online car booking clients. If many online car booking clients are installed, the demand for receiving service requests is relatively large, and the possibility of receiving service requests is relatively large for relatively active drivers.

The service requester characteristic information in the history service may include: the number of days a passenger registers for a passenger, the number of services requested at a client, the average price of services requested at a client, the highest price of services requested at a client, the consumer power score at a client, the consumer power city ranking at a client, the number of cities for which services were requested at a client, etc. Generally, the higher the quality of passengers served by a driver during historical service, the more arousing the interest of the driver in providing service. For example: the serviced passengers have more requirements on the request network car booking service, the request service price is higher, and the drivers can have better service experience in the process of providing the service, so that more service amount is expected.

Optionally, the feature information in the present application is not limited to the above-mentioned various feature information, and more feature information of the service provider and the service requester may be specifically referred to for prediction of the service data. According to various characteristic information of a service requester, after service data of the service requester in the future is predicted, the service platform can be helped to better distribute services for a service provider, meanwhile, for the service provider with higher service quality, the service platform can also adopt corresponding incentive policies to encourage the service provider to complete more services, and for the service provider, the service requester and the service platform, the service experience degree can be better improved.

It should be noted that, in this embodiment, the network car booking service is only used as an example, but not limited to, and the network car booking service is also applicable to take-out service, express service, and the like, and the present application is not particularly limited.

The service data prediction method can adopt the feature information of the service provider sample and the actual service times of N days to train a prediction model, calculate the prediction probability corresponding to the feature information of the service provider according to the feature information of the service provider and the trained prediction model, and determine the predicted service times of the service provider in N days according to the mapping relation between the prediction probability and the predicted service times. By comprehensively evaluating the various characteristic information of the service provider sample and combining the actual service times of the service provider for N days, the service data is predicted, and the accuracy of service data prediction is effectively improved. In addition, the prediction service data of the service provider is compared with the actual service data, the prediction accuracy is calculated, and the prediction model is optimized and updated according to the prediction accuracy, so that the service data prediction method is more optimized and has better prediction effect.

Fig. 10 is a schematic structural diagram of a service data prediction apparatus according to an embodiment of the present application, and as shown in fig. 10, the apparatus includes: an obtaining module 810, a calculating module 820, and a determining module 830.

An obtaining module 810, configured to obtain feature information of a service provider.

A calculating module 820, configured to obtain a prediction probability of a service provider according to feature information of the service provider and a prediction model, where the prediction model is obtained by training according to feature information of multiple service provider samples and N-day actual service times of the service provider samples, and is used to indicate the prediction probability corresponding to the feature information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are larger than a first preset threshold, and N is an integer larger than 0.

And the determining module 830 is configured to determine the predicted service times of the service provider in N days according to the prediction probability and the mapping relationship between the prediction probability and the predicted service times.

Fig. 11 shows a schematic structural diagram of another service data prediction apparatus provided in the embodiment of the present application, as shown in fig. 11, further, the apparatus further includes a building module 840 and a training module 850.

The obtaining module 810 is further configured to obtain feature information of a plurality of the service provider samples and N-day actual service times of the service provider samples; the establishing module 840 is used for establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the plurality of service provider samples and the N-day actual service times of the service provider samples; the training module 850 is specifically configured to train and obtain the prediction model according to a mapping relationship between the feature information and the actual service times of N days.

Further, the obtaining module 810 is specifically configured to divide the feature information of the multiple service provider samples into first type sample feature information and second type sample feature information according to a mapping relationship between the feature information and the N-day actual service times, where the N-day actual service times corresponding to the feature information in the first type sample feature information are greater than or equal to a second preset threshold, and the N-day actual service times corresponding to the feature information in the second type sample feature information are smaller than the second preset threshold; and training to obtain the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

Further, the calculating module 820 is further configured to obtain a prediction probability of each service provider sample by using a prediction model; sequencing the plurality of service provider samples according to the prediction probability of each service provider sample; dividing the plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples; acquiring a predicted service frequency corresponding to each gear according to the N-day actual service frequency of the service provider sample in each gear and a preset rule; and acquiring the mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of the service provider sample contained in each gear.

Further, the actual service times of the service provider sample in N days are the actual service times of the service provider sample in N days after the service provider sample provides service for the first time; the predicted service times of the service provider in N days are the predicted service times of the service provider in N days after the service provider provides the service for the first time.

Further, the calculating module 820 is specifically configured to obtain a preset ordered target service provider sample from the multiple service provider samples in each gear; and taking the actual service times of the target service provider samples as the predicted service times corresponding to the gear.

Further, in some embodiments, referring to fig. 10, the apparatus further includes a determining module 830, where the obtaining module 810 is further configured to rank the N-day actual service times of the multiple service provider samples, and obtain a preset ranked target service provider sample; and a determining module 830, configured to use the N-day actual service times of the target service provider sample as a second preset threshold.

Further, the calculating module 820 is further configured to count and obtain the actual service times of the service provider in N days; and calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

The apparatus may be configured to execute the method provided by the method embodiment, and the specific implementation manner and the technical effect are similar and will not be described herein again.

Fig. 12 is a schematic structural diagram illustrating a further service data prediction apparatus according to an embodiment of the present application, and as shown in fig. 12, the apparatus includes: a processor 901 and a memory 902, wherein: the memory 902 is used for storing programs, and the processor 901 calls the programs stored in the memory 902 to execute the above method embodiments. The specific implementation and technical effects are similar, and are not described herein again.

The apparatus may be integrated in a device such as a terminal or a server, and is not limited in this application.

Optionally, the invention also provides a program product, for example a computer-readable storage medium, comprising a program which, when being executed by a processor, is adapted to carry out the above-mentioned method embodiments.

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 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.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for service data prediction, the method comprising:

acquiring characteristic information of a service provider;

obtaining the prediction probability of the service provider according to the feature information of the service provider and a prediction model, wherein the prediction model is obtained according to the feature information of a plurality of service provider samples and the N-day actual service times of the service provider samples and is used for indicating the prediction probability corresponding to the feature information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are greater than a first preset threshold, and N is an integer greater than 0;

and determining the predicted service times of the service provider in N days according to the predicted probability and the mapping relation between the predicted probability and the predicted service times.

2. The method of claim 1, wherein before obtaining the prediction probability of the service provider according to the feature information of the service provider and the prediction model, the method further comprises:

acquiring characteristic information of a plurality of service provider samples and N-day actual service times of the service provider samples;

establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the plurality of service provider samples and the N-day actual service times of the service provider samples;

and training to obtain the prediction model according to the mapping relation between the characteristic information and the actual service times of N days.

3. The method according to claim 2, wherein the training to obtain the prediction model according to the mapping relationship between the feature information and the actual service times of N days comprises:

dividing the characteristic information of a plurality of service provider samples into first type sample characteristic information and second type sample characteristic information according to the mapping relation between the characteristic information and the N-day actual service times, wherein the N-day actual service times corresponding to the characteristic information in the first type sample characteristic information are greater than or equal to a second preset threshold value, and the N-day actual service times corresponding to the characteristic information in the second type sample characteristic information are less than the second preset threshold value;

and training to obtain the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

4. The method according to claim 3, wherein after training and obtaining the prediction model according to the mapping relationship between the feature information and the actual service times of N days, the method further comprises:

acquiring the prediction probability of each service provider sample by adopting the prediction model;

sequencing the plurality of service provider samples according to the prediction probability of each service provider sample;

dividing the plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples;

acquiring the predicted service times corresponding to each gear according to the actual service times of the service provider sample in N days in each gear and a preset rule;

and acquiring the mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of a service provider sample contained in each gear.

5. The method according to any of claims 1-4, wherein the actual number of services of the service provider sample for N days is the actual number of services of the service provider sample for N days after the first provision of services;

the predicted service times of the service provider in N days are the predicted service times of the service provider in N days after the service provider provides service for the first time.

6. The method according to claim 4, wherein the obtaining the predicted service times corresponding to each of the gear positions according to the actual service times of the service provider in each of the gear positions in N days and a preset rule comprises:

obtaining preset ordered target service provider samples from the plurality of service provider samples of each gear;

and taking the actual service times of the target service provider samples as predicted service times corresponding to the gears.

7. The method according to claim 3 or 4, wherein before dividing the feature information of the plurality of service provider samples into the first type sample feature information and the second type sample feature information according to the mapping relationship between the feature information and the actual service times of N days, the method further comprises:

sequencing the N-day actual service times of the plurality of service provider samples to obtain preset sequenced target service provider samples;

and taking the actual service times of the target service provider sample in N days as the second preset threshold value.

8. The method according to claim 3 or 4, wherein said determining the predicted service times of the service provider in N days according to the predicted probability and the mapping relationship between the predicted probability and the predicted service times further comprises:

counting and acquiring the actual service times of the service provider in N days;

and calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

9. The method of claim 1, wherein the feature information comprises one or more of: identity characteristic information, registration characteristic information, service behavior characteristic information, terminal characteristic information and service requester characteristic information in historical service of the service provider.

10. A service data prediction apparatus, the apparatus comprising: the device comprises an acquisition module, a calculation module and a determination module;

the acquisition module is used for acquiring the characteristic information of the service provider;

the calculation module is used for obtaining the prediction probability of the service provider according to the characteristic information of the service provider and a prediction model, wherein the prediction model is obtained according to the characteristic information of a plurality of service provider samples and the N-day actual service times of the service provider samples and is used for indicating the prediction probability corresponding to the characteristic information; the prediction probability is used for indicating the probability that the service times of the service provider in N days are greater than a first preset threshold, and N is an integer greater than 0;

and the determining module is used for determining the predicted service times of the service provider in N days according to the predicted probability and the mapping relation between the predicted probability and the predicted service times.

11. The apparatus of claim 10, further comprising: establishing a module and a training module;

the acquisition module is further used for acquiring the characteristic information of the plurality of service provider samples and the actual service times of the service provider samples in N days;

the establishing module is used for establishing a mapping relation between the characteristic information and the N-day actual service times according to the characteristic information of the plurality of service provider samples and the N-day actual service times of the service provider samples;

12. The apparatus according to claim 11, wherein the training module is specifically configured to divide feature information of a plurality of samples of the service provider into first-class sample feature information and second-class sample feature information according to a mapping relationship between the feature information and N-day actual service times, where the N-day actual service times corresponding to the feature information in the first-class sample feature information is greater than or equal to a second preset threshold, and the N-day actual service times corresponding to the feature information in the second-class sample feature information is less than the second preset threshold; and training to obtain the prediction model according to the first type sample characteristic information and the second type sample characteristic information.

13. The apparatus of claim 12, wherein the computing module is further configured to obtain a prediction probability for each of the service provider samples using the prediction model; sequencing the plurality of service provider samples according to the prediction probability of each service provider sample; dividing the plurality of service provider samples into a plurality of gears according to the sequence of the plurality of service provider samples; acquiring the predicted service times corresponding to each gear according to the actual service times of the service provider sample in N days in each gear and a preset rule; and acquiring the mapping relation between the prediction probability and the prediction service times according to the prediction service times corresponding to each gear and the prediction probability of a service provider sample contained in each gear.

14. The apparatus according to any of claims 10-13, wherein the N-day actual service count of the service provider sample is N-day actual service count of the service provider sample after first providing service;

15. The apparatus according to claim 13, wherein the computing module is specifically configured to obtain a preset ordered target service provider sample from a plurality of service provider samples in each of the gears; and taking the actual service times of the target service provider samples as predicted service times corresponding to the gears.

16. The apparatus of claim 12 or 13, further comprising: a determination module;

and the determining module is used for taking the actual service times of the target service provider sample in N days as the second preset threshold.

17. The apparatus according to claim 12 or 13, wherein the computing module is further configured to statistically obtain the actual service times of the service provider in N days; and calculating and obtaining the prediction accuracy according to the actual service times of the service provider in N days and the predicted service times of the service provider in N days.

18. The apparatus of claim 10, wherein the characteristic information comprises one or more of: identity characteristic information, registration characteristic information, service behavior characteristic information, terminal characteristic information and service requester characteristic information in historical service of the service provider.

19. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the service data prediction method according to any one of claims 1 to 9.

20. A storage medium, characterized in that the storage medium has stored thereon a computer program which, when being executed by a processor, performs the steps of the service data prediction method according to one of claims 1 to 9.