CN111353093B

CN111353093B - Problem recommendation method, device, server and readable storage medium

Info

Publication number: CN111353093B
Application number: CN201811585301.XA
Authority: CN
Inventors: 张姣姣
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2018-12-24
Filing date: 2018-12-24
Publication date: 2023-05-23
Anticipated expiration: 2038-12-24
Also published as: CN111353093A

Abstract

The embodiment of the application provides a problem recommending method, a device, a server and a readable storage medium, wherein history service record data are input into a pre-trained collaborative filtering model, a main analoging problem is predicted first, then a target main class problem is selected and continuously input the history service record data into a deep neural network DNN model corresponding to the target main class problem, a specific sub-class recommending problem under the target main class problem is predicted, and finally an output problem recommending menu can simultaneously comprise the main analoging problem and the specific sub-class recommending problem. Therefore, by utilizing the characteristics of prominent problem characteristics and high prediction accuracy of the main referral problem, the main referral problem is predicted first and then the sub-class recommendation problem is predicted, so that the prediction accuracy can be effectively improved, on the other hand, the problem recommendation menu directly comprises the specific sub-class recommendation problem, and a user does not need to repeatedly perform multi-layer menu operation, so that the waiting time of the user is reduced, and the user problem is accurately positioned.

Description

Problem recommendation method, device, server and readable storage medium

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, a server, and a readable storage medium for problem recommendation.

Background

Currently, with the popularization of intelligent terminals, various Applications (APP) for providing life convenience services are also layered, and provide services (such as travel services, takeaway services, etc.) for users to eat and wear. In the process of using the services, users often encounter various problems to be solved and need to consult with a service platform. The existing service platform is difficult to realize accurate personalized recommendation aiming at the problem to be solved by the user, and particularly, the problem of the user in the service using process can not be rapidly positioned when the user consults, so that the user experience is reduced.

Disclosure of Invention

In view of the foregoing, an object of an embodiment of the present application is to provide a problem recommendation method, apparatus, server and readable storage medium, so as to solve or improve the above problem.

According to one aspect of embodiments of the present application, an electronic device is provided that may include one or more storage media and one or more processors in communication with the storage media. One or more storage media store machine-readable instructions executable by a processor. When the electronic device is in operation, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the problem recommendation method.

According to another aspect of the embodiments of the present application, there is provided a problem recommendation method, applied to a server, the method may include:

according to a problem acquisition request sent by a service requester terminal, acquiring historical service record data of the service requester, wherein the historical service record data comprises one or more combinations of service requester portrait features, service order data, service provider portrait features and service space-time information;

inputting the history service record data into a pre-trained collaborative filtering model to obtain the confidence coefficient of each candidate main class problem, and selecting a target main class problem from each candidate main class problem according to the confidence coefficient of each candidate main class problem;

inputting the history service record data into a deep neural network DNN model corresponding to the target main class problem to obtain the confidence coefficient of each candidate sub-class problem under the target main class problem;

and sending a problem recommendation menu to the service requester terminal according to the confidence coefficient of each candidate main class problem and the confidence coefficient of each candidate sub-class problem under the target main class problem, wherein the problem recommendation menu comprises main analogized problems and sub-class recommendation problems under the target main class problem.

In one possible implementation manner, the step of acquiring the historical service record data of the service requester according to the problem acquisition request sent by the service requester terminal may include:

acquiring user information of the service requester from the problem acquisition request, wherein the user information comprises at least one of an incoming call number, a user account number and a user biological feature;

and acquiring the historical service record data of the service requester from a historical service record database stored by the server according to the user information.

In one possible implementation, the service requester profile includes at least basic information of the service requester, service usage frequency, service type distribution and complaint problem distribution, the service provider profile includes at least basic information of the service provider, average service statistics and complaint problem distribution, the service order data includes order statistics of a latest preset number of service orders, the service space-time information includes time information, location information and thermodynamic diagram information when the problem acquisition request is sent, wherein the basic information of the service requester includes age, sex, service level and occupation, the basic information of the service provider includes age, sex and service level, and the average service statistics includes average daily service duration and average daily income.

In a possible implementation manner, the method may further include a step of pre-training a collaborative filtering model, specifically including:

acquiring a main class training sample set, wherein the main class training sample set comprises main class problem service record data of each service requester and candidate main class problems associated with the main class problem service record data, and the problem service record data comprises service requester portraits, service order data, service provider portraits and service space-time information;

and training by adopting a collaborative filtering algorithm according to the training sample set to obtain the collaborative filtering model.

In a possible implementation manner, the method may further include a step of training a DNN model corresponding to each candidate main class problem, and specifically includes:

obtaining a sub-class training sample set of each candidate main class problem, wherein the sub-class training sample set comprises sub-class problem service record data of each service requester and candidate sub-class problems associated with the sub-class problem service record data, and the problem service record data comprises service requester portraits, service order data, service provider portraits and service space-time information;

And training by adopting a DNN algorithm according to each subclass training sample set to obtain a DNN model of each candidate main class problem.

In a possible implementation manner, the step of selecting the target main class question from the candidate main class questions according to the confidence level of the candidate main class questions may include:

generating a sequencing result of each candidate main class problem according to the confidence degree of each candidate main class problem and the descending order of the confidence degree;

selecting at least one candidate main class problem with the forefront ranking as the target main class problem according to the ranking result; or alternatively

And taking the candidate main class problems with the confidence coefficient larger than the preset confidence coefficient as the target main class problems.

In one possible implementation manner, the step of sending a question recommendation menu to the service requester terminal according to the confidence level of each candidate main class question and the confidence level of each candidate sub-class question under the target main class question may include:

selecting a first preset number of candidate main class questions as main analogies according to the confidence degree of each candidate main class question and the descending order of the confidence degree;

selecting a second preset number of candidate sub-class questions as sub-class recommendation questions according to the confidence level of each candidate sub-class question under the target main class question and the descending order of the confidence level;

And generating a problem recommendation menu according to the main analoging problem and the sub-class recommendation problem under the target main class problem, and sending the problem recommendation menu to the service requester terminal.

according to the confidence coefficient of each candidate main class problem, taking the candidate main class problem with the confidence coefficient larger than the first confidence coefficient as a main analoging problem;

according to the confidence coefficient of each candidate sub-class problem under the target main class problem, taking the candidate sub-class problem with the confidence coefficient larger than the second confidence coefficient as a sub-class recommendation problem;

In one possible implementation manner, the step of sending a question recommendation menu to the service requester terminal may include:

and sending the voice data of the main analoging problem and the sub-class recommending problem under the target main class problem included in the problem recommending menu to the service requester terminal, so that the service requester terminal plays the voice data of the main analoging problem and the sub-class recommending problem under the target main class problem to the service requester.

and sending the main analogies and the sub-category recommendation questions under the target main category questions included in the question recommendation menu to the service requester terminal, so that the service requester terminal displays the main analogies and the sub-category recommendation questions under the target main category questions to a user.

In a possible implementation manner, after the step of sending a question recommendation menu to the service requester terminal according to the confidence level of each candidate main class question and the confidence level of each candidate sub-class question under the target main class question, the method may further include:

acquiring a target recommendation problem selected by the service requester terminal from the main analogized problem and the sub-class recommendation problem under the target main class problem;

judging whether the target recommendation problem is a main analoging problem or not;

if the target recommendation problem is a main analoging problem, inputting the history service record data into a DNN model corresponding to the main analoging problem to obtain the confidence coefficient of each candidate sub-class problem under the main analoging problem;

Transmitting sub-class recommendation questions under the main analoging questions to the service requester terminal according to the confidence degrees of the candidate sub-class questions;

and if the target recommended problem is a sub-class recommended problem, processing the sub-class recommended problem according to a problem processing strategy of the sub-class recommended problem.

According to another aspect of the embodiments of the present application, there is provided a problem recommendation apparatus applied to a server, the apparatus including:

the acquisition module can be used for acquiring historical service record data of the service requester according to a problem acquisition request sent by the service requester terminal, wherein the historical service record data comprises one or more combinations of service requester portrait features, service order data, service provider portrait features and service space-time information;

the first input module can be used for inputting the history service record data into a pre-trained collaborative filtering model to obtain the confidence coefficient of each candidate main class problem, and selecting a target main class problem from each candidate main class problem according to the confidence coefficient of each candidate main class problem;

the second input module can be used for inputting the history service record data into a deep neural network DNN model corresponding to the target main class problem to obtain the confidence coefficient of each candidate sub-class problem under the target main class problem;

The menu sending module may be configured to send a question recommendation menu to the service requester terminal according to the confidence degrees of the candidate main class questions and the confidence degrees of the candidate sub-class questions under the target main class questions, where the question recommendation menu includes a main analogized question and a sub-class recommendation question under the target main class questions.

According to another aspect of the embodiments of the present application, there is provided a readable storage medium having stored thereon a computer program which, when executed by a processor, can perform the steps of the problem recommendation method described above.

Based on any one of the above aspects, the embodiments of the present application provide a method, an apparatus, a server, and a readable storage medium for problem recommendation, by inputting historical service record data into a pre-trained collaborative filtering model, predicting a main analoging problem first, then selecting a target main class problem, continuously inputting the historical service record data into a deep neural network DNN model corresponding to the target main class problem, predicting a specific sub-class recommendation problem under the target main class problem, and finally outputting a problem recommendation menu that may include the main analoging problem and the specific sub-class recommendation problem at the same time. Therefore, by utilizing the characteristics of prominent problem characteristics and high prediction accuracy of the main referral problem, the main referral problem is predicted first and then the sub-class recommendation problem is predicted, so that the prediction accuracy can be effectively improved, on the other hand, the problem recommendation menu directly comprises the specific sub-class recommendation problem, and a user does not need to repeatedly perform multi-layer menu operation, so that the waiting time of the user is reduced, and the user problem is accurately positioned.

The foregoing objects, features and advantages of embodiments of the present application will be more readily apparent from the following detailed description of the embodiments taken in conjunction with the accompanying drawings.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 shows an interactive schematic block diagram of a problem recommendation system provided by an embodiment of the present application;

FIG. 2 illustrates a schematic diagram of exemplary hardware and software components of an electronic device that may implement the server, service requester terminal, service provider terminal of FIG. 1, as provided by embodiments of the present application;

FIG. 3 is a schematic flow chart of a problem recommendation method according to an embodiment of the present disclosure;

FIG. 4 is a functional block diagram of a problem recommendation device according to an embodiment of the present application;

FIG. 5 is a block diagram of another functional module of the problem recommendation device according to the embodiment of the present application;

Fig. 6 is a block diagram of another functional module of the problem recommendation device according to the embodiment of the present application.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it should be understood that the accompanying drawings in the present application are only for the purpose of illustration and description, and are not intended to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this application, illustrates operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to the flow diagrams and one or more operations may be removed from the flow diagrams as directed by those skilled in the art.

In addition, the described embodiments are only some, but not all, of the embodiments of the present application. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, are intended to be within the scope of the present application.

In order to enable those skilled in the art to use the present disclosure, the following embodiments are presented in connection with a specific application scenario "network about car scenario". It will be apparent to those having ordinary skill 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 present application. Although the present application is primarily described in terms of a "net jockey scenario," it should be understood that this is but one exemplary embodiment. The present application may be applied to any other traffic type. For example, the present application may be applied to different transportation system environments, including land, sea, or air, among others, or any combination thereof. The transportation means of the transportation system may include taxis, private cars, windmills, buses, trains, bullet trains, high speed railways, subways, ships, airplanes, spacecraft, hot air balloons, or unmanned vehicles, etc., or any combination thereof. The present application may also include any service system for network about a drive, e.g., a system for sending and/or receiving express, a service system for a business of both parties. Applications of the systems or methods of the present application may include web pages, plug-ins to browsers, client terminals, customization systems, internal analysis systems, or artificial intelligence robots, etc., or any combination thereof.

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

The terms "passenger," "requestor," "attendant," "service requestor," and "customer" are used interchangeably herein to refer to a person, entity, or tool that may request or subscribe to a service. The terms "driver," "provider," "service provider," and "provider" are used interchangeably herein to refer to a person, entity, or tool that can provide a service. The term "user" in this application may refer to a person, entity, or tool requesting, subscribing to, providing, or facilitating the provision of a service. For example, the user may be a passenger, driver, operator, etc., or any combination thereof. In this application, "passenger" and "passenger terminal" may be used interchangeably, and "driver" and "driver terminal" may be used interchangeably.

According to the technical problems known from the background, a scenario in which a user dials a customer service call is further described below. At present, when a user dials a customer service call, a recording menu is issued to the user for playing. Taking the network taxi service as an example, the recording menu issued to the user may include these service options: passenger questions 1, driver questions 2, cost questions 3, security questions 4, manual service questions 5, etc., and generally, passenger questions, driver questions, cost questions, security questions may include at least one hierarchy of other service options. For example, when the user presses the fee question 3, a plurality of questions under the fee question, such as the detour question is pressed 1, the additional fee question is pressed 2, the fee is pressed 3 even if the fee is not finished, the fee is pressed 4 when the user is not seated, and so on, are also played.

Therefore, the user needs to listen to the menu broadcast in a multi-level manner when consulting the problem, and accurate recommendation and personalized recommendation cannot be realized, so that the waiting time of the user is prolonged, the lost problem is serious, and the problem of the user is difficult to solve in time.

Before the technical solutions provided in the following embodiments are provided, the present inventors have found through research that, in the existing solutions, a plurality of specific problems known by a service platform are generally collected as classification targets, and are ranked according to probability values of occurrence of the respective classification targets, so that a problem with top ranking, for example, a problem with top ten ranking is recommended for a user consulting the problem. However, the problem with the above solution is that the problem of low probability is not easily predicted, resulting in low overall prediction accuracy. And features among various classification targets are not obvious, so that even if the method is used for training the deep network model to predict, the accuracy of the actual prediction of the finally trained deep network model is low.

In addition, the inventor also discovers that in the existing scheme, a multi-layer menu mode can be used, the overall large-class problem of the first-layer menu is firstly predicted, and the detailed small-class problem prediction is performed on the second menu under the overall large-class problem. However, the mode of multi-layer menu always requires the user to repeatedly perform multi-layer menu operation, which increases the waiting time of the user and makes it difficult to accurately locate the user problem. Because it is desirable for users to directly solve the problem of interest, i.e., it is more desirable to directly predict the problem of own needs to solve, specific solutions to specific problems.

Based on the above technical problem discovery, the embodiments of the present application provide a problem recommendation method, apparatus, server and readable storage medium, which analyze historical service record data of a user according to a target service item selected by the user from a issued problem recommendation menu, determine a problem recommendation path according to an analysis result, and then perform a distribution process on the problem acquisition request according to the determined problem recommendation path, so as to implement intelligent distribution of the problem acquisition request according to historical service record data of different users, enable the user to find a channel for solving the problem of the user more quickly, and promote user experience.

Fig. 1 is a schematic architecture diagram of a problem recommendation system 100 according to an alternative embodiment of the present application. For example, the problem recommendation system 100 may be an online transportation service platform for a transportation service such as a taxi service, a ride service, a express service, a carpool service, a bus service, a driver rental service, or a class service, or a combination service of any of the above. The problem recommendation system 100 may include a server 110, a network 120, a service requester terminal 130, a service provider terminal 140, and a database 150, and a processor executing instruction operations may be included in the server 110. The question recommending system 100 shown in fig. 1 is only one possible example, and in other possible embodiments, the question recommending system 100 may include only a part of the components shown in fig. 1 or may include other components.

In some embodiments, the server 110 may be a single server or a group of servers. The server farm may be centralized or distributed (e.g., server 110 may 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 stored in the service requester terminal 130, the service provider terminal 140, and 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 information and/or data stored therein. In some embodiments, server 110 may be implemented on a cloud platform; for example only, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud (community cloud), distributed cloud, inter-cloud (inter-cloud), multi-cloud (multi-cloud), and the like, or any combination thereof. In some embodiments, server 110 may be implemented on an electronic device 200 having one or more of the components shown in fig. 2 herein.

In some embodiments, server 110 may include a processor. The processor may process information and/or data related to the service request to perform one or more functions described herein. For example, in a express service, the processor may determine the target vehicle based on a service request obtained from the service requester terminal 130. The processor may include one or more processing cores (e.g., a single core processor (S) or a multi-core processor (S)). By way of example only, the Processor may include a central processing unit (Central Processing Unit, CPU), application specific integrated circuit (Application Specific Integrated Circuit, ASIC), special instruction set Processor (Application Specific Instruction-set Processor, ASIP), graphics processing unit (Graphics Processing Unit, GPU), physical processing unit (Physics Processing Unit, PPU), digital signal Processor (Digital Signal Processor, DSP), field programmable gate array (Field Programmable Gate Array, FPGA), programmable logic device (Programmable Logic Device, PLD), controller, microcontroller unit, reduced instruction set computer (Reduced Instruction Set Computing, RISC), 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 problem recommendation system 100 (e.g., the server 110, the service requester terminal 130, the service provider terminal 140, and the 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, network 120 may be any type of wired or wireless network, or a combination thereof. By way of example only, 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 (Local Area Network, LAN), a wide area network (Wide Area Network, WAN), a wireless local area network (Wireless Local Area Networks, WLAN), a metropolitan area network (Metropolitan Area Network, MAN), a wide area network (Wide Area Network, WAN), a public switched telephone network (Public Switched Telephone Network, PSTN), a bluetooth network, a ZigBee network, a near field communication (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, the 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 problem recommendation system 100 may connect to the network 120 to exchange data and/or information.

In some embodiments, the user of the service requester terminal 130 may be a person other than the actual consumer of the service. For example, user a of service requester terminal 130 may use service requester terminal 130 to initiate a service request for service actual requester B (e.g., user a may call his own friend B), or receive service information or instructions from server 110, etc. In some embodiments, the user of the service provider terminal 140 may be the actual service provider or may be a person other than the actual service provider. For example, user C of service provider terminal 140 may use service provider terminal 140 to receive a service request for providing a service by service actual provider D (e.g., user C may pick up for driver D employed by himself), and/or information or instructions from server 110. In some embodiments, "service requester" and "service requester terminal" may be used interchangeably and "service provider" and "service provider terminal" may be used interchangeably.

In some embodiments, the service requester terminal 130 may include a mobile device, a tablet computer, a laptop computer, or a built-in device in a motor vehicle, or the like, or any combination thereof. In some embodiments, the mobile device may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, or an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home device may include a smart lighting device, a control device for a smart appliance device, a smart monitoring device, a smart television, a smart video camera, or an intercom, or the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, a smart lace, a smart glass, a smart helmet, a smart watch, a smart garment, a smart backpack, a smart accessory, etc., or any combination thereof. In some embodiments, the smart mobile device may include a smart phone, a personal digital assistant (Personal Digital Assistant, PDA), a gaming device, a navigation device, or a point of sale (POS) device, or the like, or any combination thereof. In some embodiments, the virtual reality device and/or the augmented reality device may include a virtual reality helmet, a virtual reality glass, a virtual reality patch, an augmented reality helmet, an augmented reality glass, an augmented reality patch, or the like, or any combination thereof. For example, the virtual reality device and/or the augmented reality device may include various virtual reality products, and the like. In some embodiments, the built-in devices in the motor vehicle may include an on-board computer, an on-board television, and the like.

Database 150 may store data and/or instructions. In some embodiments, database 150 may store data obtained from service requester terminal 130 and/or service provider terminal 140. In some embodiments, database 150 may store data and/or instructions for the exemplary methods described in this application. In some embodiments, database 150 may include mass storage, removable storage, volatile Read-write Memory, or Read-Only Memory (ROM), or the like, 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, magnetic tape, and the like; the volatile read-write memory may include random access memory (Random Access Memory, RAM); the RAM may include dynamic RAM (Dynamic Random Access Memory, DRAM), double data Rate Synchronous dynamic RAM (DDR SDRAM); static Random-Access Memory (SRAM), thyristor RAM (T-RAM) and Zero-capacitor RAM (Zero-RAM), etc. By way of example, ROM may include Mask Read-Only Memory (MROM), programmable ROM (Programmable Read-Only Memory, PROM), erasable programmable ROM (Programmable Erasable Read-Only Memory, PEROM), electrically erasable programmable ROM (Electrically Erasable Programmable Read Only Memory, EEPROM), compact disk ROM (CD-ROM), digital versatile disk ROM, and the like. In some embodiments, database 150 may be implemented on a cloud platform. For example only, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, cross-cloud, multi-cloud, or other similar, or the like, or any combination thereof.

In some embodiments, database 150 may be connected to network 120 to communicate with one or more components in problem recommendation system 100 (e.g., server 110, service requester terminal 130, service provider terminal 140, etc.). One or more components in the problem recommendation system 100 may access data or instructions stored in the database 150 via the network 120. In some embodiments, database 150 may be directly connected to one or more components in the problem recommendation system 100 (e.g., server 110, service requester terminal 130, service provider terminal 140, etc.); alternatively, in some embodiments, database 150 may also be part of server 110.

In some embodiments, one or more components in the problem recommendation system 100 (e.g., server 110, service requester terminal 130, service provider terminal 140, etc.) may have access to the database 150. In some embodiments, one or more components in the problem recommendation system 100 may read and/or modify information related to a service requester, a service provider, or the public, or any combination thereof, when certain conditions are met. For example, server 110 may read and/or modify information of one or more users after receiving a service request.

In some embodiments, the exchange of information of one or more components in the problem recommendation system 100 may be accomplished by requesting a service. The object of the service request may be any product. In some embodiments, the product may be a tangible product or a non-physical product. The tangible product may include a food, a pharmaceutical, a merchandise, a chemical product, an appliance, a garment, an automobile, a house, a luxury item, or the like, or any combination thereof. The non-substance product may include a service product, a financial product, a knowledge product, an internet product, or the like, or any combination thereof. The internet product may include a host product alone, a web product, a mobile internet product, a commercial host product, an embedded product, or the like, or any combination thereof. The internet product may be used in software, a program, a system, etc. of the mobile terminal, or any combination thereof. The mobile terminal may include a tablet computer, a notebook computer, a mobile phone, a personal digital assistant (Personal Digital Assistant, PDA), a smart watch, a Point of sale (POS) device, a car computer, a car television, or a wearable device, or the like, or any combination thereof. For example, the internet product may be any software and/or application used in a computer or mobile phone. The software and/or applications may involve social, shopping, shipping, entertainment time, learning, or investment, or the like, or any combination thereof. In some embodiments, the transportation related software and/or applications may include travel software and/or applications, vehicle scheduling software and/or applications, drawing software and/or applications, and the like. In the vehicle scheduling software and/or applications, the vehicle may include horses, dollies, rickshaw (e.g., wheelbarrows, bicycles, tricycles, etc.), automobiles (e.g., taxis, buses, private cars, etc.), trains, subways, watercraft, aircraft (e.g., aircraft, helicopters, space shuttles, rockets, hot air balloons, etc.), and the like, or any combination thereof.

Fig. 2 shows a schematic diagram of exemplary hardware and software components of an electronic device 200 provided by some embodiments of the present application that may implement the concepts of the present application, a server 110, a service requester terminal 130, and a service provider terminal 140. For example, the processor 220 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 problem recommendation method of the present application. Although only one computer is shown, the functionality described herein may be implemented in a distributed fashion across multiple similar platforms for convenience 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 various forms of storage media 240, such as magnetic disk, ROM, or RAM, or any combination thereof. By way of example, 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 methods 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. It should be noted, however, that the electronic device 200 in the present application may also include multiple processors, and thus steps performed by one processor described in the present application may also be performed jointly by multiple processors or separately. For example, if the processor of the electronic device 200 performs steps a and B, it should be understood that steps a and B may also be performed by two different processors together or performed separately in one processor. For example, the first processor performs step a, the second processor performs step B, or the first processor and the second processor together perform steps a and B.

Fig. 3 is a flow chart illustrating a problem recommendation method provided in some embodiments of the present application, which may be performed by the server 110 shown in fig. 1. It should be understood that, in other embodiments, the order of some steps in the problem recommendation method of the present embodiment may be interchanged according to actual needs, or some steps may be omitted or deleted. The detailed steps of the problem recommendation method are described below.

Step S110, according to the problem acquisition request sent by the service requester terminal 130, the history service record data of the service requester is acquired.

During the use of various services (e.g., travel service, takeaway service), a service requester (e.g., passenger, driver, etc.) may establish session communication with the server 110 through the service requester terminal 130 to consult various questions to be resolved through a single round of dialogue. For example, customer service telephones for these services may be placed by the service requester terminal 130 to send corresponding incoming call requests to the server 110. Taking a travel service as an example, a corresponding incoming call request may be sent to the server 110 of the travel APP through a customer service phone provided on the travel APP installed on the service requester terminal 130. For another example, the problem acquisition request may also be generated by an operation of some indication control (e.g., contact customer service, complaint feedback, online inquiry, etc.) in a service interface (e.g., a service interface provided by an application, a WeChat applet, a WEB page, a WeChat public number, etc.) presented by the service requester terminal 130 and sent to the server 110.

Before that, the server 110 may collect historical service record data for each service requester during use of the services by each service requester, and build a historical service record database, where the historical service record data may include one or more combinations of service requester portrayal features, service order data, service provider portrayal features, and service temporal and spatial information.

In detail, the service requester portrayal feature includes at least basic information of the service requester, service usage frequency, service type distribution and complaint problem distribution, the service provider portrayal feature includes at least basic information of the service provider, average service statistics and complaint problem distribution, the service order data includes order statistics of the latest preset number of service orders, the service spatiotemporal information includes time information, location information and thermodynamic diagram information when a problem acquisition request is sent, wherein the basic information of the service requester includes age, sex, service grade and occupation, the basic information of the service provider includes age, sex and service grade, and the average service statistics includes average daily service duration and average daily income.

Taking travel services as an example, service requester image features may include the age, sex, class of service, and occupation of passengers, travel service usage frequency, carpool/share/non-carpool distribution, and complaint problem distribution. Service requester portrayal features may include age, sex, and class of service of the driver, average length of service per day, average revenue per day, and complaint problem distribution. The service order data may include order statistics of two recent travel orders, and the order statistics may include statistics of order estimated cost, actual cost, estimated time length, actual time length, bridge fee, high-speed fee, estimated price/actual cost price value, estimated time/actual time ratio, and time interval from beginning trip to ending trip to payment. The thermodynamic diagram information may include the travel order quantity of the section where the position information and the time information at the time of sending the problem acquisition request are located.

It should be noted that, in order to improve the referenceability of the historical service record data, the historical service record data is more capable of indicating the service usage of each user recently, and the service order data of each service requester should be the historical data within a certain period (for example, 20 days) before the current time node, or the order statistics data of the last two orders.

Thus, when the server 110 receives the problem acquisition request, first, the user information of the service requester is acquired from the problem acquisition request, and the history service record data of the service requester is acquired from the history service record database stored in the server 110 according to the user information.

Optionally, the user information may include at least one of an incoming call number, a user account number, and a user biometric feature. For example, if the question acquisition request is an incoming call request generated by the service requester terminal 130 making a customer service call, the incoming call number of the target user may be acquired from the incoming call request. For another example, if the problem acquisition request is a problem acquisition request generated when some indication controls in the application program are triggered, a user account or a user biometric of the target user may be acquired from the problem acquisition request. The user biometric characteristic may be any identifiable biometric characteristic such as a fingerprint characteristic, a face characteristic, an iris characteristic, etc., which is not limited in this embodiment.

Step S120, inputting the history service record data into a pre-trained collaborative filtering model to obtain the confidence coefficient of each candidate main class problem, and selecting a target main class problem from each candidate main class problem according to the confidence coefficient of each candidate main class problem.

Wherein, prior to step S120, the collaborative filtering model may be trained by:

first, a main class training sample set is obtained. The main class training sample set may include main class question service record data of each service requester and candidate main class questions associated with the main class question service record data, where the question service record data includes service requester portraits, service order data, service provider portraits, and service space-time information.

Illustratively, still taking the travel service as an example, the above candidate major class questions may include "cost questions", "time questions", "driver side questions", "passenger side questions", and "security questions". The main class question service record data may respectively include the history service record data of the passenger, which is obtained when the passenger consults some candidate main class questions, such as consulting "expense questions" and "time questions", and the candidate main class questions associated with the main class question service record data are "expense questions" and "time questions".

Wherein each candidate master class question may be determined based on previously recorded questions that each service requester terminal 130 consults during a single round of session. For example, in one possible implementation, the total number of times each of the questions requested to be consulted by the different service requester terminals 130 in the preset time period may be counted, and then the questions whose counted total number of times satisfies the preset condition may be used as candidate main class questions. For example, each problem with the counted total number of times exceeding the preset number of times may be used as a candidate main class problem, or the total number of times corresponding to each problem may be arranged in order from large to small, and each problem with the total number of times arranged in the first M number of times may be used as a candidate main class problem, where M is a positive integer.

And then training by adopting a collaborative filtering algorithm according to the training sample set to obtain a collaborative filtering model.

The candidate main class problem features are more outstanding, the prediction accuracy is high, the problem is more definite, the positioning problem is fast, and the regularity is stronger, for example:

cost problems: the cost is abnormal, and the actual price is obviously higher than the estimated price.

Time problem: the time is abnormal, and the time running time is longer than the average time of the route.

Driver side problem: poor service attitudes, poor vehicle sanitation, etc.

Passenger side problem: lost articles, etc.

Safety problem: driver income, identity, age, complaint conditions, etc.

Therefore, the confidence coefficient of each candidate main class problem can be accurately predicted by inputting the history service record data into the trained collaborative filtering model. And then, according to the confidence level of each candidate main class problem, generating a sequencing result of each candidate main class problem according to the descending order of the confidence level, and selecting at least one candidate main class problem with the highest sequencing as a target main class problem according to the sequencing result. Alternatively, the candidate main class problem with the confidence degree larger than the preset confidence degree may be used as the target main class problem.

For example, the ranking result of generating the respective candidate master class questions in descending order of confidence may be: TOP1 (cost question, confidence 9), TOP2 (time question, confidence 8), TOP3 (driver side question, confidence 7), TOP4 (passenger side question, confidence 6), and TOP5 (security question, confidence 5), the first candidate major class question, namely "cost question", can be selected as the target major class question. Alternatively, a candidate major class question with a confidence level greater than 9, i.e., a "cost question" may be selected as the target major class question.

And step S130, inputting the history service record data into a deep neural network DNN model corresponding to the target main class problem to obtain the confidence coefficient of each candidate sub-class problem under the target main class problem.

Wherein, prior to step S130, the collaborative filtering model may be trained by:

first, a subclass training sample set is obtained for each candidate main class problem. The subclass training sample set may include subclass problem service record data of each service requester and candidate subclass problems associated with the subclass problem service record data, where the problem service record data includes service requester portraits, service order data, service provider portraits, and service temporal and spatial information.

Still taking travel service as an example, sub-class problem service record data under the conditions of 'expense problem', 'time problem', 'driver side problem', 'passenger side problem', and 'safety problem' and candidate sub-class problems associated with the sub-class problem service record data are respectively acquired. For example, when a passenger consults the candidate sub-class problems of 'more generation of fees by detours', 'not sitting on the vehicle', 'more collection of additional fees', the obtained historical service record data of the passenger is the sub-class training sample set of 'expense problems', and the candidate sub-class problems associated with the sub-class service record data are the candidate sub-class problems of 'more generation of fees by detours', 'not sitting on the vehicle', 'more collection of additional fees'.

Based on the above, according to each subclass training sample set, a DNN (Deep Neural Network ) algorithm may be used to train to obtain a DNN model for each candidate main class problem.

The DNN model may include an input layer (inputlayer), a hidden layer (hidenlayer), and an output layer (outputlayer), where: the input layer, i.e. the first layer of the DNN model, may comprise a plurality of input nodes, e.g. when the extracted feature vector comprises 200-dimensional features, the input nodes 15 may be 200; the output layer, i.e. the last layer of the DNN model, includes the number of output nodes depending on the kind of candidate sub-class questions included in each candidate main class question, for example, when 10 candidate sub-class questions are included in the candidate main class questions, then the output layer may include 10 output nodes; the hidden layers are positioned between the input layer and the output layer, the hidden layers can be multiple layers, and the more the hidden layers are, the more nodes each hidden layer contains, the stronger the expression capability of the DNN model is.

Wherein each candidate sub-class question may be determined based on previously recorded questions that each service requester terminal 130 consults during a single round of session. For example, in one possible implementation, the total number of times each of the questions that the different service requester terminal 130 requests the consultation within a preset period of time (for example, within the past three months) may be counted, and then the counted questions that the total number of times satisfies the preset condition are taken as the candidate sub-category questions. For example, each problem with the counted total number of times exceeding the preset number of times may be used as a candidate sub-class problem, or the total number of times corresponding to each problem may be arranged in order from large to small, and each problem with the total number of times arranged in the first M number of times may be used as a candidate sub-class problem, where M is a positive integer.

Thus, the historical service record data is input into the deep neural network DNN model corresponding to the target main class problem, and the confidence coefficient of each candidate sub-class problem under the target main class problem can be obtained. For example, the historical service record data is input into a deep neural network DNN model corresponding to the "cost problem", so that the confidence level of the "more generated cost by-pass", "more additional cost", "more generated cost after charging not being finished in time", "more generated cost after getting consumed" and more generated cost after not sitting on the car "of each candidate sub-class problem under the" cost problem "can be obtained.

Step S140, a question recommending menu is sent to the service requester terminal 130 according to the confidence coefficient of each candidate main class question and the confidence coefficient of each candidate sub-class question under the target main class question, wherein the question recommending menu comprises main analogies and sub-class recommending questions under the target main class question.

As one possible implementation, a first preset number of candidate main class questions may be selected as main analogies according to the confidence level of each candidate main class question in descending order of the confidence level. Meanwhile, according to the confidence coefficient of each candidate sub-class problem under the target main class problem, selecting a second preset number of candidate sub-class problems as sub-class recommendation problems according to the descending order of the confidence coefficient. Finally, a question recommendation menu is generated according to the main analogized questions and the sub-class recommendation questions under the target main class questions, and the question recommendation menu is sent to the service requester terminal 130.

For example, suppose that the ranking result of generating the respective candidate main class questions in descending order of confidence is: TOP1 (cost question, confidence 9), TOP2 (time question, confidence 8), TOP3 (driver side question, confidence 7), TOP4 (passenger side question, confidence 6), and TOP5 (security question, confidence 5), taking "cost question" as the target main class question as an example, the ranking result of each candidate sub-class question is generated in descending order of confidence: TOP1 (detour with more charge, confidence 9), TOP2 (with more charge, confidence 8), TOP3 (without timely ending the charge with more charge, confidence 7), TOP4 (with less charge, confidence 6), TOP5 (without sitting with more charge, confidence 5).

On the basis of the above, candidate main class questions of TOP2 can be selected as main analogies, namely 'cost questions' and 'time questions'. Meanwhile, with the "expense problem" as the target main class problem, the candidate subclass problem of TOP3 under the "expense problem" can be selected as the subclass recommendation problem, namely "detour multi-generation expense", "multi-charge additional expense" and "multi-generation expense without timely ending the charging. Finally, the generated problem recommendation menu is "more generation cost on the detour", "more additional cost, more generation cost when charging is not finished in time" and "time problem".

As another possible implementation manner, the preset main class questions with the confidence degree larger than the first confidence degree can be used as main analogies according to the confidence degrees of the preset main class questions. Meanwhile, according to the confidence coefficient of each preset sub-class problem under the target main class problem, the preset sub-class problem with the confidence coefficient larger than the second confidence coefficient is used as the sub-class recommendation problem. Finally, a question recommendation menu is generated according to the main analogized questions and the sub-class recommendation questions under the target main class questions, and the question recommendation menu is sent to the service requester terminal 130.

Still taking the above example as an example, candidate major class questions with confidence greater than 8 may be selected as major referral questions, namely "cost questions" and "time questions. Meanwhile, with the ' expense problem ' as the target main class problem, the candidate subclass problem with the confidence degree larger than 7 under the ' expense problem ' can be selected as the subclass recommendation problem, namely ' detour multi-generation expense ', ' multi-charge and ' charging multi-generation expense not ended in time '. Finally, the generated problem recommendation menu is "more generation cost on the detour", "more additional cost, more generation cost when charging is not finished in time" and "time problem".

After the above-described question recommendation menu is generated, the user may be prompted to select a question that meets the condition in a voice playing manner with respect to the incoming call request transmitted from the service requester terminal 130. For example, the voice data of the main referral questions and the sub-class recommendation questions under the target main class questions included in the question recommendation menu may be transmitted to the service requester terminal 130, so that the service requester terminal 130 plays the voice data of the main referral questions and the sub-class recommendation questions under the target main class questions to the service requester. For example, the voice data may prompt the passenger: the "detour multiple generation cost" is requested to be pressed 1, the "multiple additional cost is requested to be pressed 2, the" not timely ending the charging multiple generation cost "is requested to be pressed 3, etc., and the" time problem "is requested to be pressed 4.

Or, the target user can be prompted to select the service recommendation item meeting the conditions in a page display mode. For example, the main referral questions and the sub-class recommendation questions under the target main class questions included in the question recommendation menu may be sent to the service requester terminal 130, so that the service requester terminal 130 displays the main referral questions and the sub-class recommendation questions under the target main class questions to the user. For example, "detour more generation cost", "more additional cost", "not timely ending of charging more generation cost", "time problem" are displayed to passengers, respectively.

Based on the design, the embodiment predicts the main analogies, then selects the target main class problem, continuously inputs the history service record data into the deep neural network DNN model corresponding to the target main class problem, predicts the specific subclass recommendation problem under the target main class problem, and finally outputs the problem recommendation menu which can simultaneously comprise the main analogies and the specific subclass recommendation problem. Therefore, by utilizing the characteristics of prominent problem characteristics and high prediction accuracy of the main referral problem, the main referral problem is predicted first and then the sub-class recommendation problem is predicted, so that the prediction accuracy can be effectively improved, on the other hand, the problem recommendation menu directly comprises the specific sub-class recommendation problem, and a user does not need to repeatedly perform multi-layer menu operation, so that the waiting time of the user is reduced, and the user problem is accurately positioned.

Further, the user may select a target recommendation problem meeting the conditions from the main recommendation problem and the sub-category recommendation problem under the target main category problem according to the actual requirement, and send the target recommendation problem to the server 110 through the service requester terminal 130. For example, if the passenger encounters a "detour multi-occurrence cost" question, then "detour multi-occurrence cost" may be selected as the target recommended question. For another example, if the passenger encounters a "time question," the "time question" may be selected as the target recommendation question.

After the server 110 obtains the target recommendation problem, it determines whether the target recommendation problem is a main analogized problem. If the target recommendation problem is the main recommendation problem, the history service record data is input into a DNN model corresponding to the main recommendation problem, the confidence coefficient of each preset sub-class problem under the main recommendation problem is obtained, and the sub-class recommendation problem under the main recommendation problem is sent to the service requester terminal 130 according to the confidence coefficient of each preset sub-class problem. If the target recommendation problem is a sub-category recommendation problem, the sub-category recommendation problem is processed according to a problem processing strategy of the sub-category recommendation problem.

For example, if the target recommendation problem is the main analoging problem of "time problem", the history service record data is input into the DNN model corresponding to "time problem", so as to obtain the confidence level of each preset sub-category problem under "time problem", and the sub-category recommendation problem under "time problem" is sent to the service requester terminal 130 according to the confidence level of each preset sub-category problem.

For another example, if the target recommendation problem is a sub-category recommendation problem of "detour multi-generation cost", the "detour multi-generation cost" is processed according to the problem processing policy of "detour multi-generation cost". For example, a voice or online interaction may be performed with a passenger in the service requester to prompt the passenger to enter a refund operation flow.

Therefore, the problems of the user in the service using process can be rapidly positioned, and the user experience is improved.

Fig. 4 is a functional block diagram of a problem recommendation device 300 according to some embodiments of the present application, where functions implemented by the problem recommendation device 300 may correspond to steps performed by the above-described method. The problem recommending apparatus 300 may be understood as the above-mentioned server 110, or a processor of the server 110, or may be understood as a component that implements the functions of the present application under the control of the server 110, which is independent of the above-mentioned server 110 or processor, as shown in fig. 4, the problem recommending apparatus 300 may include an obtaining module 310, a first input module 320, a second input module 330, and a menu transmitting module 340, and the functions of the respective functional modules of the problem recommending apparatus 300 are described in detail below.

The obtaining module 310 may be configured to obtain, according to a problem obtaining request sent by the service requester terminal 130, historical service record data of the service requester, where the historical service record data includes one or more combinations of service requester portrait features, service order data, service provider portrait features, and service space-time information. It will be appreciated that the acquisition module 310 may be configured to perform step S110 described above, and reference may be made to the details of implementation of the acquisition module 310 regarding step S110 described above.

The first input module 320 may be configured to input the history service record data into a pre-trained collaborative filtering model, obtain a confidence coefficient of each candidate main class problem, and select a target main class problem from each candidate main class problem according to the confidence coefficient of each candidate main class problem. It is understood that the first input module 320 may be used to perform the step S120 described above, and reference may be made to the details of the implementation of the first input module 320 related to the step S120.

The second input module 330 may be configured to input the history service record data into a deep neural network DNN model corresponding to the target main class problem, to obtain the confidence level of each candidate sub-class problem under the target main class problem. It is understood that the second input module 330 may be used to perform the step S130, and reference may be made to the details of the implementation of the second input module 330 related to the step S130.

The menu sending module 340 may be configured to send, to the service requester terminal 130, a question recommendation menu according to the confidence level of each candidate main class question and the confidence level of each candidate sub-class question under the target main class question, where the question recommendation menu includes a main analogized question and a sub-class recommendation question under the target main class question. It will be appreciated that the menu sending module 340 may be used to perform step S140 described above, and reference may be made to the details of the implementation of the menu sending module 340 described above with respect to step S140.

In one possible implementation, the obtaining module 310 may specifically obtain the historical service record data of the service requester by:

acquiring user information of the service requester from a problem acquisition request, wherein the user information comprises at least one of an incoming call number, a user account number and a user biological feature;

the history service record data of the service requester is acquired from a history service record database stored in the server 110 according to the user information.

In one possible embodiment, the service requester profile includes at least basic information of the service requester including age, sex, service class, and occupation, service type distribution, and complaint problem distribution, the service provider profile includes at least basic information of the service provider including average service duration and average income per day, average service statistics including order statistics of a latest preset number of service orders, service spatiotemporal information including time information, location information, and thermodynamic diagram information when the problem acquisition request is sent, and average service statistics including average service duration per day and average income per day.

In a possible implementation manner, referring further to fig. 5, the problem recommendation device may further include a first training module 301 for pre-training the collaborative filtering model, where the first training module 301 may specifically pre-train the collaborative filtering model by:

and training by adopting a collaborative filtering algorithm according to the training sample set to obtain a collaborative filtering model.

In a possible implementation manner, referring to fig. 6, further, the problem recommending apparatus may further include a second training module 302 for pre-training the DNN model corresponding to each candidate main class problem, where the second training module 302 may specifically pre-train the DNN model corresponding to each candidate main class problem by:

acquiring a sub-class training sample set of each candidate main class problem, wherein the sub-class training sample set comprises sub-class problem service record data of each service requester and candidate sub-class problems associated with the sub-class problem service record data, and the problem service record data comprises service requester portraits, service order data, service provider portraits and service space-time information;

In one possible implementation, the first input module 320 may specifically select the target main class question from the candidate main class questions by:

selecting at least one candidate main class problem with the forefront ranking as a target main class problem according to the ranking result; or alternatively

And taking the candidate main class problem with the confidence coefficient larger than the preset confidence coefficient as the target main class problem.

In one possible implementation, the menu transmitting module 340 may specifically transmit the question recommendation menu to the service requester terminal 130 by:

and generating a question recommendation menu according to the main analogized questions and the sub-class recommendation questions under the target main class questions, and sending the question recommendation menu to the service requester terminal 130.

and sending the voice data of the main analoging problem and the sub-class recommendation problem under the target main class problem included in the problem recommendation menu to the service requester terminal 130, so that the service requester terminal 130 plays the voice data of the main analoging problem and the sub-class recommendation problem under the target main class problem to the service requester.

In one possible implementation, the menu sending module 340 may specifically send the question recommendation menu by:

And sending the main analogies and the sub-category recommended questions under the target main category questions included in the question recommendation menu to the service requester terminal 130, so that the service requester terminal 130 displays the main analogies and the sub-category recommended questions under the target main category questions to the user.

In one possible implementation, the menu sending module 340 is further configured to:

acquiring a target recommendation problem selected by the service requester terminal 130 from the sub-class recommendation problem under the main analogized problem and the target main class problem;

judging whether the target recommendation problem is a main analogized problem or not;

if the target recommendation problem is a main recommendation problem, the history service record data is input into a DNN model corresponding to the main recommendation problem, and the confidence coefficient of each candidate sub-class problem under the main recommendation problem is obtained;

transmitting sub-category recommendation questions under the main analoging questions to the service requester terminal 130 according to the confidence level of each candidate sub-category question;

if the target recommendation problem is a sub-category recommendation problem, the sub-category recommendation problem is processed according to a problem processing strategy of the sub-category recommendation problem.

The modules may be connected or communicate with each other via wired or wireless connections. The wired connection may include a metal cable, optical cable, hybrid cable, or the like, or any combination thereof. The wireless connection may include a connection through a LAN, WAN, bluetooth, zigBee, or NFC, or any combination thereof. Two or more modules may be combined into a single module, and any one module may be divided into two or more units.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described system and apparatus may refer to corresponding procedures in the method embodiments, which are not described in detail in this application. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, and the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, and for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, indirect coupling or communication connection of devices or modules, electrical, mechanical, or other form.

The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in 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 may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.

The foregoing is merely a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes or substitutions are covered in the protection 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 problem recommendation method, applied to a server, comprising:

according to the confidence coefficient of each candidate main class problem and the confidence coefficient of each candidate sub-class problem under the target main class problem, a problem recommendation menu is sent to the service requester terminal in a voice broadcasting or page displaying mode, and the problem recommendation menu comprises main analogies and sub-class recommendation problems under the target main class problem.

2. The question recommending method according to claim 1, wherein the step of acquiring the history service record data of the service requester according to the question acquisition request transmitted by the service requester terminal comprises:

3. The problem recommendation method according to claim 1, wherein the service requester profile includes at least basic information of the service requester, service usage frequency, service type distribution, and complaint problem distribution, the service provider profile includes at least basic information of the service provider, average service statistics including a most recent order statistics of a preset number of service orders, and average service statistics including time information, location information, and thermodynamic diagram information at the time of sending the problem acquisition request, wherein the basic information of the service requester includes age, sex, service class, and occupation, and the basic information of the service provider includes age, sex, and service class, and the average service statistics includes average daily service duration and average daily incomes.

4. The problem recommendation method according to claim 1, further comprising the step of pre-training a collaborative filtering model, comprising in particular:

5. The problem recommendation method according to claim 1, further comprising the step of training a DNN model corresponding to each candidate main class problem, specifically comprising:

6. The question recommending method according to claim 1, wherein the step of selecting a target main class question from the respective candidate main class questions according to the confidence levels of the respective candidate main class questions comprises:

7. The question recommending method according to any one of claims 1 to 6, wherein the step of transmitting a question recommending menu to the service requester terminal according to the confidence level of each candidate major class question and the confidence level of each candidate minor class question under the target major class question comprises:

8. The question recommending method according to any one of claims 1 to 6, wherein the step of transmitting a question recommending menu to the service requester terminal according to the confidence level of each candidate major class question and the confidence level of each candidate minor class question under the target major class question comprises:

9. The question recommending method according to claim 1, wherein the step of transmitting a question recommending menu to the service requester terminal comprises:

10. The question recommending method according to claim 1, wherein the step of transmitting a question recommending menu to the service requester terminal comprises:

11. The question recommending method according to claim 1, wherein after the step of transmitting a question recommending menu to the service requester terminal according to the confidence level of each candidate major class question and the confidence level of each candidate sub-class question under the target major class question, the method further comprises:

12. A question recommending apparatus, applied to a server, comprising:

the acquisition module is used for acquiring historical service record data of the service requester according to a problem acquisition request sent by the service requester terminal, wherein the historical service record data comprises one or more combinations of service requester portrait features, service order data, service provider portrait features and service space-time information;

The first input module is used for inputting the history service record data into a pre-trained collaborative filtering model to obtain the confidence coefficient of each candidate main class problem, and selecting a target main class problem from each candidate main class problem according to the confidence coefficient of each candidate main class problem;

the second input module is used for inputting the history service record data into a deep neural network DNN model corresponding to the target main class problem to obtain the confidence coefficient of each candidate sub-class problem under the target main class problem;

and the menu sending module is used for sending a problem recommendation menu to the service requester terminal in a voice broadcasting or page displaying mode according to the confidence coefficient of each candidate main class problem and the confidence coefficient of each candidate sub-class problem under the target main class problem, wherein the problem recommendation menu comprises main analogized problems and sub-class recommendation problems under the target main class problem.

13. The question recommending apparatus according to claim 12, wherein the acquiring module acquires the history service record data of the service requester specifically by:

14. The question recommending apparatus according to claim 12, wherein the service requester profile includes at least basic information of the service requester, service use frequency, service type distribution, and complaint question distribution, the service provider profile includes at least basic information of the service provider, average service statistics including a most recent preset number of service orders, and complaint question distribution, the service order data includes order statistics including time information, location information, and thermodynamic diagram information at the time of sending the question acquisition request, wherein the basic information of the service requester includes age, sex, service class, and occupation, and the basic information of the service provider includes age, sex, and service class, and the average service statistics includes average daily service duration and average daily incomes.

15. The question recommending apparatus of claim 12, further comprising a first training module for pre-training a collaborative filtering model;

The first training module pre-trains the collaborative filtering model specifically by:

16. The question recommending apparatus according to claim 12, further comprising a second training module for pre-training a DNN model corresponding to each of the candidate main class questions;

the second training module specifically pre-trains a DNN model corresponding to each candidate main class problem in the following manner:

17. The question recommending apparatus according to claim 12, wherein the first input module selects a target main class question from the respective candidate main class questions by:

18. The question recommending apparatus according to any of the claims 12-17, wherein the menu transmitting module transmits the question recommending menu to the service requester terminal, in particular by:

19. The question recommending apparatus according to any of the claims 12-17, wherein the menu transmitting module transmits the question recommending menu to the service requester terminal, in particular by:

20. The question recommending apparatus according to claim 12, wherein the menu transmitting module transmits the question recommending menu to the service requester terminal specifically by:

21. The question recommending apparatus according to claim 12, wherein the menu transmitting module transmits the question recommending menu by, in particular:

22. The question recommending apparatus of claim 12, wherein the menu transmitting module is further configured to:

23. A server, 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 over the bus when run by a server, the processor executing the machine-readable instructions to perform the steps of the problem recommendation method according to any one of claims 1-11 when executed.

24. A readable storage medium, characterized in that the readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the problem recommendation method according to any one of claims 1-11.