CN116757771A

CN116757771A - Scheme recommendation method, device, equipment and storage medium based on artificial intelligence

Info

Publication number: CN116757771A
Application number: CN202310666304.0A
Authority: CN
Inventors: 孟繁烨
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2023-06-06
Filing date: 2023-06-06
Publication date: 2023-09-15

Abstract

The embodiment of the application belongs to the field of artificial intelligence and the field of financial science and technology, and relates to a scheme recommendation method based on artificial intelligence, which comprises the following steps: acquiring user data of a target user; determining a user type of a user based on the user data, and determining a target recommendation model corresponding to the user type; extracting user characteristic data from the user data, and acquiring target user characteristics from the user characteristic data based on a characteristic analysis model; acquiring the characteristics of an activation scheme to be recommended; processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate a prediction result; and determining a recommended promotion scheme from the promotion schemes to be recommended based on the prediction result. The application also provides a scheme recommending device, computer equipment and a storage medium based on the artificial intelligence. Furthermore, the present application relates to blockchain technology, and recommended activation schemes may be stored in the blockchain. The method and the device can be applied to the promotion scheme recommendation scene in the financial field, and improve the accuracy of promotion scheme recommendation.

Description

Scheme recommendation method, device, equipment and storage medium based on artificial intelligence

Technical Field

The application relates to the technical field of artificial intelligence and the technical field of finance, in particular to a scheme recommendation method, a scheme recommendation device, computer equipment and a storage medium based on artificial intelligence.

Background

With the entrance of the internet field into a fierce stock market, various large internet finance companies, such as insurance companies, banks and the like, face the dilemma of continuous rising of the acquisition cost, new and increased difficulty of users, low user activity, loss of users and the like, so that the operation strategies of various large finance companies are greatly changed from large-scale growth to operation type growth. The user liveness is used as an important operation index, reflects the application health degree and the operation effect, and feeds back whether the application function is effective or not to a certain extent, and whether the application experience is good or not.

In order to improve the activity of users, various promotion activities such as sales promotion activities, check-in activities, point card activities, hot spot activities, festival activities and the like are promoted by various large finance companies, and are pushed through channels such as short messages, APP messages, official micro-operations and micro-operations, so that the promotion activities and the pushing channels are combined together to form a promotion scheme. The traditional operation strategy is to recommend the activation activity to the whole number of users through a specified channel, or define a specific user to send according to the portrait characteristic of a certain user, and the processing mode cannot achieve the effect of personalized recommendation to different users, so that the problems of low recommendation accuracy and poor user experience are solved.

Disclosure of Invention

The embodiment of the application aims to provide an artificial intelligence-based scheme recommendation method, an artificial intelligence-based scheme recommendation device, computer equipment and a storage medium, so as to solve the technical problems that an existing activation scheme pushing processing mode cannot achieve personalized recommendation effects for different users, recommendation accuracy is low, and user experience is poor.

In order to solve the technical problems, the embodiment of the application provides an artificial intelligence based scheme recommendation method, which adopts the following technical scheme:

acquiring user data of a target user;

determining a user type of the target user based on the user data, and determining a target recommendation model corresponding to the user type;

extracting user characteristic data of the target user from the user data, and acquiring target user characteristics from the user characteristic data based on a preset characteristic analysis model;

acquiring activation scheme data of an activation scheme to be recommended, and extracting features of the activation scheme data to obtain corresponding activation scheme features;

processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate corresponding prediction results;

And determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule based on the prediction result.

Further, the step of obtaining the target user feature from the user feature data based on the preset feature analysis model specifically includes:

invoking the feature analysis model;

inputting the user characteristic data into the characteristic analysis model, and sorting the importance of the user characteristic data through the characteristic analysis model to obtain a corresponding sorting result;

acquiring a preset number of feature data with highest importance from the sequencing result;

and taking the characteristic data as the target user characteristic.

Further, the user type includes a hot start user or a cold start user; the step of processing the target user feature and the activation proposal feature based on the target recommendation model to generate a corresponding prediction result specifically comprises the following steps:

if the target user is a cold start user, coding the target user characteristic by using a first coder in a preset double-tower model to obtain a corresponding user coding vector, and coding the activation scheme characteristic by using a second coder in the double-tower model to obtain a corresponding scheme coding vector;

Processing the user coding vector by using a deep neural network layer in the double-tower model to obtain a corresponding user vector, and processing the scheme coding vector to obtain a corresponding scheme vector;

performing similarity calculation on the user vector and the scheme vector by using a calculation layer in the double-tower model, and outputting a corresponding similarity calculation result;

and predicting the similarity calculation result by using a prediction layer in the double-tower model, and outputting a first prediction result of the target user for the to-be-recommended activation scheme.

if the target user is a hot start user, acquiring historical activation scheme data corresponding to the target user;

extracting features of the historical activation scheme data to obtain corresponding historical activation scheme features;

and inputting the target user characteristics, the activation scheme characteristics and the historical activation scheme characteristics into a preset depth interest model, and outputting a second prediction result of the target user on the to-be-recommended activation scheme through the depth interest model.

Further, the step of determining, based on the prediction result, a recommended activation scheme corresponding to the target user from the to-be-recommended activation schemes according to a preset recommendation rule specifically includes:

screening target prediction results meeting preset conditions from the prediction results;

acquiring a target promotion scheme corresponding to the target prediction result from the promotion scheme to be recommended;

and taking the target activation scheme as the recommended activation scheme.

Further, the step of determining the user type of the target user based on the user data specifically includes:

acquiring activation scheme push record data of the target user from the user data;

counting the number of the promotion scheme pushing records of the target user based on the promotion scheme pushing record data;

and determining the user type of the target user based on the number.

Further, after the step of determining the recommended activation scheme corresponding to the target user from the to-be-recommended activation schemes according to a preset recommendation rule based on the prediction result, the method further includes:

acquiring communication information of the target user;

Acquiring the working time of the target user;

determining push time of the target user based on the working time;

and pushing the recommended promotion scheme to the target user based on the pushing time and the communication information.

In order to solve the technical problems, the embodiment of the application also provides a scheme recommending device based on artificial intelligence, which adopts the following technical scheme:

the first acquisition module is used for acquiring user data of a target user;

the first determining module is used for determining the user type of the target user based on the user data and determining a target recommendation model corresponding to the user type;

the first extraction module is used for extracting user characteristic data of the target user from the user data and acquiring target user characteristics from the user characteristic data based on a preset characteristic analysis model;

the second extraction module is used for acquiring the activation scheme data of the activation scheme to be recommended, and extracting the characteristics of the activation scheme data to obtain corresponding activation scheme characteristics;

the generation module is used for processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate corresponding prediction results;

And the second determining module is used for determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule based on the prediction result.

In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical schemes:

acquiring user data of a target user;

In order to solve the above technical problems, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical schemes:

acquiring user data of a target user;

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

firstly, acquiring user data of a target user; then determining the user type of the target user based on the user data, and determining a target recommendation model corresponding to the user type; then extracting user characteristic data of the target user from the user data, and acquiring target user characteristics from the user characteristic data based on a preset characteristic analysis model; subsequently acquiring activation scheme data of an activation scheme to be recommended, and extracting features of the activation scheme data to obtain corresponding activation scheme features; processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate corresponding prediction results; and finally, determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule based on the prediction result. According to the embodiment of the application, the corresponding user type is determined based on the user data of the target user, different recommendation models are selected for the target user of different user types to provide recommendation processing of the targeted activation scheme for the target user, so that the adaptive activation scheme recommendation for different types of users is realized, the recommended activation scheme corresponding to the target user is intelligently determined from the to-be-recommended activation scheme based on the preset recommendation rule, the accurate screening of the to-be-recommended activation scheme is realized to determine the final recommended activation scheme for recommending to the target user, and the accuracy of the activation scheme recommendation is effectively improved.

Drawings

In order to more clearly illustrate the solution of the present application, a brief description will be given below of the drawings required for the description of the embodiments of the present application, it being apparent that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without the exercise of inventive effort for a person of ordinary skill in the art.

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of one embodiment of an artificial intelligence based solution recommendation method in accordance with the present application;

FIG. 3 is a schematic diagram of one embodiment of an artificial intelligence based solution recommender in accordance with the present application;

FIG. 4 is a schematic structural diagram of one embodiment of a computer device in accordance with the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the description of the drawings above are intended to cover a non-exclusive inclusion. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture Experts Group Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.

It should be noted that, the method for recommending a solution based on artificial intelligence provided by the embodiment of the present application is generally executed by a server/terminal device, and accordingly, the device for recommending a solution based on artificial intelligence is generally disposed in the server/terminal device.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow chart of one embodiment of an artificial intelligence based approach recommendation method in accordance with the present application is shown. The order of the steps in the flowchart may be changed and some steps may be omitted according to various needs. The method for recommending the scheme by the artificial intelligence can be applied to any scene needing scheme recommendation, and can be applied to products in the scenes, such as promotion scheme recommendation in the field of financial insurance. The scheme recommendation method based on artificial intelligence comprises the following steps:

Step S201, user data of a target user is acquired.

In this embodiment, the electronic device (e.g., the server/terminal device shown in fig. 1) on which the scheme recommendation method based on artificial intelligence operates may acquire the user data of the target user through a wired connection manner or a wireless connection manner. It should be noted that the wireless connection may include, but is not limited to, 3G/4G/5G connection, wiFi connection, bluetooth connection, wiMAX connection, zigbee connection, UWB (ultra wideband) connection, and other now known or later developed wireless connection. The user data at least comprises user characteristic data of a target user, activation scheme pushing record data and activation scheme-caused user active record data. The user characteristic data is recorded as data_feature, the total number of the user characteristic data is recorded as M, the promotion scheme pushing record data is recorded as data_record, the promotion scheme number is recorded as N, and the user active record data is recorded as data_active. In addition, only user data of the promotion scheme push record is usually reserved, and as long as any 1 promotion scheme is transmitted by the N promotion schemes, the promotion scheme push record is considered, and the processed data set is recorded as data_total. The activation scheme may include various types of activation activities, such as promotion activities initiated by insurance companies or banks, check-in activities, point card activities, hot spot activities, festival activities, etc., where the activation activities are often pushed through channels such as sms, APP messages, official micro, micro-operations, etc., and the activation activities are combined with the pushing channels to form the activation scheme. In addition, the user activity record data may include business data, transaction data, payment data, etc. generated by the user in participating in the activation activity.

Step S202, determining the user type of the target user based on the user data, and determining a target recommendation model corresponding to the user type.

In this embodiment, the user type includes a hot start user or a cold start user. The above implementation of determining the user type of the target user based on the user data will be described in further detail in the following embodiments, which will not be described here. If the user type is a hot start user, the corresponding target recommendation model is a preset depth interest model; and if the user type is a cold start user, the corresponding target recommendation model is a preset double-tower model.

Step S203, extracting user feature data of the target user from the user data, and acquiring target user features from the user feature data based on a preset feature analysis model.

In this embodiment, the data corresponding to the user feature information type may be extracted from the user data, so as to extract the user feature data of the target user. The specific implementation process of acquiring the target user feature from the user feature data based on the preset feature analysis model will be described in further detail in the following specific embodiments, which will not be described herein.

Step S204, acquiring the activation scheme data of the activation scheme to be recommended, and extracting the characteristics of the activation scheme data to obtain the corresponding characteristics of the activation scheme.

In this embodiment, the above-mentioned to-be-recommended activation schedule is an activation schedule that is preliminarily sorted out, and the number of to-be-recommended activation schedules includes a plurality of to-be-recommended activation schedules.

Step S205, processing the target user feature and the activation proposal feature based on the target recommendation model, and generating a corresponding prediction result.

In this embodiment, if the user type is a hot start user, the target user feature and the activation scheme feature are processed by using a preset depth interest model to generate a corresponding prediction result; and if the user type is a cold start user, processing the target user characteristic and the activation proposal characteristic by using a preset double-tower model to generate a corresponding prediction result.

Step S206, determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to preset recommendation rules based on the prediction result.

In this embodiment, the specific implementation process of the recommended promotion scheme corresponding to the target user is determined from the to-be-recommended promotion schemes according to a preset recommendation rule based on the prediction result, and further details of this will be described in the following specific embodiments, which will not be described herein.

Firstly, acquiring user data of a target user; then determining the user type of the target user based on the user data, and determining a target recommendation model corresponding to the user type; then extracting user characteristic data of the target user from the user data, and acquiring target user characteristics from the user characteristic data based on a preset characteristic analysis model; subsequently acquiring activation scheme data of an activation scheme to be recommended, and extracting features of the activation scheme data to obtain corresponding activation scheme features; processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate corresponding prediction results; and finally, determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule based on the prediction result. According to the method and the system for recommending the activation promoting schemes, the corresponding user types are determined based on the user data of the target users, different recommendation models are selected for the target users of different user types to provide recommendation processing of the targeted activation promoting schemes for the target users, the adaptive activation promoting scheme recommendation for the different types of users is achieved, the recommended activation promoting scheme corresponding to the target users is intelligently determined from the to-be-recommended activation promoting schemes based on preset recommendation rules, accurate screening of the to-be-recommended activation promoting scheme is achieved, the final recommended activation promoting scheme for recommending the target users is determined, and accuracy of the activation promoting scheme recommendation is effectively improved.

In some alternative implementations, step S203 includes the steps of:

and calling the feature analysis model.

In this embodiment, the feature analysis model may specifically be generated after training the Xgboost model based on the training data collected in advance. Specifically, firstly, user characteristic data of a plurality of users are collected, and label labeling processing is carried out on the user characteristic data; the method comprises the steps of dividing users into active users and inactive users, wherein the active records caused by at least one activation scheme of the users are active users, the active labels are marked as 1, the active records caused by the fact that the users have no activation scheme are inactive users, and the active labels are marked as 0; and training the Xgboost model by using the training data, adjusting model parameters of the Xgboost model until the Xgboost model meeting the preset convergence condition is obtained, and taking the Xgboost model meeting the preset convergence condition as the characteristic analysis model.

Inputting the user characteristic data into the characteristic analysis model, and sorting the importance of the user characteristic data through the characteristic analysis model to obtain a corresponding sorting result.

In this embodiment, the above-mentioned importance ranking may refer to ranking all user feature data in order of importance values from high to low.

And acquiring the feature data with the highest importance from the sequencing result.

In this embodiment, the preset number of values is not limited in particular, and may be set according to actual service usage requirements. For example, topK may be selected from M user features as the activation key user feature, i.e., the target user feature described above.

And taking the characteristic data as the target user characteristic.

The application calls the feature analysis model; inputting the user characteristic data into the characteristic analysis model, and sorting the importance of the user characteristic data through the characteristic analysis model to obtain a corresponding sorting result; further, obtaining the feature data with the highest importance and the preset quantity from the sorting result; and taking the characteristic data as the target user characteristic. According to the application, the user activity factor analysis is performed based on the use of the feature analysis model to obtain the target user features, and compared with the traditional manual feature screening mode, the method is more scientific and strict, and the generation efficiency and the generation intelligence of the target user features are effectively improved. And the obtained target user characteristics are only required to be processed in the follow-up process to recommend the activation proposal to the target user, and all the obtained user characteristic data are not required to be processed, so that the processing workload of the electronic equipment is effectively reduced, and the processing efficiency and the processing intelligence of recommending the activation proposal to the target user are improved.

In some optional implementations of the present embodiment, the user type includes a hot start user or a cold start user; step S205 includes the steps of:

and if the target user is a cold start user, coding the target user characteristic by using a first coder in a preset double-tower model to obtain a corresponding user coding vector, and coding the activation scheme characteristic by using a second coder in the double-tower model to obtain a corresponding scheme coding vector.

In this embodiment, the above-described activation scheme features specifically include an activation scheme transmission channel, an activation scheme ID, an activation scheme classification (e.g., a coupon-like activity, a check-in-type activity, etc.). The double-tower model is a pre-constructed model comprising a first encoder, a second encoder, a deep neural network layer, a calculation layer and a prediction layer. The first encoder specifically refers to a User tower (User tower) in the double-tower model, the second encoder specifically refers to a solution tower (item tower) in the double-tower model, the deep neural network layer specifically refers to a DNN layer in the double-tower model, the calculation layer specifically refers to a similarity layer in the double-tower model, and the prediction layer specifically refers to a sigmoid layer in the double-tower model. The calculation layer is located after the first encoder and the second encoder, and the prediction layer is located after the calculation layer. In addition, the first encoder and the second encoder are used for encoding processing, so that the length consistency of the target user characteristics and the activation promoting scheme characteristics can be ensured, the subsequent similarity calculation of the target user characteristics and the activation promoting scheme characteristics is facilitated, and the activation promoting scheme recommendation of the target user is realized. In addition, the training generation process of the double-tower model is similar to the application process of the double-tower model, and is not described herein. Preferably, cross-Entropy is selected as a loss function, and based on pre-collected training data, an Adam optimizer is used for optimizing the loss function to train the double-tower model, and when the loss function tends to be stable and the accuracy of the model is not improved, training is stopped, so that a trained double-tower model is obtained.

And processing the user coding vector by using the deep neural network layer in the double-tower model to obtain a corresponding user vector, and processing the scheme coding vector to obtain a corresponding scheme vector.

And performing similarity calculation on the user coding vector and the scheme coding vector by using a calculation layer in the double-tower model, and outputting a corresponding similarity calculation result.

In this embodiment, in the above-mentioned calculation layer, similarity calculation is performed on the user vector and the scheme vector by using a cosine similarity algorithm, so as to output a corresponding similarity calculation result.

In this embodiment, in the prediction layer, the similarity calculation result may be predicted based on a sigmoid function, so as to output a first prediction result of the target user for the to-be-recommended activation solution.

When the target user is judged to be a cold start user, firstly, a first encoder in a preset double-tower model is utilized to encode the target user characteristic to obtain a corresponding user encoding vector, and a second encoder in the double-tower model is utilized to encode the activation scheme characteristic to obtain a corresponding scheme encoding vector; then, similarity calculation is carried out on the user vector and the scheme vector by utilizing a calculation layer in the double-tower model, and a corresponding similarity calculation result is output; and further, predicting the similarity calculation result by using a prediction layer in the double-tower model, and outputting a first prediction result of the target user for the to-be-recommended activation proposal. According to the method and the system, the corresponding double-tower model is intelligently selected for the cold start user to process the target user characteristics and the activation promoting scheme characteristics, so that the first prediction result of the target user for the to-be-recommended activation promoting scheme is quickly and accurately generated, the accurate recommendation of the activation promoting scheme of the target user is facilitated to be realized based on the obtained first prediction result, and the use experience of the target user is improved. In addition, the target user is characterized by selecting the target user characteristic from the user characteristic data of the target user as the activation key user characteristic, and the generated recommendation result is more accurate due to the fact that more comprehensive user information is contained in the target user characteristic.

In some alternative implementations, the user type includes a hot start user or a cold start user; step S205 includes the steps of:

and if the target user is a hot start user, acquiring historical activation scheme data corresponding to the target user.

In this embodiment, the user data of the target user further includes an activation solution push record of the target user, where the historical activation solution data refers to the activation solution push record, and the historical activation solution data may be obtained from the user data.

And extracting the characteristics of the historical activation scheme data to obtain corresponding historical activation scheme characteristics.

In this embodiment, the above-mentioned history activation scheme features include at least the activation scheme transmission channel, activation scheme ID, activation scheme classification, time interval from last activation scheme transmission, whether or not the user is caused to be active, and the like, which correspond to the history activation scheme associated with the target user.

In this embodiment, the activation program features at least include features of an activation program transmission channel, an activation program ID, an activation program classification, a time interval from the last activation program transmission, whether the user is caused to be active, and the like, which correspond to the activation program to be recommended. Specifically, the depth interest model is a pre-constructed DIN model comprising an attention network, a pooling and aggregation layer, a connection layer, an MLP layer and a sigmoid output layer. Calculating the relation weight of each historical activation scheme and the to-be-recommended activation scheme according to the fusion characteristics (comprising the historical activation scheme characteristics and the target user characteristics) of each historical activation scheme and the activation scheme characteristics of the to-be-recommended activation scheme by using the attention network; pooling the fusion characteristics of each history article and the corresponding relation weight by using a pooling and aggregation layer to obtain the history behavior characteristic data of the target user; connecting the historical behavior characteristic data of the target user with the activation proposal characteristic of the target article by using a connecting layer to obtain combined characteristic data; performing feature extraction on the combined feature data by using an MLP layer; and carrying out nonlinear mapping on the output of the MLP layer by utilizing a sigmoid output layer, and outputting a second prediction result of the target user for the to-be-recommended activation proposal. In addition, the training generation process of the depth interest model is similar to the application process of the depth interest model, and will not be described herein. Preferably, cross-Entropy is selected as a loss function, and based on pre-acquired training data, an Adam optimizer is used for optimizing the loss function to train the depth interest model, and when the loss function tends to be stable and the accuracy of the model is not improved, training is stopped, so that a trained depth interest model is obtained.

When the target user is judged to be a hot start user, firstly, historical activation proposal data corresponding to the target user is acquired; then, extracting features of the historical activation scheme data to obtain corresponding historical activation scheme features; and further inputting the target user characteristics, the promotion scheme characteristics and the historical promotion scheme characteristics into a preset depth interest model, and outputting a second prediction result of the target user on the promotion scheme to be recommended through the depth interest model. According to the method and the system, the corresponding deep interest model is intelligently selected for the hot start user to process the target user characteristics, the activation promoting scheme characteristics and the historical activation promoting scheme characteristics, so that the second prediction result of the target user for the to-be-recommended activation promoting scheme is quickly and accurately generated, the accurate recommendation of the activation promoting scheme of the target user is facilitated to be realized based on the obtained second prediction result, and the use experience of the target user is improved. In addition, the target user is characterized by selecting the target user characteristic from the user characteristic data of the target user as the activation key user characteristic, and the generated recommendation result is more accurate due to the fact that more comprehensive user information is contained in the target user characteristic.

In some alternative implementations, step S206 includes the steps of:

and screening target prediction results meeting preset conditions from the prediction results.

In this embodiment, the preset conditions are not specifically limited, and may be set according to actual service usage requirements. Specifically, the to-be-recommended promotion schemes can be ordered according to the sequence from the big to the small of the prediction result; the preset condition may refer to a target activation scheme of a target number of bits before sorting, or a target activation scheme of which the predicted result is greater than or equal to a preset threshold.

And acquiring a target activation scheme corresponding to the target prediction result from the activation scheme to be recommended.

And taking the target activation scheme as the recommended activation scheme.

The target prediction result meeting the preset condition is screened out from the prediction results; and further acquiring a target promotion scheme corresponding to the target prediction result from the promotion scheme to be recommended, and taking the target promotion scheme as the recommended promotion scheme. According to the method and the device for recommending the target user, based on the prediction result, the recommended promoting scheme corresponding to the target user is determined from the promoting schemes to be recommended according to the preset recommending rule, so that the promoting schemes to be recommended are accurately screened based on the prediction result, the final recommended promoting scheme for recommending the target user is determined, and the recommending accuracy of the promoting scheme is effectively improved.

In some alternative implementations of the present embodiment, step S202 includes the steps of:

and acquiring the activation scheme push record data of the target user from the user data.

In this embodiment, the data corresponding to the type of the activation scheme push record information may be extracted from the user data, so as to extract the activation scheme push record data of the target user.

And counting the number of the promotion scheme pushing records of the target user based on the promotion scheme pushing record data.

And determining the user type of the target user based on the number.

In this embodiment, the users may be classified into a cold start user and a hot start user according to the promotion scheme push record data, where the cold start user is only 1 promotion scheme push record user, and the hot start user is at least 2 promotion scheme push records.

Firstly, acquiring activation proposal pushing record data of the target user from the user data; then counting the number of the promotion scheme pushing records of the target user based on the promotion scheme pushing record data; and further determining a user type of the target user based on the number. According to the method and the device for determining the user type of the target user, the record data is pushed by the activation proposal in the user data and analyzed, so that the user type of the target user can be determined rapidly and accurately, and the confirmation efficiency of the user type is improved.

In some optional implementations of this embodiment, before step S206, the electronic device may further perform the following steps:

and acquiring the communication information of the target user.

In this embodiment, the communication information may include a mobile phone number or a mail address.

And acquiring the working time of the target user.

In this embodiment, the user data further includes work information of the target user, and the work information of the target user may be extracted from the user data of the target user, so as to analyze the work information to obtain the work time of the target user.

And determining the push time of the target user based on the working time.

In this embodiment, a general rest time is determined, and then a time other than the working time and the rest time in one day (24 h) is taken as the push time.

In this embodiment, when the current time is within the pushing time, the recommended activation promoting scheme is pushed to a terminal device corresponding to the communication information of the target user.

The application obtains the communication information of the target user; then, the working time of the target user is obtained; then determining the push time of the target user based on the working time; and pushing the recommended promotion scheme to the target user based on the pushing time and the communication information. According to the method and the device for promoting the target user to the recommended activation, when the recommended activation proposal of the target user is generated, the corresponding pushing time is further determined based on the working time of the target user, and the recommended activation proposal is pushed to the target user within the pushing time, so that the pushing of the recommended activation proposal does not disturb the target user, the pushing intelligence of the recommended activation proposal is improved, and the use experience of the target user is improved.

It is emphasized that the recommended promotion scheme may also be stored in a blockchain node in order to further ensure privacy and security of the recommended promotion scheme.

The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by computer readable instructions stored in a computer readable storage medium that, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

With further reference to fig. 3, as an implementation of the method shown in fig. 2, the present application provides an embodiment of an artificial intelligence-based scheme recommendation device, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 2, and the device may be specifically applied to various electronic devices.

As shown in fig. 3, the artificial intelligence based scheme recommendation device 300 according to the present embodiment includes: a judging module 301, a first determining module 302, a first obtaining module 303, a second determining module 304, a screening module 305, a second obtaining module 306 and a third determining module 307. Wherein:

A first obtaining module 301, configured to obtain user data of a target user;

a first determining module 302, configured to determine a user type of the target user based on the user data, and determine a target recommendation model corresponding to the user type;

a first extraction module 303, configured to extract user feature data of the target user from the user data, and obtain target user features from the user feature data based on a preset feature analysis model;

the second extraction module 304 is configured to obtain activation solution data of an activation solution to be recommended, and perform feature extraction on the activation solution data to obtain corresponding activation solution features;

a generating module 305, configured to process the target user feature and the activation solution feature based on the target recommendation model, and generate a corresponding prediction result;

and a second determining module 306, configured to determine, based on the prediction result, a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule.

In this embodiment, the operations performed by the modules or units respectively correspond to the steps of the method for recommending a solution based on artificial intelligence in the foregoing embodiment, which is not described herein again.

In some alternative implementations of the present embodiment, the first extraction module 303 includes:

the calling sub-module is used for calling the feature analysis model;

the sequencing sub-module is used for inputting the user characteristic data into the characteristic analysis model, and sequencing the importance of the user characteristic data through the characteristic analysis model to obtain a corresponding sequencing result;

the first acquisition sub-module is used for acquiring the preset number of characteristic data with highest importance from the sequencing result;

and the first determination submodule is used for taking the characteristic data as the target user characteristic.

In some optional implementations of the present embodiment, the user type includes a hot start user or a cold start user; the generation module 305 includes:

the first processing sub-module is used for coding the target user characteristics by using a first coder in a preset double-tower model to obtain corresponding user coding vectors if the target user is a cold start user, and coding the activation scheme characteristics by using a second coder in the double-tower model to obtain corresponding scheme coding vectors;

The second processing sub-module is used for processing the user coding vector by utilizing the deep neural network layer in the double-tower model to obtain a corresponding user vector, and processing the scheme coding vector to obtain a corresponding scheme vector;

the third processing sub-module is used for calculating the similarity between the user vector and the scheme vector by utilizing a calculation layer in the double-tower model and outputting a corresponding similarity calculation result;

and the third processing sub-module is used for carrying out prediction processing on the similarity calculation result by utilizing a prediction layer in the double-tower model and outputting a first prediction result of the target user on the to-be-recommended activation proposal.

In this embodiment, the operations performed by the modules or units respectively correspond to the steps of the scheme recommendation method based on artificial intelligence in the foregoing embodiment one by one, which is not described herein again.

the second acquisition sub-module is used for acquiring historical activation scheme data corresponding to the target user if the target user is a hot start user;

The extraction submodule is used for extracting the characteristics of the historical activation scheme data to obtain corresponding historical activation scheme characteristics;

and the fourth processing submodule is used for inputting the target user characteristics, the promotion scheme characteristics and the historical promotion scheme characteristics into a preset depth interest model, and outputting a second prediction result of the target user on the promotion scheme to be recommended through the depth interest model.

In some alternative implementations of the present embodiment, the second determining module 306 includes:

the screening sub-module is used for screening target prediction results meeting preset conditions from the prediction results;

a third obtaining sub-module, configured to obtain a target activation solution corresponding to the target prediction result from the to-be-recommended activation solution;

a second determination sub-module for regarding the target promotion program as the recommended promotion program.

In some optional implementations of this embodiment, the first determining module 302 includes:

a fourth obtaining sub-module, configured to obtain, from the user data, activation solution push record data of the target user;

a statistics sub-module for counting the number of the promotion scheme pushing records of the target user based on the promotion scheme pushing record data;

and a third determination sub-module that determines a user type of the target user based on the number.

In some optional implementations of this embodiment, the artificial intelligence based scheme recommendation device further includes:

the second acquisition module is used for acquiring the communication information of the target user;

a third obtaining module, configured to obtain a working time of the target user;

a third determining module, configured to determine a push time of the target user based on the working time;

and the pushing module is used for pushing the recommended promotion scheme to the target user based on the pushing time and the communication information.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 4, fig. 4 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It should be noted that only computer device 4 having components 41-43 is shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 41 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 4. Of course, the memory 41 may also comprise both an internal memory unit of the computer device 4 and an external memory device. In this embodiment, the memory 41 is typically used to store an operating system and various application software installed on the computer device 4, such as computer readable instructions of an artificial intelligence based scheme recommendation method. Further, the memory 41 may be used to temporarily store various types of data that have been output or are to be output.

The processor 42 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions stored in the memory 41 or process data, such as executing computer readable instructions of the artificial intelligence based solution recommendation method.

The network interface 43 may comprise a wireless network interface or a wired network interface, which network interface 43 is typically used for establishing a communication connection between the computer device 4 and other electronic devices.

in the embodiment of the application, firstly, user data of a target user is acquired; then determining the user type of the target user based on the user data, and determining a target recommendation model corresponding to the user type; then extracting user characteristic data of the target user from the user data, and acquiring target user characteristics from the user characteristic data based on a preset characteristic analysis model; subsequently acquiring activation scheme data of an activation scheme to be recommended, and extracting features of the activation scheme data to obtain corresponding activation scheme features; processing the target user characteristics and the activation proposal characteristics based on the target recommendation model to generate corresponding prediction results; and finally, determining a recommended promotion scheme corresponding to the target user from the promotion schemes to be recommended according to a preset recommendation rule based on the prediction result. According to the embodiment of the application, the corresponding user type is determined based on the user data of the target user, different recommendation models are selected for the target user of different user types to provide recommendation processing of the targeted activation scheme for the target user, so that the adaptive activation scheme recommendation for different types of users is realized, the recommended activation scheme corresponding to the target user is intelligently determined from the to-be-recommended activation scheme based on the preset recommendation rule, the accurate screening of the to-be-recommended activation scheme is realized to determine the final recommended activation scheme for recommending to the target user, and the accuracy of the activation scheme recommendation is effectively improved.

The present application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the artificial intelligence based solution recommendation method as described above.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

It is apparent that the above-described embodiments are only some embodiments of the present application, but not all embodiments, and the preferred embodiments of the present application are shown in the drawings, which do not limit the scope of the patent claims. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a thorough and complete understanding of the present disclosure. Although the application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing description, or equivalents may be substituted for elements thereof. All equivalent structures made by the content of the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the scope of the application.

Claims

1. An artificial intelligence based scheme recommendation method is characterized by comprising the following steps:

acquiring user data of a target user;

2. The method for recommending a solution based on artificial intelligence according to claim 1, wherein the step of obtaining the target user feature from the user feature data based on a preset feature analysis model specifically comprises:

Invoking the feature analysis model;

and taking the characteristic data as the target user characteristic.

3. The artificial intelligence based scheme recommendation method according to claim 1, wherein the user type includes a hot start user or a cold start user; the step of processing the target user feature and the activation proposal feature based on the target recommendation model to generate a corresponding prediction result specifically comprises the following steps:

4. The artificial intelligence based scheme recommendation method according to claim 3, wherein the user type includes a hot start user or a cold start user; the step of processing the target user feature and the activation proposal feature based on the target recommendation model to generate a corresponding prediction result specifically comprises the following steps:

5. The method for recommending a solution based on artificial intelligence according to claim 1, wherein the step of determining a recommended activation solution corresponding to the target user from the to-be-recommended activation solutions according to a preset recommendation rule based on the prediction result specifically comprises:

and taking the target activation scheme as the recommended activation scheme.

6. The method for recommending a solution based on artificial intelligence according to claim 1, wherein the step of determining the user type of the target user based on the user data specifically comprises:

and determining the user type of the target user based on the number.

7. The method according to claim 1, further comprising, after the step of determining a recommended promotion program corresponding to the target user from the promotion programs to be recommended according to a preset recommendation rule based on the prediction result:

Acquiring communication information of the target user;

acquiring the working time of the target user;

determining push time of the target user based on the working time;

8. An artificial intelligence based scheme recommendation device, comprising:

the first acquisition module is used for acquiring user data of a target user;

9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the artificial intelligence based solution recommendation method of any of claims 1 to 7.

10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the artificial intelligence based solution recommendation method according to any of claims 1 to 7.