CN118013128A

CN118013128A - Material recommendation method, device, equipment and storage medium based on artificial intelligence

Info

Publication number: CN118013128A
Application number: CN202410355552.8A
Authority: CN
Inventors: 顾聪聪
Original assignee: Ping An Health Insurance Company of China Ltd
Current assignee: Ping An Health Insurance Company of China Ltd
Priority date: 2024-03-26
Filing date: 2024-03-26
Publication date: 2024-05-10

Abstract

The application belongs to the field of artificial intelligence and the field of financial science and technology, and relates to a material recommendation method based on artificial intelligence, which comprises the following steps: receiving a recommendation request triggered by a user; acquiring target user data of a user based on a target dimension; generating a target crowd label of a user based on a first service system; inquiring materials to be recommended corresponding to the target crowd labels; analyzing the target user data and the materials to be recommended based on the recommendation model, and determining the recommended materials from the materials to be recommended; filtering the recommended materials based on a recommendation strategy to obtain target recommended materials; and pushing the target recommended materials to the user. The application also provides a material recommending device, computer equipment and a storage medium based on the artificial intelligence. In addition, the application also relates to a blockchain technology, and target recommendation materials can be stored in the blockchain. The method and the device can be applied to the material recommendation scene in the financial field, finish the accurate material recommendation of the user, and effectively improve the accuracy and the intelligence of the material recommendation.

Description

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

Technical Field

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

Background

At present, with the high-speed development of the Internet industry, the daily life behavior mode of people is gradually changed, and the development of daily operation and back technology of the industry is further driven. Before the internet is raised, the company business of insurance companies typically interacts with people by means of televisions, newspapers, magazines, etc. The interaction has the conditions of low efficiency, poor hit rate, resource waste and the like, and the expected effect is often not achieved. With the rapid development of the internet field, people gradually throw their eyes towards terminals on the subscriber line, such as: the mobile phone, the tablet, the computer and the like perform relevant business development through various software embedded into various terminals, thereby becoming a main matrix of marketing of various businesses.

At present, the recommended treatment for the insurance materials in the insurance field is often a split and multi-type manner, such as: when opening the insurance APP, there is a need to conduct simultaneous activities on insurance products, service products, content articles, and coupons. Because the inlets of the APP pages, the source contacts and the like often operate independently, different strategies and the like are not taken into consideration by the background of each strategy. There are situations where the user is not interested in the recommended insurance material and the subsequent conversion rate is low. Therefore, the existing insurance material recommendation method has the problem of low recommendation accuracy.

Disclosure of Invention

The embodiment of the application aims to provide an artificial intelligence-based material recommendation method, an artificial intelligence-based material recommendation device, computer equipment and a storage medium, so as to solve the technical problem of low recommendation accuracy in the existing insurance material recommendation mode.

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

receiving a recommendation request triggered by a user;

acquiring target user data corresponding to the user based on a preset target dimension;

Generating a target crowd label corresponding to the user based on a preset first service system;

Inquiring materials to be recommended corresponding to the target crowd labels from a preset second service system;

analyzing and processing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended; the recommendation model is generated by training XGBOOST models based on sample data acquired in advance, a first preset index and a second preset index;

filtering the recommended materials based on a preset recommended strategy to obtain corresponding target recommended materials;

And pushing the target recommended materials to the user.

Further, the step of acquiring target user data corresponding to the user based on a preset target dimension specifically includes:

acquiring user information of the user;

acquiring initial user data corresponding to the user based on the user information;

screening the initial user data based on the target dimension to obtain specified data corresponding to the target dimension;

And taking the specified data as the target user data.

Further, the step of querying the material to be recommended corresponding to the target crowd label from the preset second service system specifically includes:

acquiring source contact position information and channel information corresponding to the recommendation request;

Invoking the second service system;

acquiring first materials matched with the source contact position information and the channel information from all initial materials contained in the second service system;

Screening second materials corresponding to the target crowd labels from the first materials;

and taking the second material as the material to be recommended.

Further, the step of analyzing the target user data and the material to be recommended based on the preset recommendation model and determining the recommended material corresponding to the target user data from the material to be recommended specifically includes:

inputting the target user data and the materials to be recommended into the recommendation model;

analyzing and processing the user data and the materials to be recommended through the recommendation model to obtain the value of the degree of interest of the user on each material to be recommended;

screening out appointed interested degree values larger than a preset threshold value from all the interested degree values;

acquiring a specified material corresponding to the specified interest degree value from the material to be recommended;

and taking the appointed material as the recommended material.

Further, the step of filtering the recommended materials based on a preset recommended policy to obtain corresponding target recommended materials specifically includes:

acquiring the content of the issued material corresponding to the user in a preset time period;

judging whether a third material with a matching relation with the content of the issued material exists in the recommended material or not;

if yes, filtering the third material from the recommended material to obtain a processed fourth material;

and taking the fourth material as the target recommended material.

Further, the step of pushing the target recommended material to the user specifically includes:

arranging all the target recommended materials in descending order according to the value of the degree of interest to obtain ordered target recommended materials;

filling the ordered target recommendation materials into a preset initial list to obtain a corresponding recommendation list;

And displaying the recommendation list to finish recommendation processing of pushing the target recommendation materials to the user.

Further, before the step of analyzing the target user data and the material to be recommended based on the preset recommendation model and determining the recommended material corresponding to the target user data from the material to be recommended, the method further includes:

acquiring pre-acquired historical user data;

Preprocessing the historical user data to obtain corresponding sample data;

Dividing the sample data into a training set and a testing set;

Training the XGBOOST model by using the training set to obtain a trained initial model;

based on the first preset index and the second preset index, testing the initial model by using the test set to obtain a specified model meeting the index condition requirements corresponding to the first preset index and the second preset index;

and taking the specified model as the recommendation model.

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

The receiving module is used for receiving a recommendation request triggered by a user;

The first acquisition module is used for acquiring target user data corresponding to the user based on a preset target dimension;

the generation module is used for generating a target crowd label corresponding to the user based on a preset first service system;

The query module is used for querying materials to be recommended corresponding to the target crowd labels from a preset second service system;

The first processing module is used for analyzing and processing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended; the recommendation model is generated by training XGBOOST models based on sample data acquired in advance, a first preset index and a second preset index;

The second processing module is used for filtering the recommended materials based on a preset recommended strategy to obtain corresponding target recommended materials;

And the pushing module is used for pushing the target recommended materials to the user.

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:

receiving a recommendation request triggered by a user;

And pushing the target recommended materials to the 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:

receiving a recommendation request triggered by a user;

And pushing the target recommended materials to the user.

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

After receiving a recommendation request triggered by a user, the embodiment of the application firstly acquires target user data corresponding to the user based on a preset target dimension; then generating a target crowd label corresponding to the user based on a preset first service system; inquiring materials to be recommended corresponding to the target crowd labels from a preset second service system; analyzing and processing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended; filtering the recommended materials based on a preset recommended strategy to obtain corresponding target recommended materials; and finally, pushing the target recommended materials to the user. According to the method, the device and the system, the target user data corresponding to the user are obtained based on the target dimension, the material to be recommended corresponding to the target crowd label of the user is determined through the use of the service system, the recommendation material corresponding to the target user data is determined from the material to be recommended based on the use of the preset recommendation model, the recommendation material is filtered based on the preset recommendation strategy to obtain the target recommendation material, and the target recommendation material is pushed to the user, so that the recommendation material conforming to the user interests is recommended based on the use of the target dimension, the service system, the recommendation model and the recommendation strategy, the accurate material recommendation for the user is completed, the accuracy and the intelligence of the material recommendation are effectively improved, and the use experience of the user is facilitated to be 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 material recommendation method in accordance with the present application;

FIG. 3 is a schematic diagram of one embodiment of an artificial intelligence based material recommendation device 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 (Mov i ng P i cture Experts G roup Aud i o Layer I I I, dynamic video expert compression standard audio plane 3), MP4 (Mov i ng P i ctu re Experts G roup Aud i o Layer I V, 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 material recommending method based on artificial intelligence provided by the embodiment of the application is generally executed by a server/terminal device, and correspondingly, the material recommending device based on artificial intelligence is generally arranged 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 flowchart of one embodiment of an artificial intelligence based material 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 material recommending method based on the artificial intelligence provided by the embodiment of the application can be applied to any scene needing material recommending, and can be applied to products of the scenes, such as insurance product material recommending in the field of financial insurance. The artificial intelligence-based material recommendation method comprises the following steps:

step S201, a recommendation request triggered by a user is received.

In this embodiment, the electronic device (for example, the server/terminal device shown in fig. 1) on which the material recommendation method based on artificial intelligence operates may acquire the recommendation request 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 connections, wifi connections, bluetooth connections, wimax connections, Z i gbee connections, UWB (u l tra W i deband) connections, and other now known or later developed wireless connection means. The execution subject of the artificial intelligence-based material recommendation method may be a recommendation system. Under the service scene of insurance product recommendation in the field of financial insurance, the user can be specifically an insurance client in different life cycles, such as new client, current renewal, participation activity and the like. The recommendation request is a request triggered by a user for requesting the recommendation system to make a recommendation of the adapted insurance product material to the user. The recommendation request may further carry user information of the user, where the user information may include name information or identity code of the user.

Step S202, acquiring target user data corresponding to the user based on a preset target dimension.

In this embodiment, the specific implementation process of acquiring the target user data corresponding to the user based on the preset target dimension is described in further detail in the following specific embodiments, which will not be described herein.

Step S203, generating a target crowd label corresponding to the user based on a preset first service system.

In this embodiment, the first service system may be a member system that is pre-built and stores crowd labels of a plurality of insurance clients. The target crowd label corresponding to the user can be generated from the disfigurement by acquiring the user information of the user and further judging the life cycle of the first service system based on the user information. The target crowd label can comprise any one of labels of new guests, current renewal, participation in activities and the like.

Step S204, inquiring the materials to be recommended corresponding to the target crowd labels from a preset second service system.

In this embodiment, in the service scenario of the insurance material recommendation, the material to be recommended may at least include an insurance product, a service product, a medical care product, a content article, a short video, a card, an article, a live broadcast, and so on. The specific implementation process of querying the material to be recommended corresponding to the target crowd label from the preset second service system will be described in further detail in the following specific embodiments, which are not described herein.

Step S205, analyzing and processing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended; the recommendation model is generated by training XGBOOST models based on pre-collected sample data, a first preset index and a second preset index.

In this embodiment, the foregoing analysis processing is performed on the target user data and the material to be recommended based on the preset recommendation model, and a specific implementation process of the recommended material corresponding to the target user data is determined from the material to be recommended.

Step S206, filtering the recommended materials based on a preset recommended strategy to obtain corresponding target recommended materials.

In this embodiment, the filtering process is performed on the recommended materials based on the preset recommendation policy to obtain a specific implementation process of the corresponding target recommended materials, which will be described in further detail in the following specific embodiments, which are not described herein.

Step S207, pushing the target recommended materials to the user.

In this embodiment, the foregoing specific implementation process of pushing the target recommended material to the user will be described in further detail in the following specific embodiments, which will not be described herein. The present application will be described in further detail in the following embodiments, which will not be explained in any greater detail herein.

After receiving a recommendation request triggered by a user, acquiring target user data corresponding to the user based on a preset target dimension; then generating a target crowd label corresponding to the user based on a preset first service system; inquiring materials to be recommended corresponding to the target crowd labels from a preset second service system; analyzing and processing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended; filtering the recommended materials based on a preset recommended strategy to obtain corresponding target recommended materials; and finally, pushing the target recommended materials to the user. According to the method, the device and the system, the target user data corresponding to the user are obtained based on the target dimension, the material to be recommended corresponding to the target crowd label of the user is determined through the use of the service system, the recommendation material corresponding to the target user data is determined from the material to be recommended based on the use of the preset recommendation model, the recommendation material is filtered based on the preset recommendation strategy to obtain the target recommendation material, and the target recommendation material is pushed to the user, so that the recommendation material conforming to the user interests is recommended based on the use of the target dimension, the service system, the recommendation model and the recommendation strategy, the accurate material recommendation for the user is completed, the accuracy and the intelligence of the material recommendation are effectively improved, and the use experience of the user is facilitated to be improved.

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

And acquiring user information of the user.

In this embodiment, the user information may include name information or identity code information that characterizes the identity of the user.

And acquiring initial user data corresponding to the user based on the user information.

In this embodiment, the user information is used to query a preset big data portrait system, so as to query initial user data corresponding to the user information. The big data portrait system is a pre-built database storing user service data of a plurality of users. The initial user data may include at least user basic information, user behavior information, user transaction information, user claim information, user interest characteristics, and the like.

And screening the initial user data based on the target dimension to obtain specified data corresponding to the target dimension.

In this embodiment, the target dimension may specifically include dimensions corresponding to user basic information, user behavior information, user transaction information, user claim information, and user interest features.

And taking the specified data as the target user data.

The application obtains the user information of the user; then, initial user data corresponding to the user is obtained based on the user information; then screening the initial user data based on the target dimension to obtain specified data corresponding to the target dimension; the specified data is then used as the target user data. According to the method and the device for acquiring the target user data, the initial user data corresponding to the user is acquired according to the user information of the user, and then the initial user data is screened based on the target dimension, so that the target user data corresponding to the target dimension can be acquired rapidly and accurately, the acquisition efficiency of the target user data is improved, and the accuracy of the generated target user data is ensured.

In some alternative implementations of the present embodiment, step S204 includes the steps of:

and acquiring source contact position information and channel information corresponding to the recommendation request.

In this embodiment, the source contact position and the channel of the recommendation request triggered by the user may be extracted to obtain the source contact position information and the channel information corresponding to the recommendation request. The channels may include at least on-line channels, life channels, third party channels, and the like. In the field of C-terminal insurance, because products are subjected to the actual problems of powerful supervision of a supervision department, profit distribution of different channels and the like, recommendation strategies of different source contacts and the like, materials to be recommended displayed by different user groups, different channels, different source contact positions and the like are different.

And calling the second service system.

In this embodiment, the second service system is a mall center background system that is built in advance and stores a plurality of lists of materials for performing recommendation processing. And in the second service system, dividing the materials according to the source contact positions, channels and adaptive crowd labels of the materials in the list in advance so as to facilitate the accurate searching of the follow-up materials.

And acquiring first materials matched with the source contact position information and the channel information from all initial materials contained in the second service system.

In this embodiment, the material screening may be performed on all the initial materials included in the second service system according to the source contact position information and the channel information, so as to obtain the corresponding first material.

And screening second materials corresponding to the target crowd labels from the first materials.

In this embodiment, the first material may be screened by using the target crowd label to obtain a matched second material.

And taking the second material as the material to be recommended.

The method comprises the steps of obtaining source contact position information and channel information corresponding to the recommendation request; then calling the second service system; then, first materials matched with the source contact position information and the channel information are obtained from all initial materials contained in the second service system; subsequently, second materials corresponding to the target crowd labels are screened out from the first materials; and finally, taking the second material as the material to be recommended. According to the method and the device for inquiring the second business system, the source contact position information and the channel information corresponding to the recommendation request are used, and the target crowd label of the user is used for inquiring the second business system, so that the materials to be recommended, which are matched with the user, can be quickly and accurately inquired, the acquisition efficiency of the materials to be recommended is improved, and the accuracy of the obtained materials to be recommended is ensured.

In some alternative implementations, step S205 includes the steps of:

And inputting the target user data and the materials to be recommended into the recommendation model.

In this embodiment, the target user data and the material to be recommended may be first converted into a data format corresponding to a model processing, and then the converted target user data and the material to be recommended may be input into the recommendation model.

And analyzing and processing the user data and the materials to be recommended through the recommendation model to obtain the value of the degree of interest of the user on each material to be recommended.

In this embodiment, after the recommendation model analyzes and processes the user data and the materials to be recommended, the value of the degree of interest of the user for each material to be recommended is output.

And screening out the appointed interested degree values larger than a preset threshold value from all the interested degree values.

In this embodiment, the value of the preset threshold is not specifically limited, and may be set according to the actual service usage requirement.

And acquiring the appointed material corresponding to the appointed interesting degree value from the material to be recommended.

In this embodiment, the specified material is a material having an association relationship with the specified interest level value in the materials to be recommended.

And taking the appointed material as the recommended material.

Inputting the target user data and the materials to be recommended into the recommendation model; then analyzing and processing the user data and the materials to be recommended through the recommendation model to obtain the value of the degree of interest of the user on each material to be recommended; then screening out appointed interested degree values larger than a preset threshold value from all the interested degree values; and acquiring the appointed material corresponding to the appointed interest degree value from the material to be recommended, and taking the appointed material as the recommended material. According to the method and the device for recommending the material, the recommendation model is used for analyzing and processing the target user data and the material to be recommended, so that the recommended material which meets the user interest can be rapidly and accurately output, and the accuracy and the intelligence of material recommendation are effectively improved.

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

and acquiring the content of the issued material corresponding to the user in a preset time period.

In this embodiment, the value of the preset time period is not specifically limited, and may be set according to the actual use requirement, for example, may be set within the first half year from the current time.

And judging whether a third material with a matching relation with the content of the issued material exists in the recommended material.

In this embodiment, the matching processing may be performed by using the content of the delivered material and all the content of the material included in the recommended material, so as to determine whether a third material having a matching relationship with the content of the delivered material exists in the recommended material.

And if yes, filtering the third material from the recommended material to obtain a processed fourth material.

And taking the fourth material as the target recommended material.

The method comprises the steps of obtaining the content of the issued material corresponding to the user in a preset time period; then judging whether a third material with a matching relation with the content of the issued material exists in the recommended material or not; if yes, filtering the third material from the recommended material to obtain a processed fourth material; and taking the fourth material as the target recommended material. According to the method and the device for processing the target user data, after the recommended materials corresponding to the target user data are determined based on the use of the recommended model, whether the recommended materials exist third materials which have a matching relation with the content of the issued materials corresponding to the user in a preset time period or not is intelligently detected, if so, the third materials are filtered from the recommended materials, and the processed fourth materials are used as final target recommended materials, so that the situation of repeated pushing of the materials can be avoided, the accuracy and the intelligence of pushing of the materials are improved, and the use experience of the user is improved.

In some alternative implementations of the present embodiment, step S207 includes the steps of:

and arranging all the target recommended materials in descending order according to the value of the degree of interest, and obtaining the ordered target recommended materials.

And filling the ordered target recommendation materials into a preset initial list to obtain a corresponding recommendation list.

In this embodiment, the initial list may be a pre-constructed list template for performing material recommendation processing, and the content of the list template is not specifically limited, and may be written according to actual service usage requirements.

In this embodiment, the generated recommendation list may be displayed on the material display interface by determining a preset material display interface. The interface selection of the material display interface can be set according to actual use requirements.

According to the method, all the target recommended materials are arranged in descending order according to the value of the degree of interest, so that the ordered target recommended materials are obtained; filling the ordered target recommendation materials into a preset initial list to obtain a corresponding recommendation list; and displaying the recommendation list to finish recommendation processing of pushing the target recommendation materials to the user. According to the method, all the target recommended materials are arranged in descending order according to the value of the degree of interest, the ordered target recommended materials are filled into the preset initial list to obtain the corresponding recommended list, and the recommended list is displayed, so that the recommendation processing of pushing the target recommended materials to the user is completed, the recommended materials conforming to the interests of the user are pushed to the user based on the display mode of the recommended list, and the accuracy and the intelligence of material recommendation are effectively improved.

In some optional implementations of this embodiment, before step S205, the electronic device may further perform the following steps:

Acquiring pre-acquired historical user data.

In this embodiment, the historical user data may be user data of an insurance client in a previously collected historical period. The user data of the insurer may include data of the insurer's user base information, user behavior information, user transaction information, user claim information, user interest features, etc. over a historical period of time. In order to ensure more accurate material recommendation for users, the active users in the historical user data are cut off behavior data and random sampling of labels is performed, so that the influence of the active users on subsequent sample data processing is avoided. In addition, through analyzing and modeling the data of a plurality of dimensions of the user, the matching degree of the user to the current material can be accurately judged, and technical support is provided for follow-up user preservation, clicking, conversion and other fine operation. In addition, the data used in the application is based on the transaction information of full-flow automation, a real-time data summarization mechanism is created, and data support is provided for model reasoning of a follow-up online recommendation model.

And preprocessing the historical user data to obtain corresponding sample data.

In this embodiment, the preprocessing may include data cleaning, data interpolation, normalization, and the like.

The sample data is divided into a training set and a testing set.

In this embodiment, the sample data may be divided into a training set and a testing set according to a preset dividing value. The dividing value is not limited, and for example, 7:3 may be used.

And training the XGBOOST model by using the training set to obtain a trained initial model.

In this embodiment, the training process of the initial model by the training set may refer to the training process of the XGBOOST model, which is not described in detail herein. The method can support operators to analyze online user data at the same time. The XGBOOST model can calculate the importance of each feature, and the featu re _ i mportance is used for obtaining the contribution degree of the feature factors, so that operators can carry out operation strategy formulation, advertising material conception and the like according to the related feature importance, the online exposure flow is utilized to the maximum, the follow-up refined operation is realized, the user satisfaction degree is improved, and the greater value is promoted. In addition, XGBOOST supports the GPU to calculate, and the powerful calculation capability of the GPU is utilized to rapidly infer, so that the effects of high response and low delay on the line are met.

And based on the first preset index and the second preset index, testing the initial model by using the test set to obtain a specified model meeting the index condition requirements corresponding to the first preset index and the second preset index.

In this embodiment, the first preset index and the second preset index may be model evaluation indexes GAUC and Nove l-Reca l l@10 index. Wherein GAUC (group auc) is to calculate auc for each user and then weight average GAUC, which can reduce the influence of the sorting result among different users. In addition, in order to calculate GAUC more in line with the distribution of the online users, the application can further remove the data of all positive samples and all negative samples of the online users (including tourists) when calculating GAUC, so that the model effect can more conform to the recommendation result of the online users. In addition, to highlight the novelty of the recommendation system in the recommendation results, nove l-Reca l l@50: for a user u, pu (|Pu|50) is defined as a future interest commodity prediction set of the user u produced by the player, gu is defined as a real future interest commodity set of the user u, hu is defined as a commodity subset of historical behavior categories of the user u, specifically, hu is a subset of Gu, and for each commodity in Hu, category IDs belong to categories which the user has historically performed. At this time, the Nove l-Reca l l@50 index of the user is calculated by the following methodIn addition, the above-mentioned index threshold corresponding to the index condition requirement corresponding to the first preset index and the second preset index is not specifically limited, and may be set according to the actual model building service requirement. According to the embodiment of the application, based on the XGBoost model, corresponding recommendation is performed by utilizing user data with multiple dimensions, and in order to better accord with online user ordering, the abnormal data is removed by adopting GAUC to replace AUC, so that the recommendation model is better optimized. Meanwhile, due to the fact that recommended contents with multiple dimensions exist, the condition of recommended homogenization can be effectively avoided by assisting in adopting Nove l-Reca l l@10 as a novel index.

And taking the specified model as the recommendation model.

The method comprises the steps of acquiring pre-acquired historical user data; preprocessing the historical user data to obtain corresponding sample data; dividing the sample data into a training set and a testing set; training the XGBOOST model by using the training set to obtain a trained initial model; and finally, based on the first preset index and the second preset index, testing the initial model by using the test set to obtain a specified model meeting the index condition requirements corresponding to the first preset index and the second preset index, and taking the specified model as the recommended model. According to the application, the sample data is obtained by preprocessing the pre-collected historical user data, then the sample data is divided into the training set and the testing set, then the training set is used for training the XGBOOST model to obtain a trained initial model, and then the testing set is used for testing the initial model to obtain a specified model meeting the index condition requirements corresponding to the first preset index and the second preset index and serving as a final recommended model, so that the model construction process of purchasing the recommended model is completed, the model effect and the prediction accuracy of the generated recommended model are effectively ensured, and the construction efficiency of the recommended model is improved.

It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.

It should be emphasized that, to further ensure the privacy and security of the target recommended material, the target recommended material may also be stored in a node of a blockchain.

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 (B l ockcha i n), essentially a de-centralized database, is a string of data blocks that are generated in association using cryptographic methods, each of which contains information from a batch of network transactions for verifying the validity (anti-counterfeit) of its information 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 (ART I F I C I A L I NTE L L I GENCE, A I) is the theory, method, technique, and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend, and expand 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-On-y 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 material recommendation apparatus, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be specifically applied to various electronic devices.

As shown in fig. 3, the artificial intelligence based material recommendation apparatus 300 according to the present embodiment includes: a receiving module 301, a first obtaining module 302, a generating module 303, a querying module 304, a first processing module 305, a second processing module 306 and a pushing module 307. Wherein:

a receiving module 301, configured to receive a recommendation request triggered by a user;

A first obtaining module 302, configured to obtain target user data corresponding to the user based on a preset target dimension;

a generating module 303, configured to generate a target crowd label corresponding to the user based on a preset first service system;

The query module 304 is configured to query, from a preset second service system, materials to be recommended corresponding to the target crowd label;

The first processing module 305 is configured to perform analysis processing on the target user data and the material to be recommended based on a preset recommendation model, and determine a recommendation material corresponding to the target user data from the material to be recommended; the recommendation model is generated by training XGBOOST models based on sample data acquired in advance, a first preset index and a second preset index;

The second processing module 306 is configured to perform filtering processing on the recommended material based on a preset recommendation policy, so as to obtain a corresponding target recommended material;

and the pushing module 307 is configured to push the target recommended material to the user.

In this embodiment, the operations performed by the modules or units are respectively corresponding to the steps of the artificial intelligence-based material recommendation method in the foregoing embodiment, and are not described herein again.

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

The first acquisition sub-module is used for acquiring the user information of the user;

The second acquisition sub-module is used for acquiring initial user data corresponding to the user based on the user information;

the first screening submodule is used for screening the initial user data based on the target dimension to obtain specified data corresponding to the target dimension;

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

In some alternative implementations of the present embodiment, the query module 304 includes:

the third acquisition sub-module is used for acquiring source contact position information and channel information corresponding to the recommendation request;

a calling sub-module for calling the second service system;

a fourth obtaining sub-module, configured to obtain, from all initial materials included in the second service system, a first material that matches both the source contact location information and the channel information;

The second screening submodule is used for screening second materials corresponding to the target crowd labels from the first materials;

and the second determining submodule is used for taking the second material as the material to be recommended.

In this embodiment, the operations performed by the modules or units are respectively corresponding to the steps of the artificial intelligence-based material recommendation method in the foregoing embodiment, which is not described herein again.

In some alternative implementations of the present embodiment, the first processing module 305 includes:

the input sub-module is used for inputting the target user data and the materials to be recommended into the recommendation model;

the processing sub-module is used for analyzing and processing the user data and the materials to be recommended through the recommendation model to obtain the value of the degree of interest of the user on each material to be recommended;

A third screening sub-module, configured to screen out a specified interest degree value greater than a preset threshold from all the interest degree values;

a fifth obtaining sub-module, configured to obtain a specified material corresponding to the specified interest degree value from the material to be recommended;

and the third determination submodule is used for taking the appointed material as the recommended material.

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

a sixth acquisition sub-module, configured to acquire the content of the delivered material corresponding to the user in a preset time period;

the judging submodule is used for judging whether a third material with a matching relation with the content of the issued material exists in the recommended material or not;

the filtering sub-module is used for filtering the third material from the recommended material if so, and obtaining a processed fourth material;

and the fourth determining submodule is used for taking the fourth material as the target recommended material.

In some optional implementations of this embodiment, the pushing module 307 includes:

The sequencing sub-module is used for sequencing all the target recommended materials in descending order according to the value of the degree of interest to obtain sequenced target recommended materials;

The filling sub-module is used for filling the ordered target recommended materials into a preset initial list to obtain a corresponding recommended list;

and the display sub-module is used for displaying the recommendation list so as to finish the recommendation process of pushing the target recommendation materials to the user.

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

the second acquisition module is used for acquiring the pre-acquired historical user data;

The preprocessing module is used for preprocessing the historical user data to obtain corresponding sample data;

the dividing module is used for dividing the sample data into a training set and a testing set;

The training module is used for training the XGBOOST models by using the training set to obtain a trained initial model;

The test module is used for carrying out test processing on the initial model by using the test set based on the first preset index and the second preset index so as to obtain a specified model which meets the index condition requirements corresponding to the first preset index and the second preset index;

And the determining module is used for taking the specified model as the recommendation model.

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 calculation and/or information processing according to preset or stored instructions, and the hardware thereof includes, but is not limited to, a microprocessor, an application specific integrated circuit (APP L I CAT I on SPEC I F I C I NTEGRATED C I rcu I t, AS IC), a programmable gate array (Fie l d-Programmab L E GATE AR RAY, FPGA), a digital Processor (D I G I TA L S I GNA L Processor, DSP), an embedded device, and the like.

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 memory card (SMART MED I A CARD, SMC), a secure digital (Secu RE D I G I TA L, SD) card, a flash memory card (F L ASH 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 material 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 (Cent ra lProcess i ng Un i t, 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 material 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.

According to the embodiment of the application, the target user data corresponding to the user is obtained based on the target dimension, then the material to be recommended corresponding to the target crowd label of the user is determined through the use of the service system, the recommended material corresponding to the target user data is determined from the material to be recommended based on the use of the preset recommendation model, the recommended material is filtered based on the preset recommendation strategy to obtain the target recommended material, and the target recommended material is pushed to the user, so that the recommended material conforming to the user interest is recommended based on the use of the target dimension, the service system, the recommendation model and the recommendation strategy, the accurate material recommendation of the user is completed, the accuracy and the intelligence of the material recommendation are effectively improved, and the use experience of the user is facilitated to be 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 material 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. The material recommending method based on the artificial intelligence is characterized by comprising the following steps of:

receiving a recommendation request triggered by a user;

And pushing the target recommended materials to the user.

2. The method for recommending materials based on artificial intelligence according to claim 1, wherein the step of acquiring target user data corresponding to the user based on a preset target dimension specifically comprises:

acquiring user information of the user;

And taking the specified data as the target user data.

3. The method for recommending materials based on artificial intelligence according to claim 1, wherein the step of querying the materials to be recommended corresponding to the target crowd label from a preset second service system specifically comprises:

Invoking the second service system;

and taking the second material as the material to be recommended.

4. The method for recommending materials based on artificial intelligence according to claim 1, wherein the step of analyzing the target user data and the materials to be recommended based on a preset recommendation model, and determining recommended materials corresponding to the target user data from the materials to be recommended specifically comprises:

and taking the appointed material as the recommended material.

5. The method for recommending materials based on artificial intelligence according to claim 1, wherein the step of filtering the recommended materials based on a preset recommendation policy to obtain corresponding target recommended materials specifically comprises:

and taking the fourth material as the target recommended material.

6. The method for recommending materials based on artificial intelligence according to claim 1, wherein the step of pushing the target recommended materials to the user comprises:

7. The method for recommending material based on artificial intelligence according to claim 1, further comprising, before the step of analyzing the target user data and the material to be recommended based on the preset recommendation model and determining the recommended material corresponding to the target user data from the material to be recommended:

acquiring pre-acquired historical user data;

Preprocessing the historical user data to obtain corresponding sample data;

Dividing the sample data into a training set and a testing set;

and taking the specified model as the recommendation model.

8. An artificial intelligence-based material recommendation device, comprising:

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