Movatterモバイル変換


[0]ホーム

URL:


CN120387858A - A marketing method, system, device and medium based on user portrait - Google Patents

A marketing method, system, device and medium based on user portrait

Info

Publication number
CN120387858A
CN120387858ACN202510873444.4ACN202510873444ACN120387858ACN 120387858 ACN120387858 ACN 120387858ACN 202510873444 ACN202510873444 ACN 202510873444ACN 120387858 ACN120387858 ACN 120387858A
Authority
CN
China
Prior art keywords
user
data
attribute
marketing
users
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202510873444.4A
Other languages
Chinese (zh)
Other versions
CN120387858B (en
Inventor
吴伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Ifudata Information Technology Co ltd
Original Assignee
Beijing Ifudata Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Ifudata Information Technology Co ltdfiledCriticalBeijing Ifudata Information Technology Co ltd
Priority to CN202510873444.4ApriorityCriticalpatent/CN120387858B/en
Publication of CN120387858ApublicationCriticalpatent/CN120387858A/en
Application grantedgrantedCritical
Publication of CN120387858BpublicationCriticalpatent/CN120387858B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Landscapes

Abstract

Translated fromChinese

本发明公开了一种基于用户画像的营销方法、系统、设备及介质,其属于营销技术领域,包括获取多个用户的行为数据,并进行预处理操作,得到预处理行为数据,基于用户属性模型进行属性特征提取和概率预测,构建多个用户的用户画像,并进行分组,得到多个用户组;构建知识图谱,基于图算法提取得到高阶邻居节点特征,使用营销方案生成模型,得到多个营销方案,将知识图谱和多个营销方案输入至推荐模型中,随后选取最高推荐度所对应的营销方案推送给对应的用户。本发明通过用户属性模型充分处理用户的多维特征,其中用户画像包括多个属性特征,解决用户画像数据单一的问题,且选取推荐度最高的营销方案推送给客户,大幅提高营销方案推荐精准度。

The present invention discloses a marketing method, system, device and medium based on user portrait, which belongs to the field of marketing technology, including obtaining behavioral data of multiple users, performing preprocessing operations to obtain preprocessed behavioral data, performing attribute feature extraction and probability prediction based on a user attribute model, constructing user portraits of multiple users, and grouping them to obtain multiple user groups; constructing a knowledge graph, extracting high-order neighbor node features based on a graph algorithm, using a marketing plan generation model to obtain multiple marketing plans, inputting the knowledge graph and multiple marketing plans into a recommendation model, and then selecting the marketing plan corresponding to the highest recommendation degree and pushing it to the corresponding user. The present invention fully processes the multidimensional characteristics of users through a user attribute model, wherein the user portrait includes multiple attribute features, solves the problem of single user portrait data, and selects the marketing plan with the highest recommendation degree to push to the customer, greatly improving the accuracy of marketing plan recommendations.

Description

Marketing method, system, equipment and medium based on user portrait
Technical Field
The invention belongs to the technical field of marketing, and particularly relates to a marketing method, a marketing system, marketing equipment and marketing media based on user portraits.
Background
With the rapid development of internet technology and big data, modern information volume is explosive growth, people's life has been greatly enriched, but it becomes more time consuming and difficult to also make people acquire required information from data, along with the growth of user volume and interactive data, traditional data processing instrument is difficult to high-efficient processing huge amount, high-dimensional user data, and various information of user is comparatively fragmented, and it distributes in different platforms or systems, easily appears the problem of data island. With the increase of user quantity and interaction data, the traditional data processing tool is difficult to efficiently process massive and high-dimensional user data, and cannot rapidly perform deep analysis.
At present, common methods for constructing user portraits according to user data comprise a data statistics-based method, a machine learning-based method and a deep learning-based method, but the accuracy of extracted features can influence the result for more complex data distribution, and the machine learning-based method and the deep learning-based method only can mine the features of a single dimension when processing the user data, neglect the features of other dimensions, and cannot fully utilize the information in the user data. The conventional marketing system generates a marketing scheme according to the user personalized data, pushes the marketing scheme, and because the user personalized data is single, the user portrait generated based on the user personalized data may have the problem of single data, so that the marketing scheme recommendation accuracy is lower. Therefore, there is a need to provide a marketing method based on a user portrait scheme to solve the problems of insufficient multi-dimensional feature utilization and low recommendation accuracy.
Disclosure of Invention
The invention aims to provide a marketing method, a marketing system, marketing equipment and marketing media based on user portraits, which are used for solving the problems in the prior art.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
In a first aspect, the present invention provides a user portrait-based marketing method, including:
acquiring behavior data of a plurality of users, wherein the behavior data comprises behavior log data of a cross-domain platform, UGC content and social relationship graphs, and preprocessing the behavior data to obtain preprocessed behavior data;
inputting the preprocessing behavior data into a user attribute model for attribute feature extraction and prediction to obtain attribute features of users and attribute probabilities of the users, constructing user portraits of a plurality of users based on the attribute features of the users and the attribute probabilities of the users, and grouping the user portraits of the plurality of users to obtain a plurality of user groups;
Acquiring item portraits and interaction relations between users and the items, and constructing a knowledge graph based on a plurality of user groups, the item portraits and the interaction relations between the users and the items, wherein the user portraits and the item portraits in the user groups are used as entity nodes of the knowledge graph, and the interaction relations between the users and the items are used as edges of the knowledge graph;
Searching adjacent nodes of each entity node in the knowledge graph, extracting features of the adjacent nodes based on a graph algorithm to obtain high-order neighbor node features, inputting the user portrait and the high-order neighbor node features into a marketing scheme generation model to obtain a plurality of marketing schemes, inputting the knowledge graph and the plurality of marketing schemes into a recommendation model to obtain a recommendation degree sequence, and selecting a marketing scheme corresponding to a first recommendation degree in the recommendation degree sequence to be pushed to a corresponding user.
In one possible design, the preprocessing operation is performed on the behavior data to obtain preprocessed behavior data, including:
performing data cleaning on the behavior data, and performing missing value filling on the behavior data after data cleaning to obtain complete behavior data;
and denoising and normalizing the complete behavior data by a moving average method to obtain the preprocessed behavior data.
In one possible design, the user attribute model includes a pre-training encoding layer, a DPCNN channel layer, a GRU and attention mechanism channel layer and an output layer, and the step of inputting the pre-processing behavior data into the user attribute model to extract and predict the attribute characteristics and obtain the attribute characteristics and the attribute probability of the user includes:
inputting the preprocessing behavior data into a pre-training encoding layer to obtain data depth semantic information, and representing the data depth semantic information in a vectorization form to obtain a first data feature vector;
Inputting the first data feature vector into DPCNN channel layers for feature extraction to obtain a second data feature vector;
inputting the second data feature vector into the GRU and the attention mechanism channel layer to obtain first feature data;
The output layer performs splicing fusion on the first data feature vector, the second data feature vector and the first feature data to obtain attribute features of the user, and predicts the attribute features of the user by using a sigmoid function to obtain attribute probability of the user.
In one possible design, inputting the second data feature vector into the GRU and attention mechanism channel layer results in first feature data comprising:
capturing long-distance dependency relations in the second data feature vector based on the GRU model;
The attention mechanism layer is used for carrying out weight calculation on the captured long-distance dependency relationship to obtain the long-distance dependency relationship with different weights;
And extracting the characteristics of the long-distance dependency relations of different weights to obtain first characteristic data.
In one possible design, the calculation expression of the attribute feature of the user and the attribute probability of the user is:
;
In the above-mentioned description of the invention,The data representing the characteristics of the fusion is presented,A first data feature vector is represented and,Representing a feature vector of the second data,The first characteristic data is represented by a first set of characteristics,Representing the probability of the attribute of the user,The sigmoid function is represented as a function,The activation function is represented as a function of the activation,The weight is represented by a weight that,Representing the bias parameters.
In one possible design, grouping user portraits for a plurality of users to obtain a plurality of user groups includes:
Randomly selecting a user portrait of a user as an initial cluster center;
Calculating the distance from the user portrait of each user to the center of the initial cluster, and distributing the user portrait of each user to the clusters with the distance smaller than the preset distance;
Calculating the average value of the user portraits in each cluster, taking each average value as the cluster center of each cluster, calculating the distance from the user portraits of each user to the cluster center again, and reassigning the user portraits to the clusters with the distances smaller than the preset distance until the preset termination condition is reached, thereby obtaining a plurality of user groups.
In one possible design, the recommendation model comprises an embedding layer, an attention embedding propagation layer and a prediction layer, wherein the step of inputting the knowledge graph and the plurality of marketing schemes into the recommendation model to obtain a recommendation degree sequence comprises the following steps:
The embedding layer embeds the knowledge graph and the plurality of marketing schemes into a low-dimensional vector space to obtain an initial embedded vector;
inputting the initial embedded vector into an attention embedded propagation layer for calculation to obtain a marketing scheme weight coefficient;
And calculating the knowledge graph and the weight coefficients of the marketing schemes by using the graph neural network in the prediction layer to obtain a plurality of recommendation degrees, and sequencing the recommendation degrees to obtain a recommendation degree sequence.
In a second aspect, the present invention provides a user portrayal-based marketing system comprising:
The system comprises an acquisition module, a preprocessing module and a processing module, wherein the acquisition module is used for acquiring behavior data of a plurality of users, the behavior data comprises behavior log data of a cross-domain platform, UGC content and social relationship graphs, and preprocessing operation is carried out on the behavior data to obtain preprocessed behavior data;
The feature module is used for inputting the preprocessing behavior data into the user attribute model to perform attribute feature extraction and prediction to obtain attribute features of users and attribute probabilities of the users, constructing user portraits of a plurality of users based on the attribute features of the users and the attribute probabilities of the users, and grouping the user portraits of the plurality of users to obtain a plurality of user groups;
The building module is used for obtaining the project portrait and the interaction relation between the user and the project, and building a knowledge graph based on a plurality of user groups, the project portrait and the interaction relation between the user and the project, wherein the user portrait and the project portrait in the user groups are used as entity nodes of the knowledge graph, and the interaction relation between the user and the project is used as edges of the knowledge graph;
The recommendation module is used for searching adjacent nodes of each entity node in the knowledge graph, extracting features of the adjacent nodes based on a graph algorithm to obtain high-order neighbor node features, inputting user images and the high-order neighbor node features into the marketing scheme generation model to obtain a plurality of marketing schemes, inputting the knowledge graph and the plurality of marketing schemes into the recommendation model to obtain a recommendation degree sequence, and selecting a marketing scheme corresponding to a first recommendation degree in the recommendation degree sequence to be pushed to a corresponding user.
In a third aspect, the present invention provides a computer device comprising a memory, a processor and a transceiver in communication with each other in sequence, wherein the memory is configured to store a computer program, the transceiver is configured to send and receive messages, and the processor is configured to read the computer program and perform the user portrait based marketing method according to the first aspect.
In a fourth aspect, the present invention provides a computer readable storage medium having instructions stored thereon which, when executed on a computer, perform the user portrait based marketing method of the first aspect.
The beneficial effects of the invention are as follows:
The invention discloses a marketing method, a system, equipment and a medium based on user portraits, which are characterized in that firstly, behavior data of a plurality of users are obtained and preprocessed to obtain preprocessed behavior data, attribute feature extraction and prediction are carried out by using a user attribute model, the user portraits of the plurality of users are constructed based on the attribute features and attribute probabilities of the users, the user portraits are grouped to obtain a plurality of user groups, the interaction relationship between a project portraits and the user and the project is obtained, a knowledge map is constructed based on the plurality of user groups, the project portraits and the interaction relationship between the user and the project, a high-order neighbor node feature is obtained according to the knowledge map, a model is generated by using the user portraits and the high-order neighbor node feature marketing scheme, a plurality of marketing schemes are obtained, recommendation degrees are generated by using the recommendation models and aligned, recommendation degree sequences are obtained, and the marketing schemes corresponding to the first recommendation degrees are selected to be pushed to the corresponding users. The user portrait is constructed by fully processing multidimensional features of users through the user attribute model, comprises a plurality of user attribute features, solves the problem of singleness of user portrait data, constructs a knowledge graph according to the user portrait and the project portrait, intuitively reflects the preference relation between the users and the projects, sorts marketing schemes after generating the marketing schemes, selects the marketing scheme with the highest recommendation degree to push to the clients, greatly improves the recommendation accuracy of the marketing scheme, maximizes the marketing effect, reduces the marketing cost, improves the return on investment, and is convenient for practical application and popularization.
Drawings
FIG. 1 is a flow chart of a user portrait based marketing method provided in a first aspect of the present embodiment;
FIG. 2 is a block diagram of a user portrayal-based marketing system provided in accordance with a second aspect of the present embodiments.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be briefly described below with reference to the accompanying drawings and the description of the embodiments or the prior art, and it is obvious that the following description of the structure of the drawings is only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art. It should be noted that the description of these examples is for aiding in understanding the present invention, but is not intended to limit the present invention.
It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.
Examples:
As shown in fig. 1, a first aspect of the present embodiment provides a marketing method based on a user portrait, including, but not limited to, the following steps:
S1, acquiring behavior data of a plurality of users, wherein the behavior data comprises cross-domain platform behavior log data, UGC content (User-GENERATED CONTENT, user generated content) and social relationship graphs, and preprocessing the behavior data to obtain preprocessed behavior data;
Specifically, in step S1, a preprocessing operation is performed on the behavior data to obtain preprocessed behavior data, which includes:
s11, carrying out data cleaning on the behavior data, and carrying out missing value filling on the behavior data after data cleaning to obtain complete behavior data;
s12, denoising and normalizing the complete behavior data by a moving average method to obtain preprocessed behavior data.
In the embodiment, the data is cleaned for the behavior data, the missing part is filled, the influence of abnormal values and missing values on the subsequent data processing process is avoided, the moving average method is adopted for denoising and normalizing the complete behavior data, so that the influence of noise is reduced, the robustness of the model is improved, and the principle of denoising by the moving average method is that the noise in the data is removed by calculating the local average value of each data point in the data set.
S2, inputting the preprocessing behavior data into a user attribute model for attribute feature extraction and prediction to obtain attribute features of users and attribute probabilities of the users, constructing user portraits of a plurality of users based on the attribute features of the users and the attribute probabilities of the users, and grouping the user portraits of the plurality of users to obtain a plurality of user groups;
Further, the user attribute model includes a pre-training encoding layer, DPCNN (DEEP PYRAMID Convolutional Neural Networksfor Text Categorization, deep pyramid convolutional neural network) channel layer, a GRU (Gated Recurrent Unit, gated loop unit) and attention mechanism channel layer, and an output layer.
Specifically, in step S2, the preprocessing behavior data is input into a user attribute model to perform attribute feature extraction and prediction, so as to obtain attribute features of the user and attribute probability of the user, including:
s21, inputting the preprocessing behavior data into a pre-training encoding layer to obtain data depth semantic information, and representing the data depth semantic information in a vectorization form to obtain a first data feature vector;
S22, inputting the first data feature vector into a DPCNN channel layer for feature extraction to obtain a second data feature vector;
S23, inputting the second data feature vector into the GRU and the attention mechanism channel layer to obtain first feature data;
Specifically, in step S23, the second data feature vector is input into the GRU and the attention mechanism channel layer to obtain first feature data, which includes:
s231, capturing long-distance dependency relations in the second data feature vector based on the GRU model;
Specifically, in step S231, two gating units are mainly used in the GRU model, and the two gating units are mainly an update gate and a reset gate, where the update gate is used for controlling how much history information needs to be reserved in the current state and how much new information needs to be received, and the reset gate is used for determining how much information of the network state is written in at the previous moment, and the specific calculation formulas of the update gate and the reset gate are as follows:
;
In the above-mentioned description of the invention,Representing the current input and the current level of the input,The state of the last time is indicated,Indicating the state of the current moment of time,The activation function is represented as a function of the activation,Representing the update gate currently learnable connection weight matrix,Indicating that the reset gate is currently learning the connection weight matrix,Indicating that the connection weight matrix was learned a moment before the update gate,Indicating that the connection weight matrix was learned a moment prior to resetting the gate,A learnable offset vector representing an update gate,A learnable offset vector representing a reset gate.
S232, calculating weights of the captured long-distance dependency relationships by using an attention mechanism layer to obtain the long-distance dependency relationships with different weights;
In general, the attention mechanism can be divided into a hard attention mechanism, a soft attention mechanism and a self-attention mechanism, and preferably, in this embodiment, the attention mechanism is a self-attention mechanism, which can efficiently calculate the attention weights of all positions in parallel, learn the relative importance between each position and other positions, and effectively capture the relationship between different positions in the text sequence, so that the neural network focuses on important features more and suppresses the influence of irrelevant features.
S233, extracting features of long-distance dependency relations of different weights to obtain first feature data.
S24, the output layer performs splicing fusion on the first data feature vector, the second data feature vector and the first feature data to obtain attribute features of the user, and predicts the attribute features of the user by using a sigmoid function to obtain attribute probability of the user.
Further, the calculation expression of the attribute characteristics of the user and the attribute probability of the user is as follows:
;
In the above-mentioned description of the invention,The data representing the characteristics of the fusion is presented,A first data feature vector is represented and,Representing a feature vector of the second data,The first characteristic data is represented by a first set of characteristics,Representing the probability of the attribute of the user,The sigmoid function is represented as a function,The activation function is represented as a function of the activation,The weight is represented by a weight that,Representing the bias parameters.
Specifically, in step S2, user portraits of a plurality of users are grouped to obtain a plurality of user groups, including:
S25, randomly selecting a user portrait of a user as an initial cluster center;
s26, calculating the distance from the user portrait of each user to the center of the initial cluster, and distributing the user portrait of each user to the cluster with the distance smaller than the preset distance;
S27, calculating the average value of the user portraits in each cluster, taking each average value as the cluster center of each cluster, calculating the distance from the user portraits of each user to the cluster center again, and reassigning the user portraits to the clusters with the distances smaller than the preset distance until the preset termination condition is reached, so as to obtain a plurality of user groups.
S3, acquiring item portraits and interaction relations between users and items, and constructing a knowledge graph based on a plurality of user groups, the item portraits and the interaction relations between the users and the items, wherein the user portraits and the item portraits in the user groups are used as entity nodes of the knowledge graph, and the interaction relations between the users and the items are used as edges of the knowledge graph;
S4, searching adjacent nodes of each entity node in the knowledge graph, extracting features of the adjacent nodes based on a graph algorithm to obtain high-order neighbor node features, inputting the user portrait and the high-order neighbor node features into a marketing scheme generation model to obtain a plurality of marketing schemes, inputting the knowledge graph and the plurality of marketing schemes into a recommendation model to obtain a recommendation degree sequence, and selecting a marketing scheme corresponding to a first recommendation degree in the recommendation degree sequence to be pushed to a corresponding user.
In step S4, the marketing-scheme generating model is the prior art, and the cyclic neural network (Recurrent Neural Network, RNN) and the transducer (transducer) may be selected to generate the marketing-scheme, or the generation of the countermeasure network may be used to generate the marketing-scheme.
Further, the recommendation model includes an embedding layer, an attention embedding propagation layer, and a prediction layer.
Specifically, in step S4, the knowledge graph and the plurality of marketing schemes are input into a recommendation model to obtain a recommendation degree sequence, which includes:
s41, embedding the knowledge graph and a plurality of marketing schemes into a low-dimensional vector space by an embedding layer to obtain an initial embedded vector;
The expression form of the knowledge graph is a triplet (h, r, t), h and t are entity nodes in the knowledge graph, r is a connection relation in the knowledge graph (namely an edge in the knowledge graph), multiple triples can share the same entity, and the entity serves as a head or a tail in different triples and serves as an association medium of different triples.
S42, inputting the initial embedded vector into an attention embedded propagation layer for calculation to obtain a marketing scheme weight coefficient;
S43, calculating a knowledge graph and a plurality of marketing scheme weight coefficients in a prediction layer by using a graph neural network to obtain a plurality of recommendation degrees, and sequencing the plurality of recommendation degrees to obtain a recommendation degree sequence.
The embodiment provides a marketing method based on user portraits, which comprises the steps of obtaining behavior data of a plurality of users, wherein the behavior data comprise behavior log data of cross-domain platforms, UGC content and social relation graphs, preprocessing the behavior data to obtain preprocessed behavior data, inputting the preprocessed behavior data into a user attribute model to conduct attribute feature extraction and prediction to obtain attribute features of the users and attribute probabilities of the users, constructing user portraits of the plurality of users based on the attribute features of the users and the attribute probabilities of the users, grouping the user portraits of the plurality of users to obtain a plurality of user groups, obtaining interaction relations between the item portraits and the users and the items, constructing a knowledge graph based on the interaction relations between the plurality of user groups, the item portraits and the items, taking the interaction relations between the users and the items in the user groups as entity nodes of the knowledge graph, taking the interaction relations between the users and the items as edges of the knowledge graph, searching adjacent nodes of each entity node in the knowledge graph, conducting feature extraction on the adjacent nodes based on the graph algorithm to obtain high-order neighbor node features, inputting the user portraits and high-order neighbor node features into a marketing scheme, obtaining a plurality of recommended sequence, and recommending a plurality of recommended sequence, and sending the recommended sequence to a first recommendation sequence to the user to the marketing scheme. The method comprises the steps of extracting attribute characteristics and attribute probability of users through a user attribute model, constructing user portraits, grouping the user portraits of a plurality of users by adopting a clustering algorithm, constructing a knowledge graph according to user groups, projects and relations among the user groups, effectively fusing various types of data, comprehensively capturing characteristics of the users and the projects, enhancing the richness of the knowledge graph so as to improve the recommendation accuracy, generating the cooperation of the model and the recommendation model through a marketing scheme, maximizing the marketing effect, improving the user experience and enhancing the user viscosity.
As shown in fig. 2, a second aspect of the present embodiment provides a marketing system based on user portraits, including:
The system comprises an acquisition module, a preprocessing module and a processing module, wherein the acquisition module is used for acquiring behavior data of a plurality of users, the behavior data comprises behavior log data of a cross-domain platform, UGC content and social relationship graphs, and preprocessing operation is carried out on the behavior data to obtain preprocessed behavior data;
The feature module is used for inputting the preprocessing behavior data into the user attribute model to perform attribute feature extraction and prediction to obtain attribute features of users and attribute probabilities of the users, constructing user portraits of a plurality of users based on the attribute features of the users and the attribute probabilities of the users, and grouping the user portraits of the plurality of users to obtain a plurality of user groups;
The building module is used for obtaining the project portrait and the interaction relation between the user and the project, and building a knowledge graph based on a plurality of user groups, the project portrait and the interaction relation between the user and the project, wherein the user portrait and the project portrait in the user groups are used as entity nodes of the knowledge graph, and the interaction relation between the user and the project is used as edges of the knowledge graph;
The recommendation module is used for searching adjacent nodes of each entity node in the knowledge graph, extracting features of the adjacent nodes based on a graph algorithm to obtain high-order neighbor node features, inputting user images and the high-order neighbor node features into the marketing scheme generation model to obtain a plurality of marketing schemes, inputting the knowledge graph and the plurality of marketing schemes into the recommendation model to obtain a recommendation degree sequence, and selecting a marketing scheme corresponding to a first recommendation degree in the recommendation degree sequence to be pushed to a corresponding user.
The working process, working details and technical effects of the foregoing system provided in the second aspect of the present embodiment may refer to the marketing method described in the first aspect, which are not described herein.
A third aspect of the present embodiment provides a computer device for performing the marketing method according to the first aspect, comprising a memory, a processor and a transceiver in communication with each other, wherein the memory is configured to store a computer program, the transceiver is configured to send and receive messages, and the processor is configured to read the computer program and perform the marketing method according to the first aspect. By way of specific example, the Memory may include, but is not limited to, random-Access Memory (RAM), read-Only Memory (ROM), flash Memory (Flash Memory), first-in first-out Memory (First Input First Output, FIFO), and/or first-in last-out Memory (First Input Last Output, FILO), etc., and the processor may be, but is not limited to, a microprocessor of the STM32F105 family. In addition, the computer device may include, but is not limited to, a power module, a display screen, and other necessary components.
The working process, working details and technical effects of the foregoing computer device provided in the third aspect of the present embodiment may refer to the marketing method described in the first aspect, which are not described herein again.
A fourth aspect of the present embodiment provides a computer-readable storage medium storing instructions comprising the marketing method of the first aspect, i.e. the computer-readable storage medium has instructions stored thereon which, when executed on a computer, perform the marketing method of the first aspect. The computer readable storage medium refers to a carrier for storing data, and may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash Memory, and/or a Memory Stick (Memory Stick), where the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.
The working process, working details and technical effects of the foregoing computer readable storage medium provided in the fourth aspect of the present embodiment may refer to the marketing method as described in the first aspect, and are not repeated herein.
Finally, it should be noted that the above description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

CN202510873444.4A2025-06-272025-06-27Marketing method, system, equipment and medium based on user portraitActiveCN120387858B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202510873444.4ACN120387858B (en)2025-06-272025-06-27Marketing method, system, equipment and medium based on user portrait

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202510873444.4ACN120387858B (en)2025-06-272025-06-27Marketing method, system, equipment and medium based on user portrait

Publications (2)

Publication NumberPublication Date
CN120387858Atrue CN120387858A (en)2025-07-29
CN120387858B CN120387858B (en)2025-10-03

Family

ID=96485795

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202510873444.4AActiveCN120387858B (en)2025-06-272025-06-27Marketing method, system, equipment and medium based on user portrait

Country Status (1)

CountryLink
CN (1)CN120387858B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN112765486A (en)*2021-01-222021-05-07重庆邮电大学Knowledge graph fused attention mechanism movie recommendation method
CN114913028A (en)*2022-06-082022-08-16华东师范大学 A group recommendation method based on knowledge graph extraction tendency
WO2023097929A1 (en)*2021-12-012023-06-08浙江师范大学Knowledge graph recommendation method and system based on improved kgat model
CN118052624A (en)*2024-04-162024-05-17浙江同花顺智能科技有限公司Marketing strategy generation method and device, electronic equipment and storage medium
CN118193683A (en)*2024-05-142024-06-14福州掌中云科技有限公司Text recommendation method and system based on language big model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN112765486A (en)*2021-01-222021-05-07重庆邮电大学Knowledge graph fused attention mechanism movie recommendation method
WO2023097929A1 (en)*2021-12-012023-06-08浙江师范大学Knowledge graph recommendation method and system based on improved kgat model
CN114913028A (en)*2022-06-082022-08-16华东师范大学 A group recommendation method based on knowledge graph extraction tendency
CN118052624A (en)*2024-04-162024-05-17浙江同花顺智能科技有限公司Marketing strategy generation method and device, electronic equipment and storage medium
CN118193683A (en)*2024-05-142024-06-14福州掌中云科技有限公司Text recommendation method and system based on language big model

Also Published As

Publication numberPublication date
CN120387858B (en)2025-10-03

Similar Documents

PublicationPublication DateTitle
CN109960726B (en)Text classification model construction method, device, terminal and storage medium
US11782928B2 (en)Computerized information extraction from tables
CN112949647B (en)Three-dimensional scene description method and device, electronic equipment and storage medium
CN110032632A (en)Intelligent customer service answering method, device and storage medium based on text similarity
CN112380319B (en)Model training method and related device
CN113761220B (en)Information acquisition method, device, equipment and storage medium
WO2019015631A1 (en)Method for generating combined features for machine learning samples and system
CN103995804A (en)Cross-media topic detection method and device based on multimodal information fusion and graph clustering
CN116993185B (en)Time sequence prediction method, device, equipment and storage medium
CN113569041A (en) Text detection method, apparatus, computer equipment, and readable storage medium
CN111178986A (en)User-commodity preference prediction method and system
CN115248855A (en)Text processing method and device, electronic equipment and computer readable storage medium
CN111310462A (en)User attribute determination method, device, equipment and storage medium
CN119249224A (en) Power grid protocol identification method, device, computer equipment, readable storage medium and program product based on knowledge graph
US11478715B1 (en)User-controllable model-driven matchmaking
CN120387858B (en)Marketing method, system, equipment and medium based on user portrait
CN116484953B (en) A travel purpose inference method and terminal
US20250104399A1 (en)Data attribution for diffusion models
CN112559877A (en)CTR (China railway) estimation method and system based on cross-platform heterogeneous data and behavior context
CN116452281A (en)Sequence recommendation method, device and medium based on hypergraph and self-supervision learning
CN114936327A (en)Element recognition model obtaining method and device, computer equipment and storage medium
CN113705071A (en)Equipment identification method, device, equipment and storage medium
CN114155016A (en) A kind of click-through rate estimation method, device, device and readable storage medium
CN119248955B (en)Unsupervised depth image hash retrieval method based on external knowledge
CN114298118A (en)Data processing method based on deep learning, related equipment and storage medium

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp