CN116796076A - Service recommendation method, device, equipment and storage medium - Google Patents
Service recommendation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN116796076A CN116796076A CN202311094854.6A CN202311094854A CN116796076A CN 116796076 A CN116796076 A CN 116796076A CN 202311094854 A CN202311094854 A CN 202311094854A CN 116796076 A CN116796076 A CN 116796076A
- Authority
- CN
- China
- Prior art keywords
- user
- group
- recommendation
- information
- population
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000006399 behavior Effects 0.000 claims description 54
- 230000000694 effects Effects 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000007726 management method Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a service recommendation method, a device, equipment and a storage medium, comprising the following steps: acquiring a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended; determining an initial user population; acquiring target user information and historical behavior data of each user in an initial user group, and determining a recommendation scheme; based on a recommendation scheme, sending recommendation information of a service to be recommended to an initial user group, extracting user group characteristics with positive feedback and user group characteristics with negative feedback, further determining group difference characteristics and group common characteristics, and then adjusting user characteristics and characteristic weights in the initial user portrait to obtain a target user portrait; the target user group corresponding to the target user image is determined, and then the service to be recommended is recommended, and the user group requiring the service can be accurately found by the mode, so that accurate recommendation is realized. In addition, a service recommending device, equipment and a storage medium are also provided.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a storage medium for recommending services.
Background
With the development of artificial intelligence technology, many fields begin to use big data to make intelligent recommendation, and make personalized recommendation, and most typically, the e-commerce field and the news field. However, unlike e-commerce services, telecommunication services often have no much historical data to reference, and the data that can be referenced is relatively single, and many times the user may be a new user, with no more data to reference than the user information. Especially for new services, the problem of cold start is often faced, so that accurate recommendation is often not realized in the recommendation of telecommunication services, thereby influencing the popularization effect of the services.
Disclosure of Invention
Based on the method, the device, the equipment and the storage medium for realizing accurate recommendation are provided.
To achieve the above object, a first aspect of the present application provides a service recommendation method, including:
acquiring a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
acquiring an initial user group meeting the initial user portrait, including: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
Acquiring target user information and historical behavior data of each user in an initial user group, and determining a recommendation scheme based on the target user information and the historical behavior data, wherein the historical behavior data comprises: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
transmitting the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information;
determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
according to the group difference characteristics and the group common characteristics, user characteristics and characteristic weights in the initial user portrait are adjusted to obtain a target user portrait;
and determining a target user group corresponding to the target user image, and recommending the service to be recommended to the target user group.
In order to achieve the above object, a second aspect of the present application provides a service recommendation device, including:
The first determining module is used for acquiring a service to be recommended and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
the acquisition module is used for acquiring an initial user group meeting the initial user portrait and comprises the following steps: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
the second determining module is configured to obtain target user information of each user in the initial user group and historical behavior data of the user, and determine a recommendation scheme based on the target user information and the historical behavior data, where the historical behavior data includes: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
The extraction module is used for sending the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback to the recommendation information;
the third determining module is used for determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
the adjustment module is used for adjusting the user characteristics and the characteristic weights in the initial user portrait according to the group difference characteristics and the group common characteristics to obtain a target user portrait;
and the recommending module is used for determining a target user group corresponding to the target user image and recommending the service to be recommended to the target user group.
To achieve the above object, a third aspect of the present application provides a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
Acquiring a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
acquiring an initial user group meeting the initial user portrait, including: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
acquiring target user information and historical behavior data of each user in an initial user group, and determining a recommendation scheme based on the target user information and the historical behavior data, wherein the historical behavior data comprises: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
transmitting the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information;
Determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
according to the group difference characteristics and the group common characteristics, user characteristics and characteristic weights in the initial user portrait are adjusted to obtain a target user portrait;
and determining a target user group corresponding to the target user image, and recommending the service to be recommended to the target user group.
To achieve the above object, a fourth aspect of the present application provides a computer-readable storage medium comprising: a computer program is stored which, when executed by a processor, causes the processor to perform the steps of:
acquiring a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
acquiring an initial user group meeting the initial user portrait, including: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
Acquiring target user information and historical behavior data of each user in an initial user group, and determining a recommendation scheme based on the target user information and the historical behavior data, wherein the historical behavior data comprises: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
transmitting the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information;
determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
according to the group difference characteristics and the group common characteristics, user characteristics and characteristic weights in the initial user portrait are adjusted to obtain a target user portrait;
and determining a target user group corresponding to the target user image, and recommending the service to be recommended to the target user group.
According to the service recommendation method, the device, the equipment and the storage medium, firstly, the initial user portrait corresponding to the service to be recommended is determined, the initial user population is determined based on the initial user portrait, in order to accurately determine the initial user population, target user information is obtained through analyzing and perfecting related information of users, the initial user population is determined based on the target user information after perfecting, a recommendation scheme is determined according to the target user information and historical behavior data of the initial user population, population characteristics of the user population with positive feedback and population characteristics of the user population with negative feedback are extracted to determine population difference characteristics and population common characteristics, the user characteristics and characteristic weights in the initial user portrait are adjusted based on the population difference characteristics and the population common characteristics, the target user portrait is obtained, the target user population corresponding to the target user portrait is determined, and service recommendation is further carried out on the target user population. For the problem of insufficient data cold start, the initial user portrait is recommended firstly, then the initial user portrait is adjusted based on recommendation feedback so as to obtain the target user portrait, and the user group requiring the service can be accurately found by the mode, so that the accurate recommendation is realized, and the problem of inaccurate recommendation caused by insufficient data in cold start is solved.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Wherein:
FIG. 1 is a flow chart of a business recommendation method in one embodiment;
FIG. 2 is a flow diagram of a method of determining a first recommendation in one embodiment;
FIG. 3 is a flow diagram of a method for determining a recommendation corresponding to a second user group, in one embodiment;
FIG. 4 is a block diagram of a service recommendation device in one embodiment;
fig. 5 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It is noted that the terms "comprising," "including," and "having," and any variations thereof, in the description and claims of the application and in the foregoing figures, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus. In the claims, specification, and drawings of the present application, relational terms such as "first" and "second", and the like are used solely to distinguish one entity/operation/object from another entity/operation/object without necessarily requiring or implying any actual such relationship or order between such entities/operations/objects.
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.
As shown in fig. 1, a service recommendation method is provided, and the method includes:
step 102, obtaining a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features.
The service to be recommended is a service to be promoted, for example, a cloud storage service is a video monitoring service which can be provided for a user by an operator, for example, based on a video uploaded to a cloud, the user can be assisted to monitor whether an old person or a child in a home falls or not, and the like. The service to be recommended is a new service, a corresponding audience is assumed at the beginning of recommendation, and since the service to be recommended is a new service and has no corresponding reference data, an initial audience is assumed in advance, that is, an initial user portrait is set, for example, the initial user portrait is set as follows: the family is the 22-52 age group of the elderly or children.
Step 104, obtaining an initial user group meeting the initial user portrait, including: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when the associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the completed target user information; an initial user population satisfying the initial user representation is determined based on the completed target user information.
In which, the user needs to register real name when handling telecommunication service, for example, handling SIM card and the like needs to fill in real name registration information, so the user information can be obtained from the system. The user information includes: the user gender, the user age, the region where the user is located and whether the associated card exists under the name or not, and the associated card refers to other cards which are associated with the main card, such as a subsidiary card. The method comprises the steps of obtaining positioning information corresponding to the auxiliary card, determining an activity track of the auxiliary card user, deducing the identity of the auxiliary card user according to the activity track, wherein the identity is deduced to be a student if the activity track is frequently in a school, and deducing the identity to be an old person if the activity track is frequently in an old person activity center. After the identity information of the auxiliary card is analyzed, the corresponding user tag is determined to be added into the user information, so that the completed target user information is obtained.
Step 106, obtaining target user information and historical behavior data of the users in the initial user group, and determining a recommendation scheme based on the target user information and the historical behavior data, wherein the historical behavior data comprises: data of which services have been purchased historically, and events and time points that trigger the purchase of the corresponding services.
Wherein, in the initial user group, a part of users have historical behavior data, and the recommendation of the part of users has referenceable historical behavior data, so that the part can be used for determining a recommendation scheme based on target user information and the historical behavior data in the same way as other recommendation modes based on big data. The historical behavior data in the power grid domain mainly include: what packages, traffic usage, package usage, what value added services, and trigger events and time points corresponding to each purchase of the corresponding services are historically purchased. For example, when the user's traffic is fast, the system automatically recommends a traffic package to the user, and then if the user purchases the recommended traffic package, the user records the corresponding trigger event and time point accordingly.
And step 108, transmitting the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information.
The recommendation information can be summary information or link information, and it can be determined which users are positively fed back users and which users have negatively fed back users by collecting whether the users click on the recommendation information. The positive feedback and the negative feedback are opposite, for example, if the user clicks the recommendation, the positive feedback is marked, and if the user clicks the recommendation, the negative feedback is marked, and if the user purchases the recommendation, the positive feedback is marked, and if the user purchases the recommendation, the negative feedback is marked. The rules of positive feedback and negative feedback can be customized in advance for different services.
Step 110, determining group difference features and group common features according to the group features of the user group with positive feedback and the group features of the user group with negative feedback, wherein the group features are extracted based on target user information.
After the user group with positive feedback is obtained, extracting group characteristics of the user group, wherein the group characteristics are extracted based on user information, namely, analyzing the user characteristics of the user group with positive feedback and the user characteristics of the user group with negative feedback, and then finding out group difference characteristics, wherein the group difference characteristics are important characteristics for determining whether a user is interested in recommended content or not.
And step 112, adjusting the user characteristics and the characteristic weights in the initial user portrait according to the group difference characteristics and the group common characteristics to obtain the target user portrait.
The group difference features are difference features corresponding to positive feedback and negative feedback, and part of the difference features are important features for determining whether a user is interested in recommended service or not, so that the part of the difference features need to be enhanced, and then the common features are features of both positive feedback and negative feedback, which indicate that the feature is not a main factor affecting the user, and the weights of the part of the difference features can be adjusted downwards. For example, what we initial user portrayal might set is
After the recommended feedback is obtained, the user in the age group 22-52 with old people or children may find out whether the old people or children in the family are important features for determining whether to purchase or not, and the age group or area is important features for determining whether to purchase or not, so that the initial user portrait can be adjusted to the target user portrait in the age group 22-52 and XX area.
Step 114, determining a target user group corresponding to the target user image, and recommending the service to be recommended to the target user group.
The target user portrait can be accurately recommended based on the finally determined target user group, so that accurate marketing is realized.
According to the business recommendation method, firstly, an initial user portrait corresponding to a business to be recommended is determined, an initial user group is determined based on the initial user portrait, in order to accurately determine the initial user group, target user information is obtained through analysis and perfection of user related information, the initial user group is determined based on the completed target user information, a recommendation scheme is determined according to the target user information and historical behavior data of the initial user group, group difference characteristics and group common characteristics are determined by extracting group characteristics of the user group with positive feedback and group characteristics of the user group with negative feedback, the user characteristics and characteristic weights in the initial user portrait are adjusted based on the group difference characteristics and the group common characteristics, the target user group corresponding to the target user portrait is determined, and business recommendation is further carried out on the target user group. For the problem of insufficient data cold start, the initial user portrait is recommended firstly, then the initial user portrait is adjusted based on recommendation feedback so as to obtain the target user portrait, and the user group requiring the service can be accurately found by the mode, so that the accurate recommendation is realized, and the problem of inaccurate recommendation caused by insufficient data in cold start is solved.
In one embodiment, the business recommendation method is applied to a user operating marketing system which can provide more powerful micro-service architecture support by introducing Spring Cloud. Spring Cloud is an open source framework for building a distributed system, is based on Spring Boot, and provides numerous Cloud native components and tools, such as service registration and discovery, load balancing, circuit breakers, configuration management, etc., to help build highly available, scalable distributed applications.
In addition, the system adopts MySQL system, which is a common relational database management system and can be used for storing user data and task information. MySQL has good performance and reliability, supporting high concurrency access and complex query operations. Through reasonable database design and optimization, efficient management and access of the system to the user data can be ensured.
In building a user operated marketing system, the introduction of the XXL-Job framework may provide support for task scheduling and distributed task execution. XXL-Job is an open-source distributed task scheduling platform and has the characteristics of simplicity, easiness in use, high reliability and high performance. The method is developed based on Java language, and is suitable for various task scheduling scenes including timing tasks, timing scheduling, data processing and the like.
The following effects can also be achieved through the user operation marketing: firstly, the user touch is diversified, namely, the user touch of various modes such as short messages, mails, push notifications, telephone outside calling robots and the like is realized through integrating APIs of third-party service providers, and the marketing information is ensured to be accurately and timely transmitted to the users. Secondly, the task management is efficient, namely a task management module is established, the creation, the scheduling, the execution and the monitoring of the task are realized, and the task processing efficiency and the response speed are improved through technical means such as asynchronous execution, queues and schedulers. And thirdly, the data management is refined, namely, a user image module is designed, basic information, behavior data and the like of a user are collected and analyzed, so that accurate user insight and personalized marketing strategies are provided for marketing teams, and the marketing effect is improved.
In one embodiment, the initial user population comprises: the system comprises a first user group and a second user group, wherein the first user group refers to a user group corresponding to historical behavior data in an initial user group, and the second user group refers to a user group without the historical behavior data in the initial user group;
the step of sending the recommendation information of the service to be recommended to the initial user group, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information, comprises the following steps:
Sending recommendation information of a service to be recommended to a first user group in the initial user group, and extracting user group characteristics with positive feedback and user group characteristics with negative feedback for the recommendation information in the first user group;
acquiring a second user group without historical behavior data in the initial user group, and determining a recommendation scheme corresponding to the second user group based on the group difference characteristics;
and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback in the second user group.
In order to make the recommendation more accurate, when recommending the initial user group, the initial user group is divided into a first user group and a second user group, wherein the first user group has referenceable historical behavior data, and the second user group has no historical behavior data. Since the second user population is free of historical behavior data, the recommendation cannot be determined based on the historical behavior data. In order to make the recommendation of the second user group more accurate. Firstly, feedback of a first user group is obtained, and after group difference characteristics are obtained, a corresponding recommendation scheme is matched for a second user group based on the group difference characteristics. By the method, accurate recommendation can be better performed for the second user group, and the problem of recommendation for new users is solved.
In one embodiment, the obtaining the target user information and the historical behavior data of the user of each user in the initial user group, and determining the recommendation scheme based on the target user information and the historical behavior data includes:
performing cluster analysis based on the target user information corresponding to the first user group and historical behavior data of the user to obtain a plurality of first sub-user groups;
extracting group characteristics of each first sub-user group, and determining a first recommendation scheme corresponding to the corresponding first sub-user group based on the group characteristics, wherein the first recommendation scheme comprises the following steps: recommending a document, recommending a mode, recommending time and dynamically filtering a user mode;
receiving recommendation feedback of the first sub-user group;
the determining the group difference characteristic and the group common characteristic according to the group characteristic of the user group with positive feedback and the group characteristic of the user group with negative feedback comprises the following steps:
obtaining sub-population difference characteristics according to the population characteristics of the user population with positive feedback and the population characteristics of the user population with negative feedback in the first sub-user population;
determining sub-population difference characteristics corresponding to each first sub-user population;
The obtaining a second user group without historical behavior data in the initial user group, and determining a recommendation scheme corresponding to the second user group based on the group difference features comprises the following steps:
classifying the second user group based on a plurality of the sub-group difference features, and determining corresponding recommendation schemes based on classification results.
The method comprises the steps of dividing a first user group into a plurality of first sub-user groups, determining a first recommendation scheme corresponding to each first sub-user group, receiving recommendation feedback after recommendation, and extracting difference features according to positive feedback and negative feedback after recommendation to obtain corresponding sub-population difference features, so that the sub-population difference features corresponding to each first sub-user group are obtained. And carrying out matching classification on the second user group and each sub-group difference characteristic, and then determining a corresponding recommendation scheme according to a matching classification result.
Specifically, after the group difference feature is obtained, the second user group and the group difference feature can be matched, the matching degree is calculated, and if the matching degree is larger than a preset value, a recommendation scheme corresponding to the group difference feature is adopted. Each first sub-user group corresponds to a group difference feature. Specifically, the second user population is classified based on a plurality of sub-population difference features, and a corresponding recommendation scheme is determined based on the classification result.
As shown in fig. 2, in one embodiment, the extracting the population feature of each first sub-user group, and determining, based on the population feature, a first recommendation corresponding to the corresponding first sub-user group, where the first recommendation includes: recommending documents and recommending modes, including:
step 112A, obtaining service characteristics of the service to be recommended.
The service features refer to features corresponding to the service to be recommended, and include: the applicable scenario of the service, the service type, and the service purpose.
And step 112B, taking the group characteristics and the business characteristics as the input of a marketing scheme model, and obtaining the marketing characteristics output by the marketing scheme model, wherein the marketing scheme model is obtained based on the marketing data training corresponding to the history recommended business.
The marketing scheme model is obtained by training based on marketing data corresponding to a history recommendation service in advance, and the marketing data corresponding to the history recommendation service comprises: business characteristics, corresponding user groups and obtained marketing characteristics with forward feedback. The marketing features include: characteristics of the document preferred by the user and a document recommendation mode preferred by the user.
And step 112C, obtaining a corresponding first recommendation scheme according to the marketing feature matching.
Different marketing features correspond to different recommended schemes, and the recommended schemes can be determined according to the preset corresponding relation between the marketing features and the recommended features after the marketing features are obtained.
The corresponding marketing characteristic can be obtained by utilizing the marketing scheme model obtained through training according to the group characteristic and the business characteristic, and the corresponding recommendation scheme is obtained based on marketing characteristic matching, so that the determination of the recommendation scheme is time-saving, labor-saving, accurate and reliable.
As shown in fig. 3, in one embodiment, the classifying the second user group based on the plurality of the sub-population difference features, and determining the corresponding recommendation scheme based on the classification result includes:
step 302, obtaining sub-population difference characteristics corresponding to each first sub-user population;
step 304, extracting a second user characteristic of each user in the second user group;
and 306, calculating the coincidence degree between the second user characteristic and each sub-population difference characteristic, taking a first sub-user population corresponding to the sub-population difference characteristic with the highest coincidence degree as a population matched with the user, and taking a corresponding first recommendation scheme as a first recommendation scheme of the user.
The sub-population difference features are features which can reflect whether a user is willing to purchase or not, the sub-population difference features with highest coincidence degree are determined by comparing the coincidence degree of the second user features and each sub-population difference feature, then the first user population corresponding to the sub-population difference features is used as the most matched population of the user, and the first recommended scheme corresponding to the most matched first user population is correspondingly used as the first recommended scheme of the user. In this way, the most suitable recommendation scheme can be matched to the second user group without historical behavior data, so that the recommendation effect is improved.
In one embodiment, the extracting the population characteristics of the user population with positive feedback and the population characteristics of the user population with negative feedback in the first sub-user population to obtain sub-population difference characteristics includes: further subdividing the user features fed back positively, extracting first group features with commonality, further subdividing the user features fed back negatively, and extracting second group features with commonality; sub-population difference features are extracted based on the first population features and the second population features.
In order to extract the difference features, the user features need to be subdivided step by step, for example, the age group can be simply set into three stages, for example, 0-18, 18-60, and over 60 years. If roughly divided, it is likely that the user's age range for both positive and negative feedback is in the 18-60 age range, so that no differential features are extracted, so further subdivision is required until differential features can be extracted, e.g., the age range 18-60 is further subdivided into 18-25 age ranges, 25-35 age ranges, 35-45 age ranges, 45-55 age ranges, etc. Likewise, if a region is involved, it may be initially divided into south China, middle China and north China by region, and then subdivided into provinces, even cities, etc.
In one embodiment, the obtaining the positioning information corresponding to the auxiliary card determines the identity information corresponding to the auxiliary card based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the auxiliary card, including: extracting positioning information of the auxiliary card for a period of time, judging the identity information corresponding to the auxiliary card as a student when schools frequently appear in corresponding position places in the positioning information, and taking the family students as user labels of corresponding users; and when the senile activity places frequently appear in the corresponding position places in the positioning information, judging that the identity information corresponding to the auxiliary card is the old, and taking the home old as the user label of the corresponding user.
And the method comprises the steps of extracting positioning information of the auxiliary card within a period of time (within half a year), making an activity track diagram of a user of the auxiliary card, and indirectly judging identity information corresponding to the auxiliary card based on the activity track diagram.
In one embodiment, after said receiving the recommendation feedback of the first sub-user group, further comprising:
judging whether the recommended feedback is larger than a preset effect value, and when the recommended feedback is larger than the preset effect value, entering the step of extracting the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback in the first sub-user group to obtain the subgroup difference characteristics;
and when the recommended feedback is smaller than a preset effect value, adjusting the first recommended scheme to obtain an adjusted first recommended scheme.
Before receiving the recommendation feedback, it is required to determine whether the effect of the service recommendation reaches an ideal effect (a preset effect value), for example, the conversion rate is set to 10% as the preset effect value, and if the conversion rate is larger than the preset effect value, the recommendation scheme is reasonable, and if the conversion rate is not equal to the preset effect value, the first recommendation scheme needs to be adjusted again, and then the adjusted first recommendation scheme is used for recommendation.
As shown in fig. 4, a service recommendation device is provided, which includes:
a first determining module 402, configured to obtain a service to be recommended, and determine an initial user portrait corresponding to the service to be recommended, where the initial user portrait includes: a plurality of user features;
an obtaining module 404, configured to obtain an initial user group that satisfies the initial user portrait, includes: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
a second determining module 406, configured to obtain target user information of each user in the initial user group and historical behavior data of the user, and determine a recommendation scheme based on the target user information and the historical behavior data, where the historical behavior data includes: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
An extracting module 408, configured to send, to the initial user group, recommendation information of the service to be recommended based on the recommendation scheme, and extract a user group feature with positive feedback and a user group feature with negative feedback for the recommendation information;
a third determining module 410, configured to determine a group difference feature and a group common feature according to the group feature of the user group with positive feedback and the group feature of the user group with negative feedback, where the group feature is extracted based on the target user information;
an adjustment module 412, configured to adjust user features and feature weights in the initial user representation according to the group difference feature and the group common feature, so as to obtain a target user representation;
and a recommending module 414, configured to determine a target user group corresponding to the target user image, and recommend the service to be recommended to the target user group.
In one embodiment, the initial user population comprises: the system comprises a first user group and a second user group, wherein the first user group refers to a user group corresponding to historical behavior data in an initial user group, and the second user group refers to a user group without the historical behavior data in the initial user group;
The extraction module is also used for sending recommendation information of the service to be recommended to a first user group in the initial user groups, and extracting user group characteristics with positive feedback and user group characteristics with negative feedback to the recommendation information in the first user groups; acquiring a second user group without historical behavior data in the initial user group, and determining a recommendation scheme corresponding to the second user group based on the group difference characteristics; and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback in the second user group.
In one embodiment, the second determining module is further configured to perform cluster analysis based on the target user information corresponding to the first user group and historical behavior data of the user, to obtain a plurality of first sub-user groups; extracting group characteristics of each first sub-user group, and determining a first recommendation scheme corresponding to the corresponding first sub-user group based on the group characteristics, wherein the first recommendation scheme comprises the following steps: recommending a document and a recommending mode; receiving recommendation feedback of the first sub-user group;
the third determining module is further used for obtaining sub-population difference characteristics according to the population characteristics of the user population with positive feedback and the population characteristics of the user population with negative feedback in the first sub-user population; determining sub-population difference characteristics corresponding to each first sub-user population;
The second determining module is further configured to classify the second user population based on a plurality of the sub-population difference features, and determine a corresponding recommendation scheme based on a classification result.
In one embodiment, the second determining module is further configured to obtain a service feature of the service to be recommended; taking the group characteristics and the business characteristics as the input of a marketing scheme model, and acquiring the marketing characteristics output by the marketing scheme model, wherein the marketing scheme model is obtained based on marketing data training corresponding to historical recommended business; and obtaining a corresponding first recommendation scheme according to the marketing feature matching.
In one embodiment, the second determining module is further configured to obtain a subgroup difference feature corresponding to each of the first sub-user groups; extracting a second user characteristic of each user in the second user group; and calculating the coincidence degree between the second user characteristic and each sub-population difference characteristic, taking a first sub-user population corresponding to the sub-population difference characteristic with the highest coincidence degree as a population matched with the user, and taking a corresponding first recommendation scheme as a first recommendation scheme of the user.
In one embodiment, the third determining module is further configured to further segment the user features fed back in the positive direction, extract a first group feature having commonality, further segment the user features fed back in the negative direction, and extract a second group feature having commonality; sub-population difference features are extracted based on the first population features and the second population features.
The service recommendation method further comprises the following steps:
the judging module is used for judging whether the recommended feedback is larger than a preset effect value, and entering the step of extracting the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback in the first sub-user group to obtain the subgroup difference characteristics when the recommended feedback is larger than the preset effect value; and when the recommended feedback is smaller than a preset effect value, adjusting the first recommended scheme to obtain an adjusted first recommended scheme.
FIG. 5 illustrates an internal block diagram of a computer device in one embodiment. As shown in fig. 5, the computer device includes a processor, a memory, and a mesh interface connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the computer device may store an operating system, and may also store a computer program, which when executed by the processor, causes the processor to implement the service recommendation method described above. The internal memory may also store a computer program that, when executed by the processor, causes the processor to perform the service recommendation method described above. It will be appreciated by persons skilled in the art that the structure shown in fig. 5 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and does not constitute a limitation of the apparatus to which the present inventive arrangements are applied, and that a particular apparatus may include more or less components than those shown in the drawings, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided that includes a memory and a processor, the memory having stored a computer program that, when executed by the processor, causes the processor to perform the steps of the business recommendation method described above.
In one embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, causes the processor to perform the steps of the service recommendation method described above.
It will be appreciated that the above-described service recommendation method, apparatus, computer device, and computer-readable storage medium belong to a general inventive concept, and embodiments may be mutually applicable.
Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a non-volatile computer readable storage medium, and where the program, when executed, may include processes in the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A business recommendation method, the method comprising:
acquiring a service to be recommended, and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
acquiring an initial user group meeting the initial user portrait, including: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
Acquiring target user information and historical behavior data of each user in an initial user group, and determining a recommendation scheme based on the target user information and the historical behavior data, wherein the historical behavior data comprises: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
transmitting the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information;
determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
according to the group difference characteristics and the group common characteristics, user characteristics and characteristic weights in the initial user portrait are adjusted to obtain a target user portrait;
and determining a target user group corresponding to the target user image, and recommending the service to be recommended to the target user group.
2. The method of claim 1, wherein the initial user population comprises: the system comprises a first user group and a second user group, wherein the first user group refers to a user group corresponding to historical behavior data in an initial user group, and the second user group refers to a user group without the historical behavior data in the initial user group;
The step of sending the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback for the recommendation information, comprises the following steps:
sending recommendation information of a service to be recommended to a first user group in the initial user group, and extracting user group characteristics with positive feedback and user group characteristics with negative feedback for the recommendation information in the first user group;
acquiring a second user group without historical behavior data in the initial user group, and determining a recommendation scheme corresponding to the second user group based on the group difference characteristics;
and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback in the second user group.
3. The method of claim 2, wherein the obtaining target user information and historical behavior data for each user in the initial user population, determining a recommendation based on the target user information and the historical behavior data, comprises:
performing cluster analysis based on the target user information corresponding to the first user group and historical behavior data of the user to obtain a plurality of first sub-user groups;
Extracting group characteristics of each first sub-user group, and determining a first recommendation scheme corresponding to the corresponding first sub-user group based on the group characteristics, wherein the first recommendation scheme comprises the following steps: recommending a document and a recommending mode;
receiving recommendation feedback of the first sub-user group;
the determining the group difference characteristic and the group common characteristic according to the group characteristic of the user group with positive feedback and the group characteristic of the user group with negative feedback comprises the following steps:
obtaining sub-population difference characteristics according to the population characteristics of the user population with positive feedback and the population characteristics of the user population with negative feedback in the first sub-user population;
determining sub-population difference characteristics corresponding to each first sub-user population;
the obtaining a second user group without historical behavior data in the initial user group, and determining a recommendation scheme corresponding to the second user group based on the group difference features comprises the following steps:
classifying the second user group based on a plurality of the sub-group difference features, and determining corresponding recommendation schemes based on classification results.
4. The method of claim 3, wherein the extracting the population characteristics of each of the first sub-user groups determines a first recommendation corresponding to the respective first sub-user group based on the population characteristics, the first recommendation comprising: recommending documents and recommending modes, including:
Acquiring service characteristics of the service to be recommended;
taking the group characteristics and the business characteristics as the input of a marketing scheme model, and acquiring the marketing characteristics output by the marketing scheme model, wherein the marketing scheme model is obtained based on marketing data training corresponding to historical recommended business;
and obtaining a corresponding first recommendation scheme according to the marketing feature matching.
5. A method according to claim 3, wherein said classifying said second population of users based on a plurality of said sub-population difference features, determining a corresponding recommendation based on the classification results, comprises:
obtaining sub-population difference characteristics corresponding to each first sub-user population;
extracting a second user characteristic of each user in the second user group;
and calculating the coincidence degree between the second user characteristic and each sub-population difference characteristic, taking a first sub-user population corresponding to the sub-population difference characteristic with the highest coincidence degree as a population matched with the user, and taking a corresponding first recommendation scheme as a first recommendation scheme of the user.
6. A method according to claim 3, wherein the deriving sub-population difference features from the population features of the user population having positive feedback and the population features of the user population having negative feedback in the first sub-user population comprises:
Further subdividing the user features fed back positively, extracting first group features with commonality, further subdividing the user features fed back negatively, and extracting second group features with commonality;
sub-population difference features are extracted based on the first population features and the second population features.
7. The method of claim 3, further comprising, after said receiving the recommendation feedback for the first sub-user group:
judging whether the recommended feedback is larger than a preset effect value, and when the recommended feedback is larger than the preset effect value, entering the step of extracting the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback in the first sub-user group to obtain the subgroup difference characteristics;
and when the recommended feedback is smaller than a preset effect value, adjusting the first recommended scheme to obtain an adjusted first recommended scheme.
8. A service recommendation device, the device comprising:
the first determining module is used for acquiring a service to be recommended and determining an initial user portrait corresponding to the service to be recommended, wherein the initial user portrait comprises: a plurality of user features;
The acquisition module is used for acquiring an initial user group meeting the initial user portrait and comprises the following steps: obtaining user information, wherein the user information comprises: user gender, user age, region where the user is located, and associated card information; when an associated card exists, positioning information corresponding to the associated card is obtained, and identity information corresponding to the associated card is determined based on the positioning information; determining a corresponding user tag according to the identity information corresponding to the associated card, and adding the user tag into the user information to obtain the finished target user information; determining an initial user group meeting the initial user portrait based on the completed target user information;
the second determining module is configured to obtain target user information of each user in the initial user group and historical behavior data of the user, and determine a recommendation scheme based on the target user information and the historical behavior data, where the historical behavior data includes: data of which services are purchased historically and events and time points triggering the purchase of the corresponding services;
the extraction module is used for sending the recommendation information of the service to be recommended to the initial user group based on the recommendation scheme, and extracting the user group characteristics with positive feedback and the user group characteristics with negative feedback to the recommendation information;
The third determining module is used for determining group difference characteristics and group common characteristics according to the group characteristics of the user group with positive feedback and the group characteristics of the user group with negative feedback, wherein the group characteristics are extracted based on target user information;
the adjustment module is used for adjusting the user characteristics and the characteristic weights in the initial user portrait according to the group difference characteristics and the group common characteristics to obtain a target user portrait;
and the recommending module is used for determining a target user group corresponding to the target user image and recommending the service to be recommended to the target user group.
9. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the business recommendation method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a computer program is stored, which, when being executed by a processor, causes the processor to perform the steps of the service recommendation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311094854.6A CN116796076B (en) | 2023-08-29 | 2023-08-29 | Service recommendation method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311094854.6A CN116796076B (en) | 2023-08-29 | 2023-08-29 | Service recommendation method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116796076A true CN116796076A (en) | 2023-09-22 |
| CN116796076B CN116796076B (en) | 2023-11-03 |
Family
ID=88045183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311094854.6A Active CN116796076B (en) | 2023-08-29 | 2023-08-29 | Service recommendation method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116796076B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302240A1 (en) * | 2010-06-03 | 2011-12-08 | Katsu Saito | Content recommendation system, content recommendation device, and content recommendation method |
| CN108108451A (en) * | 2017-12-27 | 2018-06-01 | 合肥美的智能科技有限公司 | The group of subscribers portrait acquisition methods and device of group |
| CN110795640A (en) * | 2019-10-12 | 2020-02-14 | 华中师范大学 | Adaptive group recommendation method for compensating group member difference |
| US20210118036A1 (en) * | 2019-10-16 | 2021-04-22 | Coupang Corp. | Computerized systems and methods for providing product recommendations |
| WO2021138823A1 (en) * | 2020-01-08 | 2021-07-15 | 深圳市欢太科技有限公司 | Information pushing method and apparatus, server, and storage medium |
| CN114036403A (en) * | 2022-01-07 | 2022-02-11 | 智者四海（北京）技术有限公司 | User interest detection method, device and storage medium |
| US20220286521A1 (en) * | 2020-12-28 | 2022-09-08 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus, computer device and medium |
| CN115203543A (en) * | 2022-06-29 | 2022-10-18 | 清华大学 | Content recommendation method, and training method and device of content recommendation model |
| WO2023050143A1 (en) * | 2021-09-29 | 2023-04-06 | 华为技术有限公司 | Recommendation model training method and apparatus |
-
2023
- 2023-08-29 CN CN202311094854.6A patent/CN116796076B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302240A1 (en) * | 2010-06-03 | 2011-12-08 | Katsu Saito | Content recommendation system, content recommendation device, and content recommendation method |
| CN108108451A (en) * | 2017-12-27 | 2018-06-01 | 合肥美的智能科技有限公司 | The group of subscribers portrait acquisition methods and device of group |
| CN110795640A (en) * | 2019-10-12 | 2020-02-14 | 华中师范大学 | Adaptive group recommendation method for compensating group member difference |
| US20210118036A1 (en) * | 2019-10-16 | 2021-04-22 | Coupang Corp. | Computerized systems and methods for providing product recommendations |
| WO2021138823A1 (en) * | 2020-01-08 | 2021-07-15 | 深圳市欢太科技有限公司 | Information pushing method and apparatus, server, and storage medium |
| US20220286521A1 (en) * | 2020-12-28 | 2022-09-08 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus, computer device and medium |
| WO2023050143A1 (en) * | 2021-09-29 | 2023-04-06 | 华为技术有限公司 | Recommendation model training method and apparatus |
| CN114036403A (en) * | 2022-01-07 | 2022-02-11 | 智者四海（北京）技术有限公司 | User interest detection method, device and storage medium |
| CN115203543A (en) * | 2022-06-29 | 2022-10-18 | 清华大学 | Content recommendation method, and training method and device of content recommendation model |
Non-Patent Citations (1)
| Title |
|---|
| 谷鹏;李琳;苏畅;袁景凌;: "面向群组的社交follow推荐方法研究", 小型微型计算机***, no. 05 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116796076B (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109271420B (en) | Information pushing method, device, computer equipment and storage medium | |
| WO2018040068A1 (en) | Knowledge graph-based semantic analysis system and method | |
| CN109509010B (en) | Multimedia information processing method, terminal and storage medium | |
| CN110807655A (en) | Advertisement bidding method, device and equipment | |
| CN111737443B (en) | Answer text processing method and device and key text determining method | |
| CN111191133B (en) | Service search processing method, device and equipment | |
| CN113505272B (en) | Control method and device based on behavior habit, electronic equipment and storage medium | |
| WO2020237898A1 (en) | Personalized recommendation method for online education system, terminal and storage medium | |
| CN112104505A (en) | Application recommendation method and device, server and computer-readable storage medium | |
| EP2958062A1 (en) | Determining multiple users of a network enabled device | |
| CN112202889B (en) | Information pushing method, device and storage medium | |
| CN112561636A (en) | Recommendation method, recommendation device, terminal equipment and medium | |
| CN116796076B (en) | Service recommendation method, device, equipment and storage medium | |
| CN112115354A (en) | Information processing method, information processing apparatus, server, and storage medium | |
| CN108830298B (en) | Method and device for determining user feature tag | |
| CN108875014B (en) | Precise project recommendation method based on big data and artificial intelligence and robot system | |
| CN108520012B (en) | Mobile internet user comment mining method based on machine learning | |
| CN116091157A (en) | Resource pushing method and device, storage medium and computer equipment | |
| CN108259588A (en) | A kind of method for pushing and device of the cultural cloud platform based on big data | |
| CN109885504B (en) | Recommendation system test method, device, medium and electronic equipment | |
| CN113792211A (en) | Resource pushing processing method and device, electronic equipment and storage medium | |
| CN113706298A (en) | Deferred service processing method and device | |
| CN112527991A (en) | Information processing method, apparatus and medium | |
| CN112036576A (en) | Data processing method and device based on data form and electronic equipment | |
| CN112328779A (en) | Training sample construction method and device, terminal equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |