CN113407827A - Information recommendation method, device, equipment and medium based on user value classification - Google Patents
Information recommendation method, device, equipment and medium based on user value classification Download PDFInfo
- Publication number
- CN113407827A CN113407827A CN202110656025.7A CN202110656025A CN113407827A CN 113407827 A CN113407827 A CN 113407827A CN 202110656025 A CN202110656025 A CN 202110656025A CN 113407827 A CN113407827 A CN 113407827A
- Authority
- CN
- China
- Prior art keywords
- user
- value
- data
- type
- consumption data
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
Abstract
The invention discloses an information recommendation method, device, equipment and medium based on user value classification. The method comprises the following steps: acquiring consumption data and a user portrait of a current user; determining a value analysis correction coefficient distribution strategy of each data type in the consumption data according to a preset mapping relation between the user portrait and the correction coefficient distribution strategy of each data type in the consumption data; determining a value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient; analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user; and recommending corresponding target information to the current user according to the user value type. By adopting the method and the device, the pertinence and the accuracy of information recommendation of the user can be effectively improved, so that a good use experience is provided for the user.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an information recommendation method, device, equipment and medium based on user value classification.
Background
With the development of internet technology and the proliferation of the number of users on the internet, the information on the internet grows exponentially, and the importance of recommending proper information to the users is increasingly highlighted. In recent years, recommendation systems, which are the main techniques for solving the problem of information explosion, have been widely developed and applied.
However, the conventional information recommendation is not fine enough at present, and is mainly based on the number of clicks and browsing time of the user on information such as goods and services. However, the inventors found that the prior art has at least the following problems: the existing information recommendation method does not consider the value type of the user, is difficult to truly reflect the information content of commodities, services and the like actually wanted by the user, and has poor user experience.
Disclosure of Invention
The embodiment of the invention aims to provide an information recommendation method, device, equipment and medium based on user value classification, which can effectively improve the pertinence and accuracy of information recommendation for users, thereby providing a good use experience for the users.
In order to achieve the above object, an embodiment of the present invention provides an information recommendation method based on user value classification, including:
acquiring consumption data and a user portrait of a current user;
determining a value analysis correction coefficient distribution strategy of each data type in the consumption data according to a preset mapping relation between the user portrait and the correction coefficient distribution strategy of each data type in the consumption data;
determining a value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient;
analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user;
and recommending corresponding target information to the current user according to the user value type.
As an improvement to the above, the user representation includes at least one of: age stage of the user, gender of the user, occupation of the user, income level of the user, preferences of the user.
As an improvement of the above scheme, the acquiring a user representation of a current user specifically includes:
acquiring behavior data of a current user; the behavior data is data generated by the operation behavior of the current user on a target platform;
determining a feature tag of the current user according to the behavior data; the feature tag is used for representing feature information of the current user;
and generating a user portrait of the current user according to the feature tag.
As an improvement of the above scheme, the analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user specifically includes:
analyzing the value degree of the consumption data of each data type to obtain the data value degree of each data type of the consumption data;
and analyzing a correction coefficient and a data value degree according to the value of each data type of the consumption data to obtain the user value type of the current user.
As an improvement of the above scheme, the analyzing a correction coefficient and a data value degree according to the value of each data type of the consumption data to obtain the user value type of the current user specifically includes:
calculating the value grade of each data type of the consumption data according to the data value degree and the value analysis correction coefficient of each data type of the consumption data;
and obtaining the user value type of the current user according to the mapping relation between the preset value grade and the user value type.
As an improvement of the above scheme, the analyzing the value degree of the consumption data of each data type to obtain the data value degree of each data type of the consumption data specifically includes:
and calculating the data value degree of each data type of the consumption data based on a preset value degree calculation strategy of the consumption data of each data type.
As an improvement of the above scheme, the recommending target information to the current user according to the user value type specifically includes:
acquiring corresponding information to be recommended according to the user value type and based on a preset mapping relation between the user value type and the recommendation information; wherein the recommendation information comprises at least one of: commodity information, service information;
selecting the first N pieces of information in all information to be recommended as target information; n is an integer, and N is more than or equal to 1;
and recommending the target information to the current user.
The embodiment of the invention provides an information recommendation device based on user value classification, which comprises the following components:
the user data acquisition module is used for acquiring consumption data and a user portrait of a current user;
the correction coefficient distribution strategy determining module is used for determining a distribution strategy of the value analysis correction coefficient of each data type in the consumption data according to the mapping relation between the preset user portrait and the correction coefficient distribution strategy of each data type in the consumption data;
the value analysis correction coefficient determining module is used for determining a value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient;
the user value type obtaining module is used for analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user;
and the target information recommending module is used for recommending corresponding target information to the current user according to the user value type.
The information recommendation device based on the user value classification comprises a processor, a memory and a computer program which is stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the information recommendation method based on the user value classification is realized.
The embodiment of the invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute any one of the above information recommendation methods based on user value classification.
Compared with the prior art, the information recommendation method, the device, the equipment and the medium based on the user value classification disclosed by the embodiment of the invention have the advantages that the consumption data and the user portrait of the current user are obtained, the corresponding value analysis correction coefficient is distributed to each data type in the consumption data of the current user according to the user portrait of the current user, the value analysis is carried out on the consumption data according to the value analysis correction coefficient of each data type of the consumption data, the user value type of the current user is obtained, so that the importance degree of the consumption data of each data type is related to the user portrait in the user value analysis process, the rationality of the user value analysis is improved, the accuracy of determining the user value type is improved, and the corresponding target information can be recommended to the current user according to the user value type, the target information recommended to the user can truly reflect the information content of commodities, services and the like actually wanted by the user, the accuracy rate of information recommendation is improved, and the use experience of the user is effectively improved.
Drawings
FIG. 1 is a schematic flowchart of an information recommendation method based on user value classification according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process for obtaining a user representation according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of determining a user value type in an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an information recommendation device based on user value classification according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an information recommendation device based on user value classification according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of an information recommendation method based on user value classification according to an embodiment of the present invention. The embodiment of the invention provides an information recommendation method based on user value classification, which is specifically executed through steps S11 to S15:
and S11, acquiring consumption data and a user portrait of the current user.
In particular, said consumption data comprises at least: the time interval between the payment time nearest to the current time and the current time, the consumption frequency and the consumption amount in a preset time period.
The user representation is some characteristic attributes extracted from real user behaviors and forms a user model, and the characteristic attributes represent different user types and similar characteristics of the user types. The user representation includes at least one of: age stage of the user, gender of the user, occupation of the user, income level of the user, preferences of the user.
In the embodiment of the invention, the consumption data of the current user is obtained by acquiring the payment time interval, the consumption frequency and the consumption amount of the current user on a certain platform within the past preset time period. And constructing the user portrait of the current user based on the operation behavior data of the user on the platform.
It is understood that the age stage, sex, occupation, etc. included in the user representation may be obtained by acquiring basic information of the user, or may be inferred by operation behavior data of the user on the platform; the information of income level, preference and the like of the user included in the user portrait is obtained by predicting the operation behavior data of the user on the platform.
S12, determining the distribution strategy of the value analysis correction coefficient of each data type in the consumption data according to the mapping relation between the preset user portrait and the correction coefficient distribution strategy of each data type in the consumption data.
S13, determining the value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient.
The method comprises the steps of establishing a mapping relation of 'user portrait-consumption data type-correction coefficient distribution strategy' in advance, wherein the correction coefficient distribution strategy refers to value analysis correction coefficients given to consumption data of different data types of a user in the process of carrying out value analysis on the user, and is used for measuring the importance degree of the consumption data of each data type, so that the correction coefficient distribution strategy corresponding to each type of the obtained consumption data in the consumption data can be determined according to the user portrait of the current user, and corresponding value analysis correction coefficients are distributed to each data type in the consumption data.
And S14, analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user.
Specifically, value analysis is performed on the consumption data of each data type in the consumption data of the current user through a pre-established user value analysis model, and a value analysis result of the consumption data of each data type is obtained by combining the distributed value analysis correction coefficients, so that the user value type of the current user is determined.
And S15, recommending corresponding target information to the current user according to the user value type.
After the user value type of the current user is determined, target information corresponding to the user value type, such as commodity information and service information in a market, can be acquired and recommended to the user.
The embodiment of the invention provides an information recommendation method based on user value classification, which allocates a corresponding value analysis correction coefficient to each data type in consumption data of a current user according to a user image of the current user, performs value analysis on the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain a user value type of the current user, so that the importance degree of the consumption data of each data type is related to the user image in the user value analysis process, the rationality of user value analysis is improved, the accuracy of determining the user value type is improved, corresponding target information can be recommended to the current user according to the user value type, and the target information recommended to the user can truly reflect information contents such as commodities or services and the like actually wanted by the user, the accuracy of information recommendation is improved, and the use experience of the user is effectively improved.
In a preferred embodiment, referring to fig. 2, a schematic flow chart of acquiring a user portrait according to an embodiment of the present invention is shown. In step S11, the acquiring a user portrait of the current user specifically includes steps S21 to S23:
s21, acquiring the behavior data of the current user; the behavior data is data generated by the operation behavior of the current user on a target platform.
S22, determining the feature label of the current user according to the behavior data; wherein the feature tag is used for characterizing feature information of the current user.
And S23, generating the user portrait of the current user according to the feature tag.
As an example, a user representation has eight analysis dimensions: p stands for basic (Primary): whether the user representation is based on a contextual interview of a real user; e represents homology (Empathy): the user portrait contains description related to names, photos and commodities, and whether the user portrait has the same sense of mind or not; r stands for authenticity (Realistic): whether the user portrait looks like a real person or not; s stands for uniqueness (singleplex): whether each user is unique, with little similarity to each other; o stands for objective (Objectives): whether the user representation contains a high-level target related to the commodity or not and whether the user representation contains a keyword to describe the target or not; n represents numerical (Number): whether the number of user representations is sufficiently small so that the design team can remember the name of each user representation, and one of the primary user representations; a stands for applicability (Applicable): whether a design team can make a design decision using the user representation as a utility tool; l represents permanence (Long): longevity of the user tag.
In the embodiment of the invention, behavior data generated by operation behaviors of a current user on a target platform is acquired, wherein the behavior data can be transaction data such as transaction time of the current user on the target platform, transaction frequency in a preset time period, transaction amount in the preset time period and the like; or the current user may be the operation data of the commodities, such as the browsing time period, the browsing duration, the titles of the browsed commodities, the additional purchasing, the collection, the attention and the like, of the information of a certain commodity on the target platform. And determining a corresponding matched feature tag for characterizing feature information of the current user by performing statistical analysis on the behavior data. According to the feature tag, a user portrait of the current user can be determined.
By adopting the technical means of the embodiment of the invention, the behavior data of the user, such as consumption behavior, commodity operation behavior and other data, is used for constructing the user portrait, so that the essential characteristics of the user can be more accurately revealed from the consumption behavior level of the user, a more specific user portrait is generated, and finally commodity or service information more conforming to the individuation of the user can be recommended to the user.
In another embodiment, the obtaining of the user representation of the current user specifically includes steps S21 'to S26':
s21', establishing a historical user portrait of the current user according to the historical feature label of the current user;
s22', establishing a temporary user portrait of the current user according to the inherent feature tag in the historical feature tags; wherein the intrinsic characteristic label comprises at least one of an age label and a gender label;
s23', acquiring other feature labels except the inherent feature label of the current user within preset time;
s24', obtaining the matching degree of other feature labels except the inherent feature label of the current user in the preset time and other feature labels except the inherent feature label in the historical feature labels;
s25', if the matching degree is larger than the preset threshold value, using the historical user portrait as the user portrait of the current user;
s26', if the matching degree is not larger than the preset matching degree threshold value, establishing the latest user portrait of the current user according to other feature labels except the inherent feature label of the current user in the preset time and the inherent feature label of the current user, and taking the latest user portrait as the user portrait of the current user.
When a user image is acquired, a user image including an inherent tag, a behavior tag, and the like is created in advance from previously accumulated feature tag data. The inherent labels comprise inherent attributes of the users such as the gender, the age and the like of the users, and can be collected from the users in links such as user registration; the behavior tags comprise the behaviors of interactive operations of adding concern, canceling concern, adding a wish list, taking out the wish list, forming an order, canceling the order, paying, refunding and the like of the user, and are acquired after the authorization of the user. When a user is newly on line, a temporary user portrait is established for the user, and inherits the inherent label from the historical user portrait to reflect the inherent attributes of the user such as gender, age and the like; and the feature tag reflecting the user behavior in the temporary user portrait is obtained by acquiring the behavior in the preset time, and the matching degree of the newly acquired feature tag data and the previously accumulated feature tag data is judged. When the matching degree is larger than a preset matching degree threshold value, the latest behavior of the user is higher in consistency with the historical user portrait, so that the historical user portrait can be used as the user portrait; and when the matching degree is not greater than the preset matching degree threshold value, which indicates that the latest behavior of the user deviates from the historical user portrait, establishing the latest user portrait again according to the inherent label and the newly acquired feature label, and replacing the historical user portrait with the latest user portrait to serve as the user portrait.
By adopting the technical means of the embodiment, the user portrait can be effectively maintained and updated according to the latest behavior data of the user, particularly, under the condition that the user behavior changes suddenly, the latest user portrait of the user can be quickly established by utilizing the temporary user portrait and the latest behavior preference of the user, and the latest user portrait can more accurately reflect the latest preference of the user, so that the accuracy of information recommendation is improved.
As a preferred implementation manner, refer to fig. 3, which is a schematic flow chart illustrating the determination of the user value type in the embodiment of the present invention. The step S14 specifically includes steps S31 and S32:
and S31, analyzing the value degree of the consumption data of each data type to obtain the data value degree of each data type of the consumption data.
A monovalent value degree analysis model is constructed and trained in advance, consumption data of each data type is used as input, value degree analysis is carried out, and accordingly corresponding data value degrees are output.
And S32, analyzing the correction coefficient and the data value degree according to the value of each data type of the consumption data to obtain the user value type of the current user.
And analyzing a correction coefficient according to the value distributed by each data type, correcting the data value degree of the consumption data of each data type, and determining the user value type of the current user by combining the corrected results of the consumption data of all data types.
Specifically, the user value types may be divided into: general detained players, general developed players, general maintenance players, general value players, important detained players, important developed players, important maintenance players, important value players.
Preferably, step S31 is:
s311, calculating the data value degree of each data type of the consumption data based on the preset value degree calculation strategy of the consumption data of each data type.
Specifically, different user consumption data have corresponding value degree calculation strategies, and in the embodiment of the present invention, three value degree calculation strategies correspond to three consumption data, that is, a time interval between a payment time nearest to a current time and the current time, consumption frequency and consumption amount within a predetermined time period of a user, and are respectively: a time interval value degree calculation strategy, a consumption frequency value degree calculation strategy and a consumption amount value degree calculation strategy.
Specifically, the user is scored according to the consumption frequency value degree calculation strategy, the data value degree of the consumption frequency is calculated, and the following description is given by taking a 5-point system as an example:
new user (1 point): the consumption frequency within a period of time is within a preset minimum consumption frequency range, the user is judged as a new user, the retention rate of the user is extremely low, and the difference with other users with consumption frequency is large;
primary user (2 points): the consumption frequency in a period of time is within a preset primary consumption frequency range, the primary user is judged, and the retention rate of the part of users is increased relative to that of new users, and the range is large;
growing user (3 points): the consumption frequency within a period of time is within a preset intermediate consumption frequency range, a growing user is judged, the retention rate of the part of users is high, and a growing state with large amplitude is presented;
mature user (4 points): the consumption frequency within a period of time is within a preset advanced consumption frequency range, the user is judged to be a mature user, and the retention rate of the part of users is kept in a second highest state and is relatively stable;
faithful user (5 points): and the consumption frequency within a period of time is within the preset optimal consumption frequency range, the user is judged as a faithful user, and the retention rate of the part of users is kept in a high state and is stable.
Specifically, the average consumption time interval of each player is calculated according to the consumption frequency value degree calculation strategy, the consumption frequency of the users is binned, the consumption conditions and the consumption time interval conditions of the users with different consumption frequency data value degrees are checked, and the explanation is given by taking the case that 80% quantiles of consumption time intervals r of the players with consumption frequency fs of 5 cents, 4 cents and 3 cents (the total consumption amount of the player group accounts for more than 85% of the total consumption amount) are the last payment time interval binning threshold (4 cents) as an example:
active users (4 points): the player consumption interval of r & ltbecoming & gt 1 & ltbecoming & gt, fs & ltbecoming & gt 5 & lt & gt is t1 days with 80% quantile;
silent users (score 3): 1< r ═ 2, fs ═ 4 players consumed at 80% quantile t2 days;
early warning user (2 points): 2< r < -5, fs 4, with an 80% quantile t3 days between consumption;
attrition users (1 point): r >5, others as attrition players.
Through the value degree calculation strategy, the consumption data of the user, including the consumption frequency, the last consumption time interval and the consumption amount, can be subjected to corresponding value degree analysis, and the value degree of each data type is scored, so that the data value degree of each data type is obtained, and the value degree, the loyalty degree and the active condition of player payment are determined.
Step S32 includes:
s321, calculating the value grade of each data type of the consumption data according to the data value degree and the value analysis correction coefficient of each data type of the consumption data;
s322, obtaining the user value type of the current user according to the mapping relation between the preset value grade and the user value type.
After the data value degree of each data type of the consumption data of the current user is calculated, the data value degree is corrected according to the value analysis correction coefficient distributed by each data type, and the value score of each data type is obtained. For example, if the data worth degree of the consumption frequency of the current user is 3 points and the corresponding value analysis correction coefficient is 1.2, the value score of the consumption frequency is 3 × 1.2 to 3.6 points.
Determining a value grade corresponding to each data type according to the value score of the consumption frequency of the current user, the value score of the last consumption time interval and the value score of the consumption amount, wherein the value score of the consumption frequency is divided into 5 grades, the corresponding value grade comprises high and low, the value grade is larger than or equal to 4 grades, and the rest grades are low; the time interval value is divided into 4 grades, the corresponding value grades comprise far and near, wherein the value grade is more than or equal to 3 grades, the distance is divided, and the rest grades are near; the price value of the consumption amount is divided into 3 grades, the corresponding value grades comprise high and low, wherein the value grade is more than or equal to 2 grades, the high grade is divided, and the rest grades are low. According to the above division criteria, the user value types can be obtained by division as shown in table 1.
TABLE 1 user value type partitioning
By adopting the technical means of the embodiment of the invention, the consumption data of each data type is analyzed for the value degree through the preset value degree calculation strategy to obtain the data value degree of each data type of the consumption data, so that all value representations of users can be represented dynamically, and the accuracy of value degree analysis of the consumption data of different data types can be improved. And then, correcting the corresponding data value degree according to the value analysis correction coefficient of each data type of the consumption data, and analyzing to obtain the user value type of the current user. And recommending corresponding target information to the current user according to the user value type, so that the target information recommended to the user can truly reflect the information content of commodities, services and the like actually desired by the user, the accuracy of information recommendation is improved, and the use experience of the user is effectively improved.
As a preferred embodiment, the step S15 specifically includes steps S151 to S153:
s151, acquiring corresponding information to be recommended according to the user value type and based on a preset mapping relation between the user value type and the recommendation information; wherein the recommendation information comprises at least one of: commodity information, service information;
s152, selecting the first N pieces of information in all information to be recommended as target information; n is an integer, and N is more than or equal to 1;
s153, recommending the target information to the current user.
Specifically, in the preset mapping relationship, a specific user value type has a plurality of pieces of corresponding information to be recommended, each piece of information to be recommended has a recommendation appropriateness, and the higher the recommendation appropriateness is, the higher the probability of representing that the user is attached to the taste of the user is, the lower the recommendation appropriateness is, the lower the probability of representing that the user is attached to the taste of the user is. At this time, the N pieces of information with the highest recommendation appropriateness need to be selected from all pieces of information to be recommended as target information to be recommended to the user.
It should be noted that the value of N may be preset or may be modified by a user.
Fig. 4 is a schematic structural diagram of an information recommendation apparatus based on user value classification according to an embodiment of the present invention. The embodiment of the present invention provides an information recommendation device 40 based on user value classification, including: a user data acquisition module 41, a correction coefficient distribution strategy determination module 42, a value analysis correction coefficient determination module 43, a user value type acquisition module 44 and a target information recommendation module 45; wherein the content of the first and second substances,
the user data acquisition module 41 is configured to acquire consumption data and a user representation of a current user;
the modification coefficient distribution strategy determining module 42 is configured to determine a distribution strategy of the value analysis modification coefficient of each data type in the consumption data according to a mapping relationship between a preset user portrait and a modification coefficient distribution strategy of each data type in the consumption data;
the value analysis correction coefficient determining module 43 is configured to determine a value analysis correction coefficient for each data type in the consumption data according to a distribution policy of the value analysis correction coefficient;
the user value type obtaining module 44 is configured to perform value analysis on the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain a user value type of the current user;
and the target information recommending module 45 is configured to recommend corresponding target information to the current user according to the user value type.
The embodiment of the invention provides an information recommendation device based on user value classification, which is characterized in that a corresponding value analysis correction coefficient is distributed for each data type in consumption data of a current user according to a user image of the current user, and the value analysis correction coefficient is carried out on the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user, so that the importance degree of the consumption data of each data type is related to the user image in the user value analysis process, the rationality of user value analysis is improved, and the accuracy of determining the user value type is improved. And then, recommending corresponding target information to the current user according to the user value type, so that the target information recommended to the user can truly reflect the information content of commodities, services and the like actually desired by the user, the accuracy of information recommendation is improved, and the use experience of the user is effectively improved.
As a preferred embodiment, user data acquisition module 41 includes a consumption data acquisition unit for acquiring consumption data of a current user, and a user representation acquisition unit for acquiring a user representation of the current user;
the user portrait acquisition unit is specifically configured to:
acquiring behavior data of a current user; the behavior data is data generated by the operation behavior of the current user on a target platform;
determining a feature tag of the current user according to the behavior data; the feature tag is used for representing feature information of the current user;
and generating a user portrait of the current user according to the feature tag.
By adopting the technical means of the embodiment of the invention, the behavior data of the user, such as consumption behavior, commodity operation behavior and other data, is used for constructing the user portrait, so that the essential characteristics of the user can be more accurately revealed from the consumption behavior level of the user, a more specific user portrait is generated, and finally commodity or service information more conforming to the individuation of the user can be recommended to the user.
As a preferred embodiment, the user value type obtaining module 44 is specifically configured to:
calculating the data value degree of each data type of the consumption data based on a preset calculation strategy of the value degree of the consumption data of each data type;
calculating the value grade of each data type of the consumption data according to the data value degree and the value analysis correction coefficient of each data type of the consumption data;
and obtaining the user value type of the current user according to the mapping relation between the preset value grade and the user value type.
Specifically, the user value types may be divided into: general detained players, general developed players, general maintenance players, general value players, important detained players, important developed players, important maintenance players, important value players.
And carrying out corresponding value degree analysis on the consumption data of the user, including the consumption frequency, the last consumption time interval and the consumption amount in a preset time period, according to a preset value degree calculation strategy of the consumption data of each data type, and grading the value degree of each data type, so as to obtain the data value degree of each data type, and the data value degree is used for determining the value and the loyalty of a player and the active condition of the payment of the player. After the data value degree of each data type of the consumption data of the current user is calculated, the value analysis correction coefficient is distributed according to each data type to obtain the value score of each data type. And determining a value grade corresponding to each data type according to the value score of the consumption frequency of the current user, the value score of the last consumption time interval and the value score of the consumption amount, and determining the user value type based on the preset mapping relation between the value grade and the user value type.
By adopting the technical means of the embodiment of the invention, the consumption data of each data type is analyzed for the value degree through the preset value degree calculation strategy to obtain the data value degree of each data type of the consumption data, so that all value representations of users can be represented dynamically, and the accuracy of value degree analysis of the consumption data of different data types can be improved. And then, correcting the corresponding data value degree according to the value analysis correction coefficient of each data type of the consumption data, and analyzing to obtain the user value type of the current user. And recommending corresponding target information to the current user according to the user value type, so that the target information recommended to the user can truly reflect the information content of commodities, services and the like actually desired by the user, the accuracy of information recommendation is improved, and the use experience of the user is effectively improved.
It should be noted that the information recommendation device based on user value classification according to the embodiment of the present invention is used for executing all the process steps of the information recommendation method based on user value classification according to the above embodiment, and the working principles and beneficial effects of the two are in one-to-one correspondence, so that details are not repeated.
Fig. 5 is a schematic structural diagram of an information recommendation device based on user value classification according to an embodiment of the present invention. The information recommendation device 50 comprises a processor 51, a memory 52 and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the information recommendation method based on the user value classification as described in the above embodiment.
The embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute the information recommendation method based on user value classification according to the above embodiment.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. An information recommendation method based on user value classification is characterized by comprising the following steps:
acquiring consumption data and a user portrait of a current user;
determining a value analysis correction coefficient distribution strategy of each data type in the consumption data according to a preset mapping relation between the user portrait and the correction coefficient distribution strategy of each data type in the consumption data;
determining a value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient;
analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user;
and recommending corresponding target information to the current user according to the user value type.
2. The method of claim 1, wherein the user representation comprises at least one of: age stage of the user, gender of the user, occupation of the user, income level of the user, preferences of the user.
3. The method as claimed in claim 1 or 2, wherein the obtaining of the user representation of the current user specifically comprises:
acquiring behavior data of a current user; the behavior data is data generated by the operation behavior of the current user on a target platform;
determining a feature tag of the current user according to the behavior data; the feature tag is used for representing feature information of the current user;
and generating a user portrait of the current user according to the feature tag.
4. The information recommendation method based on user value classification as claimed in claim 1, wherein the performing a value analysis on the consumption data according to the value analysis modification coefficient of each data type of the consumption data to obtain the user value type of the current user specifically comprises:
analyzing the value degree of the consumption data of each data type to obtain the data value degree of each data type of the consumption data;
and analyzing a correction coefficient and a data value degree according to the value of each data type of the consumption data to obtain the user value type of the current user.
5. The information recommendation method based on user value classification as claimed in claim 4, wherein the analyzing the modification factor and the data value degree according to the value of each data type of the consumption data to obtain the user value type of the current user specifically comprises:
calculating the value grade of each data type of the consumption data according to the data value degree and the value analysis correction coefficient of each data type of the consumption data;
and obtaining the user value type of the current user according to the mapping relation between the preset value grade and the user value type.
6. The information recommendation method based on user value classification as claimed in claim 4, wherein the analyzing the value degree of the consumption data of each data type to obtain the data value degree of each data type of the consumption data specifically comprises:
and calculating the data value degree of each data type of the consumption data based on a preset value degree calculation strategy of the consumption data of each data type.
7. The information recommendation method based on user value classification according to any one of claims 1, 2, 4-6, wherein the recommending target information to the current user according to the user value type specifically comprises:
acquiring corresponding information to be recommended according to the user value type and based on a preset mapping relation between the user value type and the recommendation information; wherein the recommendation information comprises at least one of: commodity information, service information;
selecting the first N pieces of information in all information to be recommended as target information; n is an integer, and N is more than or equal to 1;
and recommending the target information to the current user.
8. An information recommendation apparatus classified based on user value, comprising:
the user data acquisition module is used for acquiring consumption data and a user portrait of a current user;
the correction coefficient distribution strategy determining module is used for determining a distribution strategy of the value analysis correction coefficient of each data type in the consumption data according to the mapping relation between the preset user portrait and the correction coefficient distribution strategy of each data type in the consumption data;
the value analysis correction coefficient determining module is used for determining a value analysis correction coefficient of each data type in the consumption data according to the distribution strategy of the value analysis correction coefficient;
the user value type obtaining module is used for analyzing the value of the consumption data according to the value analysis correction coefficient of each data type of the consumption data to obtain the user value type of the current user;
and the target information recommending module is used for recommending corresponding target information to the current user according to the user value type.
9. An information recommendation device based on user value classification, characterized by comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the information recommendation method based on user value classification according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, comprising a stored computer program, wherein when the computer program runs, the computer-readable storage medium controls a device to execute the information recommendation method based on user value classification according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110656025.7A CN113407827A (en) | 2021-06-11 | 2021-06-11 | Information recommendation method, device, equipment and medium based on user value classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110656025.7A CN113407827A (en) | 2021-06-11 | 2021-06-11 | Information recommendation method, device, equipment and medium based on user value classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113407827A true CN113407827A (en) | 2021-09-17 |
Family
ID=77683680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110656025.7A Pending CN113407827A (en) | 2021-06-11 | 2021-06-11 | Information recommendation method, device, equipment and medium based on user value classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113407827A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993396A (en) * | 2023-09-27 | 2023-11-03 | 北京北汽鹏龙汽车服务贸易股份有限公司 | Risk early warning method based on vehicle user tag and computer equipment |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160342911A1 (en) * | 2015-05-19 | 2016-11-24 | 24/7 Customer, Inc. | Method and system for effecting customer value based customer interaction management |
| CN106372775A (en) * | 2016-08-19 | 2017-02-01 | 国网冀北电力有限公司电力科学研究院 | Assessment method and system of comprehensive value of power grid client |
| US20170148081A1 (en) * | 2015-11-20 | 2017-05-25 | Mastercard International Incorporated | Method and system for recommending relevant merchants for a consumer at a given geographical location by evaluating the strength of the intersect between consumer vectors and merchant vectors |
| CN106980663A (en) * | 2017-03-21 | 2017-07-25 | 上海星红桉数据科技有限公司 | Based on magnanimity across the user's portrait method for shielding behavioral data |
| CN107292465A (en) * | 2016-03-31 | 2017-10-24 | 阿里巴巴集团控股有限公司 | User's evaluation method, device and equipment |
| CN108596720A (en) * | 2018-04-23 | 2018-09-28 | 广东奥园奥买家电子商务有限公司 | A method of commercial product recommending is carried out according to the behavioral data of user |
| CN108776931A (en) * | 2018-04-13 | 2018-11-09 | 上海琢学科技有限公司 | Financial client based on RFM and Canopy is worth loyalty divided method |
| US10157411B1 (en) * | 2014-03-13 | 2018-12-18 | Amazon Technologies, Inc. | Recommendation system that relies on RFM segmentation |
| CN109359244A (en) * | 2018-10-30 | 2019-02-19 | 中国科学院计算技术研究所 | A kind of recommendation method for personalized information and device |
| CN109993380A (en) * | 2017-12-29 | 2019-07-09 | 中国移动通信集团黑龙江有限公司 | A kind of information processing method, device and computer readable storage medium |
| CN112001754A (en) * | 2020-08-21 | 2020-11-27 | 上海风秩科技有限公司 | User portrait generation method, device, equipment and computer readable medium |
| CN112256973A (en) * | 2020-11-04 | 2021-01-22 | 贝壳技术有限公司 | User portrait correction method, device, medium, and electronic apparatus |
| CN112418916A (en) * | 2020-11-09 | 2021-02-26 | 北京三快在线科技有限公司 | Information providing method and device |
| CN112418956A (en) * | 2020-12-16 | 2021-02-26 | 国网雄安金融科技集团有限公司 | Financial product recommendation method and device |
| CN112612934A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | User charging behavior portrait processing method and device |
-
2021
- 2021-06-11 CN CN202110656025.7A patent/CN113407827A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10157411B1 (en) * | 2014-03-13 | 2018-12-18 | Amazon Technologies, Inc. | Recommendation system that relies on RFM segmentation |
| US20160342911A1 (en) * | 2015-05-19 | 2016-11-24 | 24/7 Customer, Inc. | Method and system for effecting customer value based customer interaction management |
| US20170148081A1 (en) * | 2015-11-20 | 2017-05-25 | Mastercard International Incorporated | Method and system for recommending relevant merchants for a consumer at a given geographical location by evaluating the strength of the intersect between consumer vectors and merchant vectors |
| CN107292465A (en) * | 2016-03-31 | 2017-10-24 | 阿里巴巴集团控股有限公司 | User's evaluation method, device and equipment |
| CN106372775A (en) * | 2016-08-19 | 2017-02-01 | 国网冀北电力有限公司电力科学研究院 | Assessment method and system of comprehensive value of power grid client |
| CN106980663A (en) * | 2017-03-21 | 2017-07-25 | 上海星红桉数据科技有限公司 | Based on magnanimity across the user's portrait method for shielding behavioral data |
| CN109993380A (en) * | 2017-12-29 | 2019-07-09 | 中国移动通信集团黑龙江有限公司 | A kind of information processing method, device and computer readable storage medium |
| CN108776931A (en) * | 2018-04-13 | 2018-11-09 | 上海琢学科技有限公司 | Financial client based on RFM and Canopy is worth loyalty divided method |
| CN108596720A (en) * | 2018-04-23 | 2018-09-28 | 广东奥园奥买家电子商务有限公司 | A method of commercial product recommending is carried out according to the behavioral data of user |
| CN109359244A (en) * | 2018-10-30 | 2019-02-19 | 中国科学院计算技术研究所 | A kind of recommendation method for personalized information and device |
| CN112001754A (en) * | 2020-08-21 | 2020-11-27 | 上海风秩科技有限公司 | User portrait generation method, device, equipment and computer readable medium |
| CN112256973A (en) * | 2020-11-04 | 2021-01-22 | 贝壳技术有限公司 | User portrait correction method, device, medium, and electronic apparatus |
| CN112418916A (en) * | 2020-11-09 | 2021-02-26 | 北京三快在线科技有限公司 | Information providing method and device |
| CN112612934A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | User charging behavior portrait processing method and device |
| CN112418956A (en) * | 2020-12-16 | 2021-02-26 | 国网雄安金融科技集团有限公司 | Financial product recommendation method and device |
Non-Patent Citations (4)
| Title |
|---|
| 冯娟娟等: "基于客户画像和GBDT算法的客户价值预测方法", 《洛阳理工学院学报(自然科学版)》 * |
| 刘小红等: "基于RFM模型的品牌服装会员价格容忍度和忠诚度的关联", 《服装学报》 * |
| 包志强等: "基于改进RFM模型的百度外卖客户价值分析", 《西安邮电大学学报》 * |
| 游晋峰: "基于RFM的商场会员画像描绘模型", 《太 原 师 范 学 院 学 报 （自然科学版）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993396A (en) * | 2023-09-27 | 2023-11-03 | 北京北汽鹏龙汽车服务贸易股份有限公司 | Risk early warning method based on vehicle user tag and computer equipment |
| CN116993396B (en) * | 2023-09-27 | 2023-12-22 | 北京北汽鹏龙汽车服务贸易股份有限公司 | Risk early warning method based on vehicle user tag and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11663493B2 (en) | Method and system of dynamic model selection for time series forecasting | |
| CN108564286B (en) | Artificial intelligent financial wind-control credit assessment method and system based on big data credit investigation | |
| US8930247B1 (en) | System and methods for content-based financial decision making support | |
| US7062477B2 (en) | Information-processing apparatus, information-processing method and storage medium | |
| US20080154668A1 (en) | System for monitoring a service provider partner | |
| CN109784779B (en) | Financial risk prediction method, device and storage medium | |
| CN111967971B (en) | Bank customer data processing method and device | |
| CN103927615A (en) | Associating An Entity With A Category | |
| CN109636482B (en) | Data processing method and system based on similarity model | |
| CN113469730A (en) | Customer repurchase prediction method and device based on RF-LightGBM fusion model under non-contract scene | |
| CN111429214B (en) | Transaction data-based buyer and seller matching method and device | |
| CN115280353A (en) | Fund combination updating method, apparatus, device and medium based on artificial intelligence | |
| Yuping et al. | New methods of customer segmentation and individual credit evaluation based on machine learning | |
| US20170345096A1 (en) | Method and system for providing a dashboard for determining resource allocation for marketing | |
| CN113407827A (en) | Information recommendation method, device, equipment and medium based on user value classification | |
| CN111445280A (en) | Model generation method, restaurant ranking method, system, device and medium | |
| US10210530B1 (en) | Selecting a report | |
| CN113537731B (en) | Design resource capability assessment method based on reinforcement learning | |
| CN112232945B (en) | Method and device for determining personal client credit | |
| CN114943582A (en) | Information recommendation method and system and recommendation server | |
| CN113762415A (en) | Neural network-based intelligent matching method and system for automobile financial products | |
| CN113793220A (en) | Stock market investment decision method based on artificial intelligence model and related equipment | |
| CN104966212B (en) | The detection method and application process of user's mobile degree | |
| Adedapo | Analysis of Non-Oil Exports–Economic Growth Relationship in Nigeria: The Role of Institutional Qualities | |
| CN113742492B (en) | Insurance scheme generation method and device, electronic 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 |