CN114996585A - Service information recommendation method, device, equipment and readable storage medium - Google Patents
Service information recommendation method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN114996585A CN114996585A CN202210751698.5A CN202210751698A CN114996585A CN 114996585 A CN114996585 A CN 114996585A CN 202210751698 A CN202210751698 A CN 202210751698A CN 114996585 A CN114996585 A CN 114996585A
- Authority
- CN
- China
- Prior art keywords
- data
- service
- recommendation model
- recommendation
- time
- 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
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a service information recommendation method, a device, equipment and a readable storage medium, which are applied to the technical field of data analysis. The method comprises the following steps: acquiring current life cycle data of a user and a plurality of material operation target data of a current page; inputting the current life cycle data and a plurality of material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model; determining a maximum match score from a plurality of the match scores; determining material operation target data to be displayed corresponding to the maximum matching score; and displaying the material operation target data to be displayed on the current page. By the service information recommendation scheme provided by the invention, the push result of the service application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Description
Technical Field
The present invention relates to the field of data analysis technologies, and in particular, to a method, an apparatus, a device, and a readable storage medium for recommending service information.
Background
When the user uses the business application program, the business application program can push business and product information to the user, and the purpose is to enable the user to know the pushed business and product and finally generate transaction behaviors. Based on this, the business application needs to recommend the favorite products or businesses to the user in a targeted manner.
In the prior art, most business applications infer the preference degree of a user based on the click behavior of the user in a recommendation result, and the exposure rate and the conversion rate of the recommendation result. Thus resulting in often pushing recommendations to the user that do not conform to their life stage and product use stage. The core requirements of the user are not caught, and the user resources are wasted.
Disclosure of Invention
In order to solve the above problems, the present application provides a service information recommendation method, device, apparatus, and readable storage medium.
In a first aspect, an embodiment of the present application provides a method for recommending service information, where the method includes:
acquiring current life cycle data of a user and a plurality of material operation target data of a current page;
inputting the current life cycle data and the material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
determining a maximum match score from a plurality of the match scores;
determining the material operation target data to be displayed corresponding to the maximum matching score;
and displaying the material operation target data to be displayed on the current page.
In a specific embodiment, the obtaining step of the target recommendation model includes:
acquiring service inferred data and service actual data at the time T;
inputting the service inferred data and the service actual data at the time T into an initial algorithm to obtain a recommendation model at the time T;
and determining the target recommendation model according to the recommendation model at the time T.
In a specific embodiment, the step of determining the target recommendation model according to the recommendation model at time T includes:
acquiring service inferred data and service actual data at the moment T + 1;
inputting the service inferred data and the service actual data at the T +1 moment into the initial algorithm to obtain a recommendation model at the T +1 moment;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
In a specific embodiment, the step of determining an optimal recommendation model from the recommendation model at time T and the recommendation model at time T +1 includes:
calling the recommendation model at the time T and the recommendation model at the time T + 1;
respectively calculating a first model score of the recommendation model at the time T and a second model score of the recommendation model at the time T + 1;
and determining the optimal recommendation model according to the first model score and the second model score.
In a specific embodiment, the step of obtaining the business inference data and the business actual data includes:
acquiring original data of a user;
and extracting the service inference data and the service actual data from the user raw data.
In a second aspect, an embodiment of the present application further provides a service information recommendation device, including:
the acquisition module is used for acquiring the current life cycle data of a user and a plurality of material operation target data of a current page;
the input module is used for inputting the current life cycle data and the material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
a first determining module for determining a maximum match score from a plurality of said match scores;
the second determining module is used for determining the material operation target data to be displayed corresponding to the maximum matching score;
and the pushing module is used for displaying the material operation target data to be displayed on the current page.
In a specific embodiment, the apparatus further comprises:
the training module is used for acquiring service inferred data and service actual data at the time T;
inputting the service inferred data and the service actual data at the T moment into an initial algorithm to obtain a recommendation model at the T moment;
and determining the target recommendation model according to the recommendation model at the time T.
In a specific embodiment, the training module is further configured to obtain service inference data and service actual data at a time T + 1;
inputting the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
In a third aspect, an embodiment of the present application provides a computer device, where the computer device includes a memory and a processor, where the memory stores a computer program, and the computer program executes the service information recommendation method according to the first aspect when the processor runs.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the service information recommendation method according to the first aspect is implemented.
The service information recommendation method, the device, the equipment and the readable storage medium are provided according to the embodiment. By training a recommendation model in advance. When a user accesses a business application program, life cycle data and material operation target data of the user are obtained, the life cycle data and the material operation target data of the user are input into a recommendation model, matching scores between the material operation target data and the life cycle data of the user can be obtained, and then the material operation target data are pushed for the user according to the matching scores. The pushing result of the business application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings required to be used in the embodiments will be briefly described below, and it should be understood that the following drawings only illustrate some embodiments of the present invention, and therefore should not be considered as limiting the scope of the present invention. Like components are numbered similarly in the various figures.
Fig. 1 shows a flow diagram of a service information recommendation method provided in an embodiment of the present application;
fig. 2 is a schematic structural diagram illustrating a service information recommendation apparatus provided in an embodiment of the present application;
fig. 3 shows a schematic structural diagram of a computer device provided in an embodiment of the present application.
An icon: 200-service information recommendation means; 201-an acquisition module; 202-an input module; 203-a first determination module; 204-a second determination module; 205-a push module; 300-a computer device; 301-a transceiver; 302-a processor; 303-memory.
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.
The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
Hereinafter, the terms "including", "having", and their derivatives, which may be used in various embodiments of the present invention, are only intended to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing, and should not be construed as first excluding the existence of, or adding to, one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing.
Furthermore, the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of the present invention belong. The terms (such as those defined in commonly used dictionaries) should be interpreted as having a meaning that is consistent with their contextual meaning in the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein in various embodiments of the present invention.
Example 1
The embodiment of the application provides a service information recommendation method. Specifically, referring to fig. 1, the service information recommendation method includes:
step S101, obtaining the current life cycle data of the user and a plurality of material management target data of the current page.
Specifically, after the user enters the service application program, the service application program pushes information such as a service and a product to be promoted to the user, and the final purpose of pushing the information to the user by the service application program is to enable the user to accept the promoted service and product, namely to complete a transaction. To better achieve the above objective, business applications need to purposefully recommend products or services that may be of interest to users. In the prior art, a business application program generally infers user preferences based on the click behavior of a user, and pushes related products or businesses for the user based on the click behavior data of the user.
For example, the user a clicks a fund financing page in a certain business application program, and after receiving the information, the business application program binds the user a with products or services related to fund financing and generates a record. When the user A visits next time, the service application program recommends products or services related to fund financing for the user A. This recommendation is too comprehensive because the user's click behavior data does not accurately reflect the user's preferences.
In the embodiment, the life cycle stage of the user and the pushing purpose of the service application program are combined, so that information such as services or products is recommended for the user, the recommendation result is more in line with the requirements of the user, and the recommendation accuracy is improved. The life cycle stage refers to the process that a client has life-like birth, growth, maturation, aging and death for an enterprise. The detailed definitions are different for different industries. For example, in the telecommunications industry, the so-called lifecycle phase refers to the process of a telecommunications customer starting to become a customer of a telecommunications company and starting to generate business consumption, growth of consumption, stability of consumption, decline of consumption, and finally leaving the network.
Specifically, the products or services included in the service application program may be various, correspondingly, the life cycle stages of the user in different products or services are different, the dividing modes of the life cycle stages in different products or services may be different, the setting of the dividing modes may be performed according to actual needs, and the setting is not limited herein. Each user is in a different lifecycle stage in each product or service, and data of the lifecycle stage in which the user is in corresponding to each product or service may be defined as lifecycle data of the user. When the life cycle data of the user is counted, the user in the same life cycle stage under the same product or service can be defined as a user group, and the service application program can use the client group as a target group for pushing when pushing the product or service, that is, the same product or service can be pushed to all users of the same user group.
Different service recommendation targets can be formulated according to the life cycle stage of the user group in the product or service, and the service recommendation targets comprise specific service recommendation modes. For example, in the loan service, a service recommendation target for improving the user consumption can be formulated for a user group at the beginning, and the specific recommendation manner included in the service recommendation target can be to push a new gift package for the user group at the beginning, and the new gift package can be bound with the specific loan service, so as to attract the user group to know about and participate in the loan service. For the user group in the dormant period, a service recommendation target of the activated user can be formulated, and the specific recommendation mode included in the service recommendation target can be that old users are pushed to return to welfare and users in the dormant period are attracted to pay attention to the loan service again.
In different products or services, different service recommendation targets can be formulated aiming at user groups in the same life cycle stage. The specific setting may be according to practical situations, and is not limited herein. The service recommendation target may be defined as a material management target, and a plurality of different service recommendation targets may be defined as material management target data. And each service recommendation target is prestored in the database and is bound with each page of the service application program. And after the user enters the service application program, the service application program calls a service recommendation target bound to the current page browsed by the user, and the service recommendation target is stored in a queue to be displayed. And then the service application program can acquire the current life cycle data of the user and the service recommendation target bound with the current page. The current life cycle data of the user is a set including life cycle phase data of the user under the current product or service. And the service application program inputs a service recommendation target bound to a current page browsed by the user and a set of life cycle stage data of the user under a current product or service into a pre-trained recommendation model, outputs a service recommendation target which best meets the user through the recommendation model, and pushes specific content for the user according to the service recommendation target. Therefore, the recommendation effect and the recommendation accuracy of the business application program are improved.
According to the service recommendation method provided by the embodiment, when a user accesses a service application program, life cycle data and material operation target data of the user are obtained, the life cycle data and the material operation target data of the user are input into a pre-trained recommendation model, a matching score between the material operation target data and the life cycle data of the user can be obtained, and then the material operation target data is pushed for the user according to the matching score. The pushing result of the business application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Step S102, inputting the current life cycle data and a plurality of material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
specifically, a recommendation model is trained in advance, and after a user enters a service application program, the service application program calls the recommendation model to push related content for the user. The recommendation model is a probability model, that is, the recommendation model can output probability values between the life cycle stage of the user under each product or service and different service recommendation targets, and the probability values can be defined as matching scores. And the service application program pushes a corresponding service recommendation target for the user according to the probability value. Probability values between life cycle stages of users under various products or services and different service recommendation targets can be obtained in a training process of a recommendation model, wherein the training process of the recommendation model is as follows:
in this embodiment, the training process of the recommendation model may include:
acquiring service inferred data and service actual data at the moment T;
inputting the service inferred data and the service actual data at the T moment into an initial algorithm to obtain a recommendation model at the T moment;
and determining the target recommendation model according to the recommendation model at the time T.
Specifically, after entering the business application, the user may start browsing the interface, clicking the product, or filling in the data. Various types of user data generated by the user's operations are stored in a designated database. And the user data can be used as sample data for recommending model training after being cleaned and processed by big data. Because the data of the business application program is continuously updated, the training of the recommendation model can adopt the user data of the preset time period as a training sample.
Time T is the time at which user data is acquired. And the life cycle stage of the user in different products or services is continuously updated along with the lapse of time, so that after the user data at the time T is obtained, the life cycle stage data of the user in different products or services at the current time can be extracted from the user data through the big data platform. In addition, the service recommendation target set at the current time and corresponding to the life cycle stage of the user in different products or services can be extracted from the user data through the big data platform.
The life cycle stage data of the user in different products or services at the current moment and the service recommendation target set in the life cycle stage corresponding to the user in different products or services can be defined as service inference data. The business recommendation targets included in the business inference data are set by business personnel according to business experiences and in combination with life cycle stages of the user in different products or businesses. During model training, the actual result data generated when the current business inference data is applied can be extracted from the user data through the big data platform, and the actual result data can be defined as the business actual data. For example, the current business inference data may be: in the loan service, an activated service recommendation target is pushed to a user group in a dormant period. The generated service actual data may be: each customer in the user group in the dormant period in the loan service can specifically judge the acceptance degree of the pushed activated service recommendation target from the exposure data, the click data and the conversion data of the activated service recommendation target.
In this embodiment, the service inferred data and the service actual data acquired at time T are input into the algorithm, and the recommendation model at time T can be trained. The algorithm used may be a LR (Logistic Regression) algorithm, which may be selected according to actual situations, and is not limited herein.
Specifically, based on the user groups in different life cycle stages in each product or service, the probability values between the life cycle stage in which the user is located and different service recommendation targets in each product or service can be obtained through the acceptance degree of each user in the user groups to the designated service recommendation target, that is, the exposure data of the service recommendation target, the click data and the conversion data of the user to the service recommendation target. And deploying the model at the time T into a service application program for use. And pushing a corresponding service recommendation target for the user according to the probability value.
In this embodiment, based on the continuous update of the user data, the recommendation model may also be continuously updated iteratively correspondingly, so as to improve the applicability of the recommendation model. Specifically, on the premise that training of the recommendation model is performed at the time T, a certain duration may be preset based on the time T, and training of the next recommendation model is performed to obtain a subsequent recommendation model of the recommendation model at the time T. The two models can be compared, and the better recommendation model of the two models can be deployed into a business application program for use. The better of the two above recommendation models may be defined as the target recommendation model.
Step S103, determining the maximum matching score from the matching scores;
specifically, after the user enters the service application program, the service application program calls the recommendation model, and the probability values between the life cycle stage of the user under each product or service and different service recommendation targets are output through the recommendation model. And the service application program determines the maximum probability value according to the probability value output by the recommendation model so as to carry out subsequent pushing work.
Step S104, determining the material operation target data to be displayed corresponding to the maximum matching score;
specifically, after the user enters the service application program, the service application program acquires the maximum probability value output by the recommendation model, and determines the service recommendation target corresponding to the maximum probability value, where the service recommendation target corresponding to the maximum probability value can be defined as the material operation target data to be displayed.
And step S105, displaying the material operation target data to be displayed on the current page.
Specifically, after the service application program determines the service recommendation target corresponding to the maximum probability value, the specific content of the service recommendation target corresponding to the maximum probability value is displayed on the currently browsed interface of the user.
In a specific embodiment, the step of determining the target recommendation model according to the recommendation model at time T includes:
acquiring service inferred data and service actual data at the moment T + 1;
inputting the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
Specifically, the training method of the recommendation model at the time T +1 may refer to the above specific description of the training method of the recommendation model at the time T, and is not described herein again to avoid repetition. It should be noted that the recommendation model at the time T +1 is the latter recommendation model of the recommendation model at the time T.
In a specific embodiment, the step of determining an optimal recommendation model from the recommendation model at time T and the recommendation model at time T +1 includes:
calling the recommendation model at the time T and the recommendation model at the time T + 1;
respectively calculating a first model score of the recommended model at the time T and a second model score of the recommended model at the time T + 1;
and determining the optimal recommendation model according to the first model score and the second model score.
Specifically, after the recommendation model at the time T +1 is obtained through training, the recommendation model at the time T and the recommendation model at the time T +1 need to be compared and evaluated. The specific evaluation mode may be: and calling the recommendation model at the time T and the recommendation model at the time T +1, respectively applying the two models to the same user group to obtain data such as total accuracy, AUC (AUC), ks (ks) value, promotion degree and the like which respectively correspond to each recommendation model, and calculating the model score of each recommendation model based on the data. For example, the score is calculated based on the preset rule on the data of the overall accuracy, AUC, ks value, lifting degree, etc. of the recommended model at time T to obtain the first model score. And calculating the score of the data such as the total accuracy, the AUC (AUC), the ks (ks) value, the promotion degree and the like of the recommended model at the T +1 moment based on the same preset rule to obtain a second model score. And comparing the magnitude relation of the first model score and the second model score, and defining the model with high score as the optimal recommendation model.
In a specific embodiment, the step of obtaining the business inference data and the business actual data includes:
acquiring original data of a user;
and extracting the service inferred data and the service actual data from the user original data.
Specifically, browsing data, click data, and filling data retained by the user in the service application system may be defined as user original data. After data cleaning is carried out on the user original data through the big data platform, sample data used for training the recommendation model can be obtained. The sample data includes service inferred data and service actual data, and for specific definition of the service inferred data and the service actual data, reference is made to the above detailed description, and for avoiding repetition, details are not repeated here.
According to the business information recommendation method provided by the embodiment, a recommendation model is trained in advance, when a user accesses a business application program, life cycle data and material operation target data of the user are obtained, the life cycle data and the material operation target data of the user are input into the recommendation model, a matching score between the material operation target data and the life cycle data of the user can be obtained, and then the material operation target data is pushed for the user according to the matching score. The pushing result of the business application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Example 2
In addition, the embodiment of the application provides a service information recommendation device.
Specifically, as shown in fig. 2, the service information recommendation apparatus 200 includes:
an obtaining module 201, configured to obtain current life cycle data of a user and multiple material operation target data of a current page;
an input module 202, configured to input the current life cycle data and the plurality of material operation target data into a target recommendation model, and calculate, through the target recommendation model, a matching score between each material operation target data and the current life cycle data;
a first determining module 203 for determining a maximum match score from a plurality of said match scores;
the second determining module 204 is configured to determine material operation target data to be displayed, which corresponds to the maximum matching score;
and the pushing module 205 is configured to display the material management target data to be displayed on the current page.
In one embodiment, the service information recommendation apparatus 200 further includes: the training module is used for acquiring service inferred data and service actual data at the time T;
inputting the service inferred data and the service actual data at the time T into an initial algorithm to obtain a recommendation model at the time T;
and determining the target recommendation model according to the recommendation model at the time T.
In an embodiment, the training module is further configured to input the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
In an embodiment, the training module is further specifically configured to call the recommendation model at the time T and the recommendation model at the time T + 1;
respectively calculating a first model score of the recommendation model at the time T and a second model score of the recommendation model at the time T + 1;
and determining the optimal recommendation model according to the first model score and the second model score.
In an embodiment, the training module is further configured to obtain user raw data;
and extracting the service inference data and the service actual data from the user raw data.
The service information recommendation device 200 provided in this embodiment may perform the specific steps of the service information recommendation method provided in embodiment 1, and is not described herein again to prevent repetition.
The service information recommendation device provided in this embodiment is implemented by the service information recommendation device provided in this embodiment. The method comprises the steps of training a recommendation model in advance, obtaining life cycle data and material operation target data of a user when the user accesses a business application program, inputting the life cycle data and the material operation target data of the user into the recommendation model, obtaining a matching score between the material operation target data and the life cycle data of the user, and pushing the material operation target data for the user according to the matching score. The pushing result of the business application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Example 3
Furthermore, an embodiment of the present application provides a computer device 300, which includes a memory and a processor, where the memory stores a computer program, and the computer program executes the service information recommendation method provided in embodiment 1 when running on the processor.
Specifically, referring to fig. 3, the computer device 300 includes: a transceiver 301, a bus interface and a processor 302, the processor 302 configured to: acquiring current life cycle data of a user and a plurality of material operation target data of a current page;
inputting the current life cycle data and a plurality of material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
determining a maximum match score from a plurality of the match scores;
determining material operation target data to be displayed corresponding to the maximum matching score;
and displaying the material operation target data to be displayed on the current page.
In one embodiment, the processor 302 is further configured to: acquiring service inferred data and service actual data at the moment T;
inputting the service inferred data and the service actual data at the time T into an initial algorithm to obtain a recommendation model at the time T;
and determining the target recommendation model according to the recommendation model at the time T.
In one embodiment, the processor 302 is further configured to: acquiring service inferred data and service actual data at the moment T + 1;
inputting the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
In one embodiment, the processor 302 is further configured to: calling the recommendation model at the time T and the recommendation model at the time T + 1;
respectively calculating a first model score of the recommendation model at the time T and a second model score of the recommendation model at the time T + 1;
and determining the optimal recommendation model according to the first model score and the second model score.
In one embodiment, the processor 302 is further configured to obtain user raw data;
and extracting the service inference data and the service actual data from the user raw data.
In the embodiment of the present invention, the computer device 300 further includes: a memory 303. In FIG. 3, the bus architecture may include any number of interconnected buses and bridges, with one or more processors, represented in particular by processor 302, and various circuits of memory, represented by memory 303, linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 301 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over a transmission medium. The processor 302 is responsible for managing the bus architecture and general processing, and the memory 303 may store data used by the processor 302 in performing operations.
The computer device 300 provided in the embodiment of the present invention may perform the specific steps of the service information recommendation method in embodiment 1, and is not described herein again to avoid repetition.
The computer device provided by the embodiment trains a recommendation model in advance. When a user accesses a business application program, life cycle data and material operation target data of the user are obtained, the life cycle data and the material operation target data of the user are input into a recommendation model, matching scores between the material operation target data and the life cycle data of the user can be obtained, and then the material operation target data are pushed for the user according to the matching scores. The pushing result of the business application program can better meet the user requirement, and the accuracy of the recommendation result is improved.
Example 4
The present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the service information recommendation method provided in embodiment 1 is implemented.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative and, for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, each functional module or unit in each embodiment of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention or a part of the technical solution that contributes to the prior art in essence can be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a smart phone, a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of the changes or substitutions within the technical scope of the present invention, and shall cover the scope of the present invention.
Claims (10)
1. A service information recommendation method is characterized in that the method comprises the following steps:
acquiring current life cycle data of a user and a plurality of material operation target data of a current page;
inputting the current life cycle data and the material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
determining a maximum match score from a plurality of the match scores;
determining material operation target data to be displayed corresponding to the maximum matching score;
and displaying the material operation target data to be displayed on the current page.
2. The service information recommendation method according to claim 1, wherein the step of obtaining the target recommendation model comprises:
acquiring service inferred data and service actual data at the time T;
inputting the service inferred data and the service actual data at the time T into an initial algorithm to obtain a recommendation model at the time T;
and determining the target recommendation model according to the recommendation model at the time T.
3. The service information recommendation method according to claim 2, wherein the step of determining the target recommendation model according to the recommendation model at time T comprises:
acquiring service inferred data and service actual data at the moment T + 1;
inputting the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
4. The method according to claim 3, wherein the step of determining an optimal recommendation model from the recommendation model at time T and the recommendation model at time T +1 includes:
calling the recommendation model at the time T and the recommendation model at the time T + 1;
respectively calculating a first model score of the recommended model at the time T and a second model score of the recommended model at the time T + 1;
and determining the optimal recommendation model according to the first model score and the second model score.
5. The method of claim 2, wherein the step of obtaining the inferred service data and the actual service data comprises:
acquiring original data of a user;
and extracting the service inference data and the service actual data from the user raw data.
6. A service information recommendation apparatus, comprising:
the acquisition module is used for acquiring the current life cycle data of the user and a plurality of material operation target data of the current page;
the input module is used for inputting the current life cycle data and the material operation target data into a target recommendation model, and respectively calculating the matching score of each material operation target data and the current life cycle data through the target recommendation model;
a first determination module to determine a maximum match score from a plurality of the match scores;
the second determining module is used for determining the material operation target data to be displayed corresponding to the maximum matching score;
and the pushing module is used for displaying the material operation target data to be displayed on the current page.
7. The service information recommendation device according to claim 6, further comprising:
the training module is used for acquiring service inferred data and service actual data at the time T;
inputting the service inferred data and the service actual data at the time T into an initial algorithm to obtain a recommendation model at the time T;
and determining the target recommendation model according to the recommendation model at the time T.
8. The service information recommendation device according to claim 7, wherein the training module is further configured to obtain service inference data and service actual data at a time T + 1;
inputting the inferred service data and the actual service data at the time T +1 into the initial algorithm to obtain a recommendation model at the time T + 1;
and determining an optimal recommendation model from the recommendation model at the T +1 moment and the recommendation model at the T moment, and taking the optimal recommendation model as the target recommendation model.
9. A computer arrangement, characterized in that the computer arrangement comprises a memory and a processor, the memory storing a computer program which, when run by the processor, performs the service information recommendation method according to any one of claims 1-5.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the service information recommendation method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210751698.5A CN114996585A (en) | 2022-06-28 | 2022-06-28 | Service information recommendation method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210751698.5A CN114996585A (en) | 2022-06-28 | 2022-06-28 | Service information recommendation method, device, equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114996585A true CN114996585A (en) | 2022-09-02 |
Family
ID=83020397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210751698.5A Pending CN114996585A (en) | 2022-06-28 | 2022-06-28 | Service information recommendation method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114996585A (en) |
-
2022
- 2022-06-28 CN CN202210751698.5A patent/CN114996585A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190034963A1 (en) | Dynamic sentiment-based mapping of user journeys | |
| US20180150783A1 (en) | Method and system for predicting task completion of a time period based on task completion rates and data trend of prior time periods in view of attributes of tasks using machine learning models | |
| EP3617952A1 (en) | Information search method, apparatus and system | |
| CN103116582B (en) | A kind of information retrieval method and related system and device | |
| CN109189931B (en) | Target statement screening method and device | |
| US20200074509A1 (en) | Business data promotion method, device, terminal and computer-readable storage medium | |
| US20220067816A1 (en) | Method and system to detect abandonment behavior | |
| CN104657437B (en) | Promote the monitoring method and device of situation data | |
| CN110782291A (en) | Advertisement delivery user determination method and device, storage medium and electronic device | |
| JP2017535866A (en) | Offline evaluation of ranking functions | |
| CN111046156B (en) | Method, device and server for determining rewarding data | |
| CN117744961A (en) | Task allocation method and device | |
| CN108090785B (en) | Method and device for determining user behavior decline tendency and electronic equipment | |
| US20220335438A1 (en) | Retaining user value through user support interaction analytics | |
| CN114996585A (en) | Service information recommendation method, device, equipment and readable storage medium | |
| US11803464B2 (en) | System for automatic identification and selection of optimization metrics and accompanying models in experimentation platforms | |
| CN115330490A (en) | Product recommendation method and device, storage medium and equipment | |
| US20220198367A1 (en) | Expert matching through workload intelligence | |
| CN110210885A (en) | Excavate method, apparatus, equipment and the readable storage medium storing program for executing of potential customers | |
| CN105408928A (en) | Methods and systems for determining relevance of advertising categories for devices | |
| CN113159926A (en) | Loan transaction repayment date determination method and device | |
| CN115170167A (en) | Advertisement diagnosis method, device, computer equipment and storage medium | |
| CN112862527B (en) | User type determining method, device, equipment and storage medium | |
| CN111784091A (en) | Method and apparatus for processing information | |
| CN110727858A (en) | Recommendation method, computer storage medium and electronic device |
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 |