CN111967998A - Product recommendation processing method and device - Google Patents
Product recommendation processing method and device Download PDFInfo
- Publication number
- CN111967998A CN111967998A CN202010836091.8A CN202010836091A CN111967998A CN 111967998 A CN111967998 A CN 111967998A CN 202010836091 A CN202010836091 A CN 202010836091A CN 111967998 A CN111967998 A CN 111967998A
- Authority
- CN
- China
- Prior art keywords
- product
- recommendation
- result
- recommendation result
- white list
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Technology Law (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a recommendation processing method and device for products, wherein the method comprises the following steps: receiving a product strategy recommendation result aiming at least one user and input by a service person; obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation; inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples; obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user. The invention can improve the accuracy and the success rate of product recommendation.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a product recommendation processing method and device.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
With the development of information technology and banking, competition between financial institutions such as banks is becoming intense. Meanwhile, under the condition that internet finance and big data analysis are generally applied, big data service value is exerted, accurate and personalized product recommendation of private customers is achieved, the accuracy level of individual customer service is comprehensively improved, the accuracy degree and the intelligent level of individual customer service can be enhanced, and the market competitiveness of financial institutions such as banks can be further improved.
At present, the recommendation of bank products is mainly based on active inquiry of a client, and occasional product recommendation is also scattered recommendation of familiar clients by a client counter or business personnel, so that the accuracy and the success rate of the recommendation of bank products are low.
Disclosure of Invention
The embodiment of the invention provides a recommendation processing method of a product, which is used for improving the precision and the success rate of product recommendation and comprises the following steps:
receiving a product strategy recommendation result aiming at least one user and input by a service person;
obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation;
inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
The embodiment of the invention also provides a product recommendation processing device, which is used for improving the product recommendation accuracy and success rate, and comprises:
the system comprises a receiving unit, a recommending unit and a recommending unit, wherein the receiving unit is used for receiving a product strategy recommending result aiming at least one user and input by service personnel;
the white list recommendation unit is used for obtaining a product white list recommendation result of at least one user according to the corresponding relation between a preset white list and a product; the preset white list is used for storing users meeting the key factors of product recommendation;
the cluster analysis recommending unit is used for inputting the plurality of product recommending factors into a product recommending model generated by pre-training to obtain a product cluster analysis recommending result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
the fusion processing unit is used for obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the recommendation processing method of the product when executing the computer program.
An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program for executing the recommendation processing method of the above product is stored.
The recommendation processing scheme of the product provided by the embodiment of the invention comprises the following steps: receiving a product strategy recommendation result aiming at least one user and input by a service person; obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation; inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples; obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; the fused recommendation result is sent to the corresponding user, and the product recommendation accuracy and success rate are improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a schematic flow chart illustrating a recommendation processing method for a product according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a product recommendation process in an embodiment of the invention;
FIG. 3 is a flowchart illustrating a product recommendation processing method according to another embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a recommendation processing apparatus for a product according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a product recommendation processing apparatus according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
Before describing embodiments of the present invention, terms related to implementation of the present invention will be described.
Clustering analysis: refers to an analytical process that groups a collection of physical or abstract objects into classes that are composed of similar objects. It is an important human behavior.
The inventor finds that: when a financial institution such as a bank recommends products (insurance products, financial products and the like), no system which can be relied on is used as a hand grip, and particularly blanket recommendation coverage of a whole number of customers and recommendation of a specific customer group cannot be realized.
As the inventor finds the technical problem, a recommended treatment scheme of the product is provided. The scheme includes four methods of recommending insurance. The implementation method adopts modes of configuration parameter management, intelligent model recommendation and the like, and has good complementarity, safety, flexibility and directivity. Specifically, the scheme recommends a single insurance product most suitable for the user to the individual client through methods of marketing strategy recommendation of banking staff, white list insurance product recommendation, cluster analysis algorithm model recommendation and real-time interactive recommendation. The recommendation mechanism gives consideration to the demands of banks, customers and products, brings benefits for the banks to sell insurance products, facilitates the customers to purchase insurance products, enlarges the sales of the insurance products, improves the accuracy of the customers to purchase the insurance products, and provides effective support for the live customers and the reserved customers of the banks.
The embodiment of the invention realizes the method of parameter configuration, rating list marketing, real-time interactive recommendation and omnibearing index data processing, and effectively improves the accuracy and the success rate of bank recommended products (such as insurance products). Specifically, the implementation of the recommendation of the insurance product according to the embodiment of the present invention is divided into the following aspects:
1. the method comprises the following steps of recommending marketing strategies of banking staff, and supporting the business staff to select recommended products from an insurance product pool for recommendation;
2. recommending the insurance products in the white list, and supporting the oriented recommendation of the insurance products in a mode of circling the white list;
3. recommending a clustering analysis algorithm model, finding a model with high accuracy, success rate and recall rate through a training model algorithm, and recommending products according to the model;
4. real-time interactive recommendation, which is to recommend products in real time according to client information, purchase will and the like provided by a client on line or face to face;
5. the four recommendation modes are comprehensively applied, the strategy recommendation is laid down, the white list recommendation and the model recommendation are accurately replaced, and the real-time interactive recommendation has more flexibility, accuracy and timeliness;
6. the recommended channel can be recommended through online recommendation or temporary cabinet recommendation, the online recommendation is for example recommended on a mobile banking, recommended by sending a short message, recommended by scanning a two-dimensional code and the like, and the temporary cabinet recommendation is mainly recommended when a customer comes to a website;
7. tracking, counting and comparing the conditions of the recommendation times, the click rate and the volume of bargain in the later stage of recommendation, optimizing the recommendation method and flexibly improving the recommendation method.
The recommended treatment of the product is described in detail below.
Fig. 1 is a schematic flow chart of a recommendation processing method for a product in an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
step 101: receiving a product strategy recommendation result aiming at least one user and input by a service person;
step 103: obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation;
step 105: inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
step 107: obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
The product recommendation processing method provided by the embodiment of the invention improves the precision and the success rate of product recommendation.
In specific implementation, the product recommendation processing method provided by the embodiment of the invention can be applied to recommendation processing of bank insurance products or financial products and the like.
The recommendation processing method for products provided by the embodiment of the invention can be applied to the recommendation of bank insurance products, and the scheme is a comprehensive, targeted, multi-directional, real-time and thousands-of-people insurance product recommendation method, which can effectively improve the accuracy and success rate of the recommendation of bank products, as shown in fig. 2, and the specific steps can include:
s1: and manually recommending the marketing strategy, and selecting key products from the insurance product pool by business personnel for recommendation.
S2: and recommending the products in the white list, and performing oriented recommendation on the insurance products by circling the white list.
S3: and recommending a clustering analysis algorithm model, wherein the recommendation is carried out by training the model algorithm model, and the recommendation is the most challenging recommendation with the best effect.
S4: and (3) real-time interactive recommendation, namely performing real-time product recommendation according to customer information, purchase wishes and the like provided by the customers on line or face to face. Such recommendations are more confident and more consistent with the expectations of the customer.
S5: and integrating the four recommendation modes through various online or offline channels to carry out omnibearing recommendation on the customers.
S6: and tracking and monitoring the conditions of the recommendation times, click rate and volume of transaction in the later stage of recommendation, optimizing the recommendation method and improving the recommendation method.
For convenience of understanding, the following describes in detail each step of the recommendation processing method of the product according to the embodiment of the present invention, taking the recommendation of the insurance product as an example, with reference to fig. 2.
First, the step 101, i.e. S1 in fig. 2, is introduced, where the step obtains a product policy recommendation result, and specifically includes:
s11: and determining business personnel for recommending the insurance products.
S12: the products that can be recommended are added to the product pool (the product pool can be a product list, a product list) so that the business personnel operating the product pool can see the products.
S13: the banking personnel select the insurance which can be generally recommended from the product pool according to various factors such as the condition of insurance products in the row, policy direction, personal experience and the like, and carry out bottom laying recommendation on the products.
S14: for products that are recommended online or offline, such recommendations are suitable for a wider range of customers, whether or not in-line, who may purchase the product.
Next, step 103, i.e. S2 in fig. 2, is introduced, where the step obtains a product white list recommendation result, and specifically includes:
s21: according to the product applicability, the client has bought factors such as insurance, client asset condition, regional condition, epidemic condition and the like, and the client in a specific area (for example, the product in Beijing area can only select the client in Beijing to recommend the product, but not select the client in Shanghai) is selected to recommend the product. In particular implementation, the product recommendation factor may include: the customer has purchased such factors as insurance, customer property status, regional status, epidemics, etc.
S22: the white list (the list of customers meeting the product recommendation factor, such as the list of people with assets exceeding millions in the row and meeting a certain condition) recommendation also takes timeliness into consideration, and the system needs to automatically screen according to the set condition, which is similar to a specific marketing activity.
S23: and performing online and offline recommendation on the recommendation result of the generated white list and product corresponding relation.
S24: the white list recommendation belongs to specific recommendations, the white list recommendation is more directional, the client range is smaller, and the recommended products can be prioritized over the result of manual recommendation of marketing strategies in the aspect of ranking.
In specific implementation, the generated white list and product corresponding relation can be used for matching and searching in the whole number of customers to obtain a product white list recommendation result.
Thirdly, next, the step 105, i.e. S3 in fig. 2, is introduced, where the step obtains a recommendation result of the product cluster analysis, and specifically may include:
s31: the product recommendation model, for example, the cluster analysis algorithm model recommendation is a recommendation of a higher-level artificial intelligence algorithm, the cluster analysis algorithm model can be obtained by pre-training according to historical data, and the specific process of obtaining the model can include: dividing a plurality of product recommendation factor samples into a training set and a testing set; pre-training a neural network model by utilizing a training set; testing the pre-trained neural network model by using a test set to obtain a final clustering analysis algorithm model, wherein the input of the model is each product recommendation factor, and the output of the model is a clustering analysis recommendation result obtained according to each product recommendation factor, for example, Zhang III is suitable for insurance products: flight delay insurance products and accident insurance products, li si suitable insurance products are heavy insurance products, and so on.
S32: the recommended conditions need to consider the situation of the purchase history of past insurance products, and model algorithm selection and model training are carried out according to a plurality of factors (product recommendation factors) such as the marketable region, the period, the insurance type, the marketable channel, the age of the client, the region to which the client belongs, the asset situation of the client, the historical product purchase situation of the client and the like. And finally finding a more suitable product recommendation model according to the historical data.
S33: according to a product recommendation model (cluster analysis algorithm model), the system generates a specific number of recommendation results suitable for thousands of people of a specific person every day, and after other recommendation results are fused in the subsequent step 107, different recommendation results are recommended to different customers, for example, flight delay insurance products and accident insurance products are recommended to Zhang III through WeChat or mobile phone banking, and serious danger products are recommended to Li IV through WeChat or mobile phone banking and the like.
S34: the recommendation of the clustering analysis algorithm model belongs to specific recommendations, the clustering analysis algorithm model has directivity and small customer range, and for the recommended product sorting, the products can be prior to the result recommended by a white list.
Fourthly, next, a further preferred step 106 after the step 105 is introduced, that is, S4 in fig. 2, where the step obtains a recommendation result of the product cluster analysis, and the current interactive recommendation result may specifically include:
s41: the interactive recommendation is more real-time. Generally, the two modes of the online mode and the offline mode are adopted. The client is enabled to gradually complete the filling of similar questionnaire surveys according to system settings through a mobile phone bank, for example, on line, so as to obtain the result information of the similar questionnaire surveys on line; and filling out the questionnaire offline or performing face-to-face communication to obtain offline questionnaire or communication information, and selecting a proper product. In specific implementation, the current output information of the user may include the result information of the on-line questionnaire-like survey, the off-line questionnaire, or the communication information.
S42: the recommendation can be carried out at any time, the autonomy selected by a customer is high, the convenience of purchasing insurance products by the user is improved, and the accuracy and the success rate of product recommendation are further improved.
S43: such recommendations not only fill in personal information, but also support product selection for parents and children.
S44: the interactive recommendation is slightly different from the three recommendation modes (the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result) and can be independently classified into a recommendation type.
As can be seen from the above, in an embodiment, as shown in fig. 3, the method for recommending a product may further include step 106: receiving a current interactive recommendation result obtained according to current output information of a user;
obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result, wherein the fusion recommendation result comprises the following steps: and obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result, the product clustering analysis recommendation result and the current interactive recommendation result.
When the method is specifically implemented, the implementation scheme of interactive recommendation by considering the user output information further improves the precision and the success rate of product recommendation.
Next, the step 107, i.e. S5 in fig. 2, is introduced, which may specifically include:
s51: and integrating the previous 4 recommendation modes (a product strategy recommendation result, a product white list recommendation result, a product clustering analysis recommendation result and a current interactive recommendation result). The first 3 types (the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result) can set the priority order to perform the ranking recommendation of the products.
S52: the set order and the recommended number of products support parameter setting (for example, 3 or 5 products are recommended at most, the number can be flexibly set when each product is the 1 st display bit or the 2 nd display bit, and the like).
S53: according to the sales channels supported by each type of product, multiple channels can be selected for displaying recommended products, the multiple channels reach customers quickly, and the recommendation timeliness is improved.
As can be seen from the above, in an embodiment, obtaining a fused recommendation result of at least one user according to the product policy recommendation result, the product white list recommendation result, and the product clustering analysis recommendation result may include:
performing priority ranking on the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result;
and obtaining a fusion recommendation result of at least one user according to the priority ranking result.
During specific implementation, the implementation mode of setting the priority ranking for the recommendation results to obtain the final fusion recommendation results further improves the precision and the success rate of product recommendation.
In specific implementation, the fusion recommendation result may include: for example, Zhang III is a suitable insurance product: flight delay insurance products and accident insurance products, wherein the insurance products suitable for lie IV are heavy insurance products, child education insurance products, accident insurance products and the like, so that the fusion recommendation result of the flight delay insurance products and the accident insurance products can be sent to Zhang III in an online or offline mode, and the heavy insurance products, the child education insurance products and the accident insurance products can be sent to lie IV in an online or offline mode.
Sixthly, next, a further preferred optimization scheme, that is, S6 in fig. 2, is introduced, which may specifically include:
s61: and (4) carrying out system statistics and tracking on click rate and purchase rate of all recommendation results.
S62: according to the change of the reference factors (product recommendation factors), manually recommended products are selected for list adjustment, particularly, model training is carried out (the model is retrained according to the change of the reference factors so as to optimize a cluster analysis algorithm model), and the recommendation accuracy and success rate are improved.
As can be seen from the above, in an embodiment, the method for recommending a product may further include: optimizing the product recommendation model according to the recommendation rate, the click rate and the purchase rate corresponding to the fused recommendation result to obtain an optimized product recommendation model;
therefore, in the step 105, inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product cluster analysis recommendation result of at least one user, may include: and inputting the plurality of product recommendation factors into the optimized product recommendation model to obtain a product clustering analysis recommendation result of at least one user.
When the product recommendation method is specifically implemented, the implementation mode of optimizing the product recommendation model further improves the precision and the success rate of product recommendation.
In specific implementation, before implementing the invention, a bank should have conditions of product data of various surrogated insurance, experience of business personnel for recommending insurance, professional personnel for artificial intelligence model training, business data, technical implementation and the like. Such as the design of model reference factors, the way of multi-channel recommendations, etc.
In summary, the product recommendation processing method provided by the embodiment of the invention has the advantages that:
1. the system is relatively perfect, and the safety of insurance recommendation is improved: the method for realizing the parameters of the insurance recommendation comprehensively considers a plurality of dimensions and carries out all-around control on possible recommendations.
2. Possess good flexibility: through parameter control, the designer selects the diversity of insurance products that are desired to be recommended.
3. The maintenance cost is low: related information can be set through parameters, only parameters need to be adjusted if the system is required to be changed in the maintenance process, and version development is generally not required.
The embodiment of the invention also provides a product recommendation processing device, which is described in the following embodiment. Because the principle of solving the problems of the device is similar to the recommendation processing method of the product, the implementation of the device can refer to the implementation of the method, and repeated parts are not described again.
Fig. 4 is a schematic structural diagram of a recommendation processing apparatus for a product according to an embodiment of the present invention, and as shown in fig. 4, the apparatus includes:
the system comprises a receiving unit 01, a recommending unit and a recommending unit, wherein the receiving unit is used for receiving a product strategy recommending result aiming at least one user and input by business personnel;
the white list recommendation unit 03 is configured to obtain a product white list recommendation result of at least one user according to a preset corresponding relationship between a white list and a product; the preset white list is used for storing users meeting the key factors of product recommendation;
the cluster analysis recommending unit 05 is used for inputting the plurality of product recommending factors into a product recommending model generated by pre-training to obtain a product cluster analysis recommending result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
the fusion processing unit 07 is configured to obtain a fusion recommendation result of at least one user according to the product policy recommendation result, the product white list recommendation result, and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
In one embodiment, as shown in fig. 5, the recommendation processing device for a product may further include: the interactive recommendation unit 06 is configured to receive a current interactive recommendation result obtained according to the current output information of the user;
the fusion processing unit 07 may be specifically configured to: and obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result, the product clustering analysis recommendation result and the current interactive recommendation result.
In one embodiment, the recommendation processing apparatus for a product may further include: the optimization processing unit is used for optimizing the product recommendation model according to the recommendation rate, the click rate and the purchase rate corresponding to the fused recommendation result to obtain an optimized product recommendation model;
the cluster analysis recommending unit is specifically configured to: and inputting the plurality of product recommendation factors into the optimized product recommendation model to obtain a product clustering analysis recommendation result of at least one user.
In an embodiment, the fusion processing unit may be specifically configured to:
performing priority ranking on the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result;
and obtaining a fusion recommendation result of at least one user according to the priority ranking result.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the recommendation processing method of the product when executing the computer program.
An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program for executing the recommendation processing method of the above product is stored.
The recommendation processing scheme of the product provided by the embodiment of the invention comprises the following steps: receiving a product strategy recommendation result aiming at least one user and input by a service person; obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation; inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples; obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; the fused recommendation result is sent to the corresponding user, and the product recommendation accuracy and success rate can be improved.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method for recommendation processing of a product, comprising:
receiving a product strategy recommendation result aiming at least one user and input by a service person;
obtaining a product white list recommendation result of at least one user according to the corresponding relation between the preset white list and the product; the preset white list is used for storing users meeting the key factors of product recommendation;
inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
2. The recommendation processing method for a product according to claim 1, further comprising: receiving a current interactive recommendation result obtained according to current output information of a user;
obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result, wherein the fusion recommendation result comprises the following steps: and obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result, the product clustering analysis recommendation result and the current interactive recommendation result.
3. The recommendation processing method for a product according to claim 1, further comprising: optimizing the product recommendation model according to the recommendation rate, the click rate and the purchase rate corresponding to the fused recommendation result to obtain an optimized product recommendation model;
inputting a plurality of product recommendation factors into a product recommendation model generated by pre-training to obtain a product clustering analysis recommendation result of at least one user, wherein the method comprises the following steps: and inputting the plurality of product recommendation factors into the optimized product recommendation model to obtain a product clustering analysis recommendation result of at least one user.
4. The method of claim 1, wherein obtaining the fused recommendation result of the at least one user according to the product policy recommendation result, the product white list recommendation result, and the product cluster analysis recommendation result comprises:
performing priority ranking on the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result;
and obtaining a fusion recommendation result of at least one user according to the priority ranking result.
5. A recommendation processing apparatus for a product, comprising:
the system comprises a receiving unit, a recommending unit and a recommending unit, wherein the receiving unit is used for receiving a product strategy recommending result aiming at least one user and input by service personnel;
the white list recommendation unit is used for obtaining a product white list recommendation result of at least one user according to the corresponding relation between a preset white list and a product; the preset white list is used for storing users meeting the key factors of product recommendation;
the cluster analysis recommending unit is used for inputting the plurality of product recommending factors into a product recommending model generated by pre-training to obtain a product cluster analysis recommending result of at least one user; the product recommendation model is generated by pre-training according to a plurality of product recommendation factor samples;
the fusion processing unit is used for obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result; and sending the fusion recommendation result to the corresponding user.
6. The recommendation processing device for a product according to claim 5, further comprising: the interactive recommendation unit is used for receiving a current interactive recommendation result obtained according to the current output information of the user;
the fusion processing unit is specifically configured to: and obtaining a fusion recommendation result of at least one user according to the product strategy recommendation result, the product white list recommendation result, the product clustering analysis recommendation result and the current interactive recommendation result.
7. The recommendation processing device for a product according to claim 5, further comprising: the optimization processing unit is used for optimizing the product recommendation model according to the recommendation rate, the click rate and the purchase rate corresponding to the fused recommendation result to obtain an optimized product recommendation model;
the cluster analysis recommending unit is specifically configured to: and inputting the plurality of product recommendation factors into the optimized product recommendation model to obtain a product clustering analysis recommendation result of at least one user.
8. The product recommendation processing apparatus of claim 5, wherein the fusion processing unit is specifically configured to:
performing priority ranking on the product strategy recommendation result, the product white list recommendation result and the product clustering analysis recommendation result;
and obtaining a fusion recommendation result of at least one user according to the priority ranking result.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 4 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010836091.8A CN111967998A (en) | 2020-08-19 | 2020-08-19 | Product recommendation processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010836091.8A CN111967998A (en) | 2020-08-19 | 2020-08-19 | Product recommendation processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111967998A true CN111967998A (en) | 2020-11-20 |
Family
ID=73388605
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010836091.8A Pending CN111967998A (en) | 2020-08-19 | 2020-08-19 | Product recommendation processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111967998A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113706266A (en) * | 2021-09-07 | 2021-11-26 | 中国银行股份有限公司 | Business recommendation method and system based on bank invoice data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160162974A1 (en) * | 2014-12-08 | 2016-06-09 | Lg Cns Co., Ltd. | Personalized recommendation method and system, and computer-readable record medium |
| CN108520444A (en) * | 2018-04-12 | 2018-09-11 | 中国平安人寿保险股份有限公司 | Insurance products recommend method, unit and computer readable storage medium |
| CN110795621A (en) * | 2019-09-26 | 2020-02-14 | 北京你财富计算机科技有限公司 | Financial product recommendation information display method and device and electronic equipment |
-
2020
- 2020-08-19 CN CN202010836091.8A patent/CN111967998A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160162974A1 (en) * | 2014-12-08 | 2016-06-09 | Lg Cns Co., Ltd. | Personalized recommendation method and system, and computer-readable record medium |
| CN108520444A (en) * | 2018-04-12 | 2018-09-11 | 中国平安人寿保险股份有限公司 | Insurance products recommend method, unit and computer readable storage medium |
| CN110795621A (en) * | 2019-09-26 | 2020-02-14 | 北京你财富计算机科技有限公司 | Financial product recommendation information display method and device and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113706266A (en) * | 2021-09-07 | 2021-11-26 | 中国银行股份有限公司 | Business recommendation method and system based on bank invoice data |
| CN113706266B (en) * | 2021-09-07 | 2024-05-28 | 中国银行股份有限公司 | Business recommendation method and system based on bank invoice data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2023258377A1 (en) | Electronic platform for managing investment products | |
| US20020091605A1 (en) | Asset allocation optimizer | |
| US20150348192A1 (en) | Electronic platform for managing investment products | |
| US20020152151A1 (en) | Integrated investment portfolio management system and method | |
| CN108734297A (en) | The machine learning commending system of the performance optimization of electronic content items for network transmission | |
| Liagkouras et al. | Efficient portfolio construction with the use of multiobjective evolutionary algorithms: Best practices and performance metrics | |
| CN111861569A (en) | Product information recommendation method and device | |
| US20140258175A1 (en) | Generating Personalized Investment Recommendations | |
| US20210216936A1 (en) | Intelligent diversification tool | |
| CN107730386A (en) | Generation method, device, storage medium and the computer equipment of investment combination product | |
| KR20170005985A (en) | System and method for providing consulting service on demand for portfolio investment according to financial plan of financial investment instruments | |
| US20180330296A1 (en) | System and method for selecting research and development project through autonomous proposals of evaluation indicators | |
| US20150095253A1 (en) | Method and system for creating a list of organizations based on an individual's preferences and personal characteristics | |
| Njenga et al. | A theoretical review of mobile commerce success determinants | |
| WO2001061579A1 (en) | System and method for creating, distributing and managing artificial agents | |
| KR20120032606A (en) | Stock investment system enabling participattion of stock investment clients and method thereof | |
| CN111967998A (en) | Product recommendation processing method and device | |
| KR102176108B1 (en) | Differential fee payment system through professional experts | |
| US20140188559A1 (en) | Market research interface for use with prediction market | |
| CN112449002B (en) | Method, device and equipment for pushing object to be pushed and storage medium | |
| Salim | Exploring Customers' Perception on Adoption and Use of Electronic Banking Services in Tanzania Commercial Banks | |
| US20210272045A1 (en) | Automatically selecting sub-contractors and estimating cost for contracted tasks | |
| CN111861674B (en) | Product recommendation method and system | |
| CN110795232A (en) | Data processing method, data processing device, computer readable storage medium and computer equipment | |
| US20230267542A1 (en) | System and method for analysing financial products |
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 |