Product recommendation method and device based on conversion rate and LTV and electronic equipment

Abstract

The invention provides a product recommendation method and device based on a conversion rate and LTV and electronic equipment, and relates to the technical field of computers. The method comprises the steps of obtaining user feature data of a target user, product feature data of a plurality of to-be-recommended products, behavior data of the target user for each to-be-recommended product and LTV evaluation factor data of the target user for each to-be-recommended product; inputting the user feature data, the product feature data and the behavior data into a target conversion rate model to obtain a value conversion rate of the target user for each to-be-recommended product; inputting the user feature data, the product feature data and the LTV evaluation factor data into a target LTV model to obtain an LTV of each to-be-recommended product by a target user; obtaining recommendation value scores of the to-be-recommended products according to the value conversion rate and LTV of the target user for each to-be-recommended product; and recommending the plurality of to-be-recommended products to the target user according to the recommendation value scores of the plurality of to-be-recommended products. The product recommendation accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a product recommendation method, device and electronic device based on conversion rate and LTV. Background Art

[0002] In the related art, product recommendation methods have always faced a problem, that is, there is a contradiction between recommendation efficiency and product value. Improving recommendation efficiency may bring short-term benefits, but in the long run it will lead to lower product value; conversely, if the product value is increased, the recommendation efficiency will be lower, which will have many impacts on consumers and merchants. For example, in order to improve the recommendation efficiency, low-priced products will be recommended to users. Low-priced products have extremely high conversion efficiency due to their price advantage, and the algorithm model also scores them higher, so they have a high exposure rate. However, low-priced products also have many disadvantages, such as less profit for merchants, and low prices for consumers. To a large extent, the quality of the goods is poor, which causes a bad experience for consumers, resulting in low consumer repurchase rates and many dissatisfactions with the consumption process, etc.

[0003] It can be seen that the current product recommendation method cannot find a balance between the long-term value and short-term value of the goods, resulting in low accuracy of product recommendations, which affects the consumer experience and also affects the profits of merchants, causing many inconveniences to consumers and merchants. Summary of the invention

[0004] The present invention provides a product recommendation method, device and electronic device based on conversion rate and LTV, so as to solve the defect in the prior art that it is impossible to recommend products to users in combination with the long-term value and short-term value of the products, resulting in low accuracy of product recommendations, and achieve the purpose of improving the accuracy of product recommendations.

[0005] The present invention provides a product recommendation method based on conversion rate and LTV, comprising the following steps.

[0006] Obtain user characteristic data of the target user, product characteristic data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and user lifetime value (LTV) evaluation factor data of the target user for each of the products to be recommended; input the user characteristic data, product characteristic data and behavior data of each product to be recommended into the target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and input the user characteristic data, product characteristic data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtain the recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; recommend multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended.

[0007] According to a product recommendation method based on conversion rate and LTV provided by the present invention, a recommendation value score of a target product to be recommended is obtained according to the value conversion rate of any target product to be recommended among multiple products to be recommended and the LTV of the target product to be recommended, including: obtaining a first value score according to a preset first coefficient and the value conversion rate of any target product to be recommended among multiple products to be recommended; obtaining a second value score according to a preset second coefficient and the LTV of the target product to be recommended; obtaining the product of the first value score and the second value score, and determining the product as the recommendation value score.

[0008] According to a product recommendation method based on conversion rate and LTV provided by the present invention, before obtaining the user characteristic data of the target user, the product characteristic data of multiple products to be recommended, and the behavior data of the target user on each of the multiple products to be recommended, the method also includes: obtaining multiple sample data; wherein the sample data includes: user characteristic sample data of the sample user, product characteristic sample data of the sample product, behavior data of the sample user on the sample product, LTV evaluation factor data of the sample product and sample comprehensive value score of the sample product; training an initial conversion rate model according to the user characteristic sample data of the sample user, the product characteristic sample data of the sample product and the behavior data of the sample user on the sample product to obtain a target conversion rate model; and training an initial LTV model according to the user characteristic sample data of the sample user, the product characteristic sample data of the sample product and the LTV evaluation factor data of the sample product to obtain a target LTV model; obtaining a first coefficient and a second coefficient according to the first output result of the target conversion rate model, the second output result of the target LTV model and the sample comprehensive value score of the sample product.

[0009] According to a product recommendation method based on conversion rate and LTV provided by the present invention, LTV evaluation factor data includes at least one of the following: purchase frequency of products of the same category, number of related matching products purchased, and product evaluation score.

[0010] According to a product recommendation method based on conversion rate and LTV provided by the present invention, multiple products to be recommended are recommended to a target user according to the recommendation value score of each product to be recommended, including: sorting the multiple products to be recommended in descending order of the recommendation value score to obtain a sorting result; and displaying a first number of products to be recommended to the target user in sequence according to the order of the sorting results.

[0011] According to a product recommendation method based on conversion rate and LTV provided by the present invention, a first number of products to be recommended are displayed to target users in sequence according to the order of sorting results, including: obtaining a first number of products to be recommended according to the order of sorting results; determining the number of replacement products if the second number of products to be recommended in a first category among the first number of products to be recommended is greater than a set quantity threshold; and replacing the number of replacement products among the products to be recommended in the first category with products to be recommended in other categories.

[0012] The present invention also provides a product recommendation device based on conversion rate and LTV, comprising the following modules: The first acquisition module is used to acquire user feature data of the target user, product feature data of multiple products to be recommended, behavior data of the target user on each of the multiple products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user on each product to be recommended; The first processing module is used to input the user feature data, product feature data and behavior data of each product to be recommended into the target conversion rate model to obtain the value conversion rate of each product to be recommended by the target user output by the target conversion rate model; and input the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of each product to be recommended by the target user output by the target LTV model; The second acquisition module is used to acquire the recommendation value score of the target product to be recommended according to the value conversion rate of any target product to be recommended and the LTV of the target product to be recommended among the multiple products to be recommended; The recommendation module is used to recommend multiple products to be recommended to a target user according to the recommendation value score of each product to be recommended.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any of the above-mentioned product recommendation methods based on conversion rate and LTV is implemented.

[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the product recommendation method based on conversion rate and LTV as described above is implemented.

[0015] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-described product recommendation methods based on conversion rate and LTV.

[0016] The product recommendation method, device and electronic device based on conversion rate and LTV provided by the present invention obtain user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and LTV evaluation factor data of the user life cycle value of each product to be recommended by the target user; input the user feature data, product feature data and behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of each product to be recommended output by the target conversion rate model; and input the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtain the recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; and recommend multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended. The present invention can recommend products to target users by combining the short-term value of the value conversion rate of the recommended product with the long-term value of the target user's LTV of the recommended product, thereby effectively solving the problem in the prior art that it is impossible to recommend products to users in combination with the long-term value and short-term value of the product, resulting in low accuracy of product recommendations, thereby achieving the purpose of improving the accuracy of product recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1This is one of the flow charts of the product recommendation method based on conversion rate and LTV provided by the present invention.

[0019] Figure 2 This is the second flow chart of the product recommendation method based on conversion rate and LTV provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of a product recommendation device based on conversion rate and LTV provided by the present invention.

[0021] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Combine the following Figure 1-Figure 2 The product recommendation method based on conversion rate and LTV of the present invention is described.

[0024] Figure 1 This is one of the flow charts of the product recommendation method based on conversion rate and LTV provided by the present invention, such as Figure 1 As shown, the method includes the following S110~S140.

[0025] S110: Obtain user feature data of the target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user for each product to be recommended.

[0026] User characteristic data may be data that characterizes the characteristics of the target user, such as gender, age, geographic location, device and network used, etc.

[0027] Product feature data refers to data related to product features, such as product category, brand, attributes, descriptive words, and product images.

[0028] Behavioral data may include, for example, browsing records, click records, order records, purchase records, and feedback information, which are data related to the user's behavior towards the product.

[0029] User life time value (LTV) evaluation factor data may include, for example, the purchase frequency of products in the same category, the number of related products purchased, and product evaluation scores, which are data related to the long-term value that the purchase of the product can bring.

[0030] For example, the higher the target user's purchase frequency of the same category of a product, the higher the target user's replacement frequency of the product. From this, it can be inferred that the user's purchase of the product is more likely to bring long-term value (because the user will purchase it frequently), and the higher the LTV score. For example, the replacement frequency of mobile phones is usually higher than that of computers. In terms of the purchase frequency of the same category of products alone, the LTV score of mobile phones is higher than that of computers.

[0031] For example, after a target user purchases a certain product, the more related products he or she purchases, the more products he or she will purchase together with the product. Therefore, it can be inferred that the user's purchase of the product is more likely to bring long-term value (because the user will purchase more other related products), and the higher the LTV score. For example, after purchasing cleaning tools, users usually also purchase detergents, cleaning protective products, etc. In terms of the number of related products purchased, cleaning tools can be given a higher LTV score.

[0032] For example, the higher the product evaluation score of a product by target users, the more users may be attracted to purchase the product, thereby bringing long-term value, and the higher the LTV score.

[0033] Of course, it can be understood that the LTV evaluation factor data listed above are merely exemplary. The LTV evaluation factor data of the embodiment of the present invention include but are not limited to the purchase frequency of the same category products listed above, the number of related supporting products purchased, and product evaluation scores. For example, the LTV evaluation factor data may also include the total value of the purchase of related supporting products, and so on.

[0034] S120: Input the user characteristic data, product characteristic data and behavior data of each product to be recommended into the target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and input the user characteristic data, product characteristic data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model.

[0035] The value conversion rate (also called conversion rate) of each product to be recommended by the target user can be, for example, the probability that the target user will purchase the product to be recommended, which can represent the short-term value generated after the user purchases the product to be recommended, that is, the value of the product to be recommended that can be realized in the short term. The LTV of each product to be recommended by the target user can represent the long-term value generated after the target user purchases the product to be recommended.

[0036] The user characteristic data of each product to be recommended is the user characteristic data of the target user obtained above.

[0037] The target conversion rate model and the target LTV model can be obtained through pre-training. The specific acquisition method can be found in the corresponding introduction in the following text (S210~S230), and will not be described in detail here.

[0038] S130: Obtain a recommendation value score of a target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended.

[0039] In some embodiments, the value conversion rate of the target user for each product to be recommended and the LTV of the target user for the product to be recommended can be directly multiplied, and the result of the multiplication is determined as the recommendation value score.

[0040] In some embodiments, a first value score can be obtained based on a preset first coefficient and the value conversion rate of any target product to be recommended among multiple products to be recommended; a second value score can be obtained based on a preset second coefficient and the LTV of the target product to be recommended; the product of the first value score and the second value score is obtained, and the product is determined as the recommended value score.

[0041] It should be noted that, for the convenience of description, in the previous and following texts, the value conversion rate of the target user for the target product to be recommended is described as: the value conversion rate of the target product to be recommended; the LTV of the target user for the target product to be recommended is described as: the LTV of the target product to be recommended.

[0042] Specifically, the product of the first coefficient and the value conversion rate of the target product to be recommended can be obtained, and the product is determined as the first value score. The product of the second coefficient and the LTV of the target product to be recommended can be obtained, and the product is determined as the second value score, and then the product of the first value score and the second value score is obtained, and the product is determined as the recommended value score. In a specific implementation, the recommended value score of the target product to be recommended can be calculated by the following formula: score = (A*prob^ɑ) * (B*ltv^β); wherein score is the recommended value score of the target product to be recommended, A is the first coefficient, prob^ɑ is the value conversion rate of the target product to be recommended, B is the second coefficient, ltv^β is the LTV of the target product to be recommended, and "*" represents the operation "multiplication".

[0043] The method for obtaining the first coefficient and the second coefficient can be found in the corresponding description below, and will not be described in detail here.

[0044] S140: Recommending multiple products to be recommended to a target user according to the recommendation value score of each product to be recommended.

[0045] The multiple products to be recommended may be sorted in descending order of recommendation value scores to obtain a sorting result; and then the first number of products to be recommended may be displayed to the target user in sequence according to the order of the sorting results.

[0046] For example, the multiple products to be recommended are A1, A2, A3, A4, A5 and A6. If the multiple products to be recommended are sorted in descending order of recommendation value scores, the sorting results are A1, A2, A3, A4, A5 and A6. According to the sorting results, when the first number is 3, A1, A2 and A3 can be displayed to the user first, and then A4, A5 and A6 can be displayed to the user.

[0047] In some embodiments, the size of the image displayed for the recommended product can also be determined based on the recommendation value score. The higher the recommendation value score, the larger the displayed image is to attract the attention of the target user; the lower the recommendation value score, the smaller the displayed image is.

[0048] In some embodiments, in order to prevent there from being too many products of the same type in the products to be recommended displayed each time, when displaying a first number of products to be recommended to the target user in sequence according to the order of the sorting results, the first number of products to be recommended can be obtained first according to the order of the sorting results; then, among the first number of products to be recommended, it is determined whether the second number of products to be recommended in the first category is greater than a set quantity threshold; if the second number of products to be recommended in the first category is greater than the set quantity threshold, the number of replacement products is determined, and the replaced number of products to be recommended in the first category of products to be recommended are replaced with products to be recommended in other categories. For example, the replaced number of products to be recommended in the first category of products to be recommended can be replaced with products to be recommended in a second category different from the first category, or the replaced number of products to be recommended in the first category of products to be recommended can be replaced with products to be recommended in multiple different categories different from the first category.

[0049] In some embodiments, before executing S110, Figure 2 As shown, the embodiment of the present invention can also execute the following S210~S230.

[0050] S210: Obtain multiple sample data; wherein the sample data includes: user feature sample data of sample users, product feature sample data of sample products, behavior data of sample users on sample products, LTV evaluation factor data of sample products and sample comprehensive value scores of sample products.

[0051] The sample comprehensive value score of the sample product may be a comprehensive value score that is set in advance based on the comprehensive value of the sample product in actual conditions.

[0052] S220: Based on the user characteristic sample data of sample users, the product characteristic sample data of sample products, and the behavior data of sample users on sample products, an initial conversion rate model is trained to obtain a target conversion rate model; and, based on the user characteristic sample data of sample users, the product characteristic sample data of sample products, and the LTV evaluation factor data of sample products, an initial LTV model is trained to obtain a target LTV model.

[0053] The user feature sample data of the sample users, the product feature sample data of the sample products, and the behavior data of the sample users on the sample products may be input into the initial conversion rate model for training to obtain a target conversion rate model.

[0054] User feature sample data of sample users, product feature sample data of sample products, and LTV evaluation factor data of sample products may be input into an initial LTV model for training to obtain a target LTV model.

[0055] S230: Obtain a first coefficient and a second coefficient according to the first output result of the target conversion rate model, the second output result of the target LTV model, and the sample comprehensive value score of the sample product.

[0056] The sample data corresponding to the first sample product is input into the target conversion probability model, and the target conversion probability model outputs a first output result. The sample data corresponding to the first sample product is input into the target LTV model, and the target LTV model outputs a second output result. The first sample product is a sample product in any sample data among the multiple sample data.

[0057] Based on the model training method, the first coefficient and the second coefficient are obtained so that the first coefficient and the second coefficient satisfy: the product of the first coefficient and the first output result is the first product, the product of the second coefficient and the second output result is the second product, and the result obtained by multiplying the first product by the second product is the sample comprehensive value score of the first sample product.

[0058] The product recommendation method based on conversion rate and LTV provided by the present invention obtains user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and LTV evaluation factor data of the user life cycle value of the target user for each of the products to be recommended; inputs the user feature data, product feature data and behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and inputs the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtains a recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; and recommends multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended. The present invention can recommend products to target users by combining the short-term value of the target user's value conversion rate for the recommended products with the long-term value of the target user's LTV for the recommended products, effectively solving the problem in the prior art that it is impossible to recommend products to users in combination with the long-term value and short-term value of the products, resulting in low accuracy of product recommendations, thereby achieving the purpose of improving the accuracy of product recommendations.

[0059] The present invention takes into account both product conversion rate and user life cycle value in the product recommendation process, which can bring maximum benefits to enterprises and better long-term experience to users. The present invention can more accurately identify user groups that have high value for recommended products, and recommend products that are most likely to bring long-term benefits to enterprises to this user group, thereby achieving accurate product recommendations.

[0060] The product recommendation device based on conversion rate and LTV provided by the present invention is described below. The product recommendation device based on conversion rate and LTV described below and the product recommendation method based on conversion rate and LTV described above can correspond to each other.

[0061] Figure 3 Schematic diagram of the structure of the product recommendation device based on conversion rate and LTV provided by the present invention. Figure 3 As shown, the product recommendation device 300 based on conversion rate and LTV includes: a first acquisition module 301 , a first processing module 302 , a second acquisition module 303 , and a recommendation module 304 .

[0062] The first acquisition module 301 is used to obtain user feature data of the target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user for each product to be recommended.

[0063] The first processing module 302 is used to input the user characteristic data, product characteristic data and behavior data of each product to be recommended into the target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and input the user characteristic data, product characteristic data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model.

[0064] The second acquisition module 303 is used to acquire the recommendation value score of the target product to be recommended according to the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended.

[0065] The recommendation module 304 is used to recommend multiple products to be recommended to the target user according to the recommendation value score of each product to be recommended.

[0066] In some embodiments, the second acquisition module 303 is specifically used to: obtain a first value score based on a preset first coefficient and the value conversion rate of any target product to be recommended among the multiple products to be recommended; obtain a second value score based on a preset second coefficient and the LTV of the target product to be recommended; obtain the product of the first value score and the second value score, and determine the product as the recommended value score.

[0067] In some embodiments, the product recommendation device based on conversion rate and LTV further includes: The third acquisition module ( Figure 3), used for obtaining a plurality of sample data before the first obtaining module obtains the user characteristic data of the target user, the product characteristic data of a plurality of products to be recommended, and the behavior data of the target user on each of the plurality of products to be recommended; wherein the sample data includes: the user characteristic sample data of the sample user, the product characteristic sample data of the sample product, the behavior data of the sample user on the sample product, the LTV evaluation factor data of the sample product, and the sample comprehensive value score of the sample product; The second processing module ( Figure 3 ), for training an initial conversion rate model according to the user feature sample data of the sample users, the product feature sample data of the sample products, and the behavior data of the sample users on the sample products to obtain the target conversion rate model; and for training an initial LTV model according to the user feature sample data of the sample users, the product feature sample data of the sample products, and the LTV evaluation factor data of the sample products to obtain the target LTV model; The third processing module ( Figure 3 ), which is used to obtain the first coefficient and the second coefficient according to the first output result of the target conversion rate model, the second output result of the target LTV model and the sample comprehensive value score of the sample product.

[0068] In some embodiments, the LTV evaluation factor data includes at least one of the following: purchase frequency of products in the same category, number of related matching products purchased, and product evaluation score.

[0069] In some embodiments, the recommendation module 304 includes: Sorting unit ( Figure 3 ), used to sort the multiple products to be recommended in descending order of the recommendation value scores to obtain a sorting result; Recommended Units ( Figure 3 ), which is used to display a first number of the products to be recommended to the target user in sequence according to the order of the sorting results.

[0070] In some embodiments, the recommendation unit is specifically configured to: Acquire the first quantity of the products to be recommended according to the order of the sorting results; If the second quantity of the to-be-recommended products of the first category is greater than a set quantity threshold among the to-be-recommended products of the first quantity, determining the quantity of replacement products; Replace the products to be recommended of the first category with the products to be recommended of the other categories.

[0071] The product recommendation device based on conversion rate and LTV provided by the present invention obtains user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and LTV evaluation factor data of the user life cycle value of the target user for each of the products to be recommended; inputs the user feature data, product feature data and behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and inputs the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtains a recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; and recommends multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended. The present invention can recommend products to target users by combining the short-term value of the target user's value conversion rate for the recommended products with the long-term value of the target user's LTV for the recommended products, effectively solving the problem in the prior art that it is impossible to recommend products to users in combination with the long-term value and short-term value of the products, resulting in low accuracy of product recommendations, thereby achieving the purpose of improving the accuracy of product recommendations.

[0072] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor (processor) 810, a communication interface (Communications Interface) 820, a memory (memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call the logic instructions in the memory 830 to execute a product recommendation method based on conversion rate and LTV, which includes: obtaining user feature data of the target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and user life cycle value LTV evaluation factor data of the target user for each product to be recommended; inputting the user feature data, product feature data and behavior data of each product to be recommended into the target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and, inputting the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtaining a recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; and recommending multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended.

[0073] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0074] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the product recommendation method based on conversion rate and LTV provided by the above-mentioned methods, which method includes: obtaining user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user for each of the multiple products to be recommended, and user life cycle value LTV evaluation factor data of the target user for each product to be recommended; inputting the user feature data, product feature data and behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and inputting the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into the target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtaining a recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; and recommending multiple products to be recommended to the target user based on the recommendation value score of each product to be recommended among the multiple products to be recommended.

[0075] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the product recommendation method based on conversion rate and LTV provided by the above-mentioned methods, the method comprising: obtaining user feature data of a target user, product feature data of a plurality of products to be recommended, behavior data of the target user for each of the plurality of products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user for each of the products to be recommended; inputting the user feature data, product feature data and behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of the target user for each product to be recommended output by the target conversion rate model; and inputting the user feature data, product feature data and LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of the target user for each product to be recommended output by the target LTV model; obtaining a recommendation value score of the target product to be recommended based on the value conversion rate of any target product to be recommended among the plurality of products to be recommended and the LTV of the target product to be recommended; and recommending a plurality of products to be recommended to the target user based on the recommendation value score of each of the plurality of products to be recommended.

[0076] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0077] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A product recommendation method based on conversion rate and LTV, characterized in that: include: Acquire user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user on each of the multiple products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user on each of the products to be recommended; Input the user feature data, the product feature data and the behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of each product to be recommended by the target user output by the target conversion rate model; and input the user feature data, the product feature data and the LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of each product to be recommended by the target user output by the target LTV model; Obtaining a recommendation value score of the target product to be recommended according to the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; The multiple products to be recommended are recommended to the target user according to the recommendation value score of each product to be recommended.

2. The product recommendation method based on conversion rate and LTV according to claim 1, characterized in that: The step of obtaining a recommendation value score of the target product to be recommended according to the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended includes: Obtaining a first value score according to a preset first coefficient and the value conversion rate of any target product to be recommended among the multiple products to be recommended; Obtaining a second value score according to a preset second coefficient and the LTV of the target product to be recommended; A product of the first value score and the second value score is obtained, and the product is determined as the recommended value score.

3. The product recommendation method based on conversion rate and LTV according to claim 2, characterized in that: Before acquiring the user characteristic data of the target user, the product characteristic data of the plurality of products to be recommended, and the behavior data of the target user on each of the plurality of products to be recommended, the method further includes: Acquire multiple sample data; wherein the sample data includes: user feature sample data of sample users, product feature sample data of sample products, behavior data of the sample users on the sample products, LTV evaluation factor data of the sample products, and sample comprehensive value scores of the sample products; Training the initial conversion rate model according to the user feature sample data of the sample users, the product feature sample data of the sample products, and the behavior data of the sample users on the sample products to obtain the target conversion rate model; and Training the initial LTV model according to the user feature sample data of the sample users, the product feature sample data of the sample products, and the LTV evaluation factor data of the sample products to obtain the target LTV model; The first coefficient and the second coefficient are obtained according to the first output result of the target conversion rate model, the second output result of the target LTV model and the sample comprehensive value score of the sample product.

4. The product recommendation method based on conversion rate and LTV according to any one of claims 1 to 3, characterized in that: The LTV evaluation factor data includes at least one of the following: purchase frequency of products of the same category, number of related matching products purchased, and product evaluation score.

5. The product recommendation method based on conversion rate and LTV according to claim 1, characterized in that: The step of recommending the multiple products to be recommended to the target user according to the recommendation value score of each product to be recommended, comprises: Sorting the multiple products to be recommended in descending order of the recommendation value scores to obtain a sorting result; According to the order of the sorting results, a first number of the products to be recommended are displayed to the target user in sequence.

6. The product recommendation method based on conversion rate and LTV according to claim 5, characterized in that: The step of sequentially displaying a first number of the products to be recommended to the target user according to the order of the sorting results includes: According to the order of the sorting results, obtaining the first quantity of the products to be recommended; If the second quantity of the to-be-recommended products of the first category is greater than a set quantity threshold among the to-be-recommended products of the first quantity, determining the quantity of replacement products; Replace the products to be recommended of the first category with the products to be recommended of the other categories.

7. A product recommendation device based on conversion rate and LTV, characterized in that: include: The first acquisition module is used to acquire user feature data of a target user, product feature data of multiple products to be recommended, behavior data of the target user on each of the multiple products to be recommended, and user life cycle value (LTV) evaluation factor data of the target user on each of the products to be recommended; A first processing module is used to input the user characteristic data, the product characteristic data and the behavior data of each product to be recommended into a target conversion rate model to obtain the value conversion rate of each product to be recommended by the target user output by the target conversion rate model; and input the user characteristic data, the product characteristic data and the LTV evaluation factor data of each product to be recommended into a target LTV model to obtain the LTV of each product to be recommended by the target user output by the target LTV model; A second acquisition module is used to acquire a recommendation value score of the target product to be recommended according to the value conversion rate of any target product to be recommended among the multiple products to be recommended and the LTV of the target product to be recommended; The recommendation module is used to recommend the multiple products to be recommended to the target user according to the recommendation value score of each product to be recommended.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the product recommendation method based on conversion rate and LTV according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the product recommendation method based on conversion rate and LTV according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the product recommendation method based on conversion rate and LTV according to any one of claims 1 to 6 is implemented.

Images