Commodity recommendation method and device based on recommendation crowd, electronic equipment and storage medium

By calculating the similarity scores and characteristics of users and commodities, and combining the characteristics of purchased and unpurned users, the recommended population is determined, which solves the problem of inaccurate recommendations in the prior art and improves the accuracy and quality of product recommendations.

CN119048197APending Publication Date: 2024-11-29DATAGRAND TECH INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411101428.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing product recommendation methods based on recommendation groups cannot accurately push products to users, resulting in low recommendation quality.

Method used

By calculating the similarity score based on the user's purchased product characteristics and target product characteristics, sorting the collection of unpurned users, and obtaining similar user collections based on the characteristics of purchased users, determining the final recommended population to improve the accuracy of recommendations.

Benefits of technology

It improves the accuracy and quality of product recommendations, and makes the characteristics of the recommended population closer to the target product, thereby improving the accuracy and user experience of recommendations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048197A_ABST
    Figure CN119048197A_ABST
Patent Text Reader

Abstract

The invention discloses a commodity recommendation method and device based on recommendation crowds and a storage medium. The method comprises the following steps: determining a similarity score of a first user according to purchased commodity characteristics of the first user and target commodity characteristics of a target commodity; sorting the first users according to the similarity scores, and determining a first user set according to a sorting result of the similarity scores; acquiring a third user similar to the second user according to the user characteristics, and determining a second user set according to the third user; determining a recommended crowd according to the first user set and the second user set; and recommending the target commodity according to the recommendation crowd. The final recommendation crowd is determined according to the combination of the first user set and the second user, the combination of the first user set and the second user is determined based on the two dimensions of the commodity characteristics and the user characteristics, and then the recommendation crowd is determined from the two dimensions of the commodity characteristics and the user characteristics, so that the recommendation accuracy of the target commodity is improved; and the recommendation quality of the target commodity is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis, and specifically relates to a commodity recommendation method, device, electronic device and storage medium based on a recommending group. Background Art

[0002] As the scale of e-commerce continues to expand, more and more products need to be recommended to users. The current method of recommending products based on the recommended group is to recommend similar products to the products that the user has purchased. If the product to be promoted belongs to the same category as the product that the user has purchased, the product to be promoted will be pushed to the user.

[0003] However, the above method cannot accurately push recommendations to users. How to improve the accuracy of recommended groups and the quality of recommendations has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a commodity recommendation method, device, electronic device and storage medium based on the recommendation population, which can accurately determine the recommendation population, improve the accuracy of commodity recommendations, and improve the recommendation quality.

[0005] According to a first aspect of the present invention, there is provided a commodity recommendation method based on a recommendation group, comprising:

[0006] Determining a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product;

[0007] Sorting the first users according to the similarity scores, and determining a first user set according to the sorting results of the similarity scores;

[0008] Acquire a third user similar to the second user according to the user characteristics, and determine a second user set according to the third user, where the second user is a user who has purchased the target product;

[0009] Determining a recommended group of people according to the first user set and the second user set;

[0010] The target product is recommended according to the recommendation group.

[0011] According to a second aspect of the present invention, there is provided a commodity recommendation device based on a recommendation group, comprising:

[0012] A similarity score determination module, configured to determine a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product;

[0013] a first user set determining module, configured to sort the first users according to the similarity scores, and determine the first user set according to the sorting results of the similarity scores;

[0014] A second user set determination module, configured to obtain a third user similar to the second user according to user characteristics, and determine a second user set according to the third user, wherein the second user is a user who has purchased the target product;

[0015] A recommended group determination module, used to determine a recommended group according to the first user set and the second user set;

[0016] The recommendation module is used to recommend the target product according to the recommendation group.

[0017] According to a third aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0018] at least one processor; and

[0019] a memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the commodity recommendation method based on the recommendation group described in any embodiment.

[0021] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the commodity recommendation method based on the recommendation group described in any embodiment of the present invention when executed.

[0022] The technical solution provided by the present invention determines the similarity score of the first user according to the characteristics of the purchased goods of the first user and the characteristics of the target goods of the target goods, wherein the first user is a user who has not purchased the target goods; sorts the first users according to the similarity score, and determines the first user set according to the sorting result of the similarity score; obtains a third user similar to the second user according to the user characteristics, and determines the second user set according to the third user, wherein the second user is a user who has purchased the target goods; determines the recommended population according to the first user set and the second user set; and recommends the target goods according to the recommended population. Compared with the current recommendation based on the classification of the purchased goods by the user, the present invention can determine the similarity score according to the characteristics of the purchased goods of the first user and the characteristics of the target goods, and obtain the first user set by sorting and selecting the similarity score. According to the user characteristics, obtains the third user similar to the second user who has purchased the target goods, and obtains the second user set according to the third user. The final recommended population is determined by combining the first user set and the second user, and the first user set and the second user set are determined based on the two dimensions of product characteristics and user characteristics, and then the recommended population is jointly determined from the two dimensions of product characteristics and user characteristics, so that the characteristics of the recommended population are closer to the target goods, and the accuracy of the recommended population is improved. Recommending target products based on the recommending population can improve the accuracy of target product recommendations and improve the quality of target product recommendations.

[0023] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 It is a flowchart of a commodity recommendation method based on a recommendation group provided by an embodiment of the present invention;

[0026] Figure 2 is a flow chart of another method for recommending products based on a recommending group provided by an embodiment of the present invention;

[0027] Figure 3 is a structural schematic diagram of a commodity recommendation device based on recommendation groups provided by an embodiment of the present invention;

[0028] Figure 4It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] As the scale of e-commerce continues to expand, more and more products need to be recommended to users. The current method of recommending products based on the recommended group is to recommend similar products to the products that the user has purchased. If the product to be promoted belongs to the same category as the product that the user has purchased, the product to be promoted will be pushed to the user. However, the above method cannot accurately push products to users. How to improve the accuracy of the recommended group and improve the quality of recommendations has become an urgent problem to be solved.

[0032] Figure 1 This is a flow chart of a method for recommending products based on a recommending group provided by an embodiment of the present invention. This embodiment is applicable to the case of recommending target products to users in e-commerce. The method can be executed by a product recommendation device based on a recommending group. The product recommendation device based on a recommending group can be implemented in the form of hardware and / or software. The method includes:

[0033] S101 : Determine a similarity score of the first user according to characteristics of a purchased product of the first user and characteristics of a target product.

[0034] The first user is a user who has not purchased the target product. The embodiment of the present invention is used to determine the recommended population of the target product. According to the user's consumption record, it can be determined whether the user has purchased the target product. If the user has not purchased the target product, the user is determined as the first user.

[0035] The heterogeneous product features and the target product features contain the same dimensions, which include but are not limited to a combination of one or more of the following dimensions: title, label, brand, or category.

[0036] Each product can be represented by a product feature, and the product vector can describe the product from multiple dimensions. The similarity between the purchased product feature of the purchased product and the target product feature of the target product can be used as the similarity score of the first user.

[0037] Optionally, determining the similarity score of the first user according to the characteristics of the purchased product of the first user and the target product characteristics of the target product can be implemented in the following manner:

[0038] Determine cosine similarity according to the characteristics of the purchased goods and the characteristics of the target goods, and determine the product similarity according to the cosine similarity; calculate the average of the product similarities of all the products purchased by the first user, and use the average as the similarity score of the first user.

[0039] The purchased product features and the target product features are feature vectors with the same dimensions, and the cosine similarity can be calculated based on the purchased product features and the target product features. The cosine similarity is multiplied by the time when the first user purchased the purchased product to obtain the product similarity between the purchased product and the target product. The product progress of each purchased product of the first user and the target product is calculated in the above manner. Then, the average value of the product similarity is calculated based on the product similarities of all the first user's products, and this value is used as the similarity score of the first user. The similarity score of the first user can be used to indicate the similarity between the first user's purchased product and the target product. The above method can accurately score the similarity of the first user who has not purchased the target product.

[0040] Optionally, determining product similarity based on cosine similarity can be implemented as follows:

[0041] A time decay factor is determined according to the purchase time and the current time of the purchased goods of the first user. The longer the purchase time is from the current time, the smaller the time decay factor is. The similarity of the goods is determined according to the product of the time decay factor and the cosine similarity.

[0042] The current time may be the time for determining the recommended population when executing the technical solution of the present invention. The time may be the time for determining the recommended population in the testing phase, or the time for determining the recommended population in the use phase after going online. Since the closer the user's purchase time is to the current time, the closer the user's purchase behavior is to the current user's purchase tendency. Therefore, the time decay factor is determined based on the purchase time of the purchased goods of the first user and the current time. The longer the purchase time is from the current time, the smaller the time decay factor. The closer the purchase time is to the current time, the larger the time decay factor.

[0043] The time decay factor can be calculated by the following formula 1 (power function of e):

[0044]

[0045] Among them, T w is the time decay factor, t current is the current time, t time The time when the first user purchased the purchased item.

[0046] After the time decay factor is obtained, the time decay factor is multiplied by the corresponding product similarity of the purchased product to obtain the optimized product similarity on Yigoo. The above method can more accurately calculate the product similarity based on the time decay factor, thereby improving the accuracy of the similarity score of the first user.

[0047] S102: Sort the first users according to the similarity scores, and determine a first user set according to the sorting results of the similarity scores.

[0048] According to step S101, the similarity score of each first user can be obtained. According to the value of the similarity score, the first users are sorted in descending order. Since the first users are users who have not purchased the target product, the number of first users will be very large, for example, tens of millions. For the convenience of calculation, the first user set can be delineated based on experience based on the sorting results of the similarity scores. For example, the tasks of the first user set are twice or more times the recommended group finally determined. The number of recommended groups can be determined according to conditions such as budget.

[0049] S103: Acquire a third user similar to the second user according to the user characteristics, and determine a second user set according to the third user.

[0050] The second user is a user who has purchased the target product. According to the second user who has purchased the target product, the user characteristics of the second user can be obtained. According to the user characteristics, a third user similar to the second user can be determined. A second user set is selected from the third user. User characteristics can be used to describe multiple attribute dimensions of a user, and multiple attribute dimensions constitute user characteristics.

[0051] Optionally, obtaining a third user similar to the second user according to user characteristics may be implemented in the following manner:

[0052] Acquire a second user feature of a second user; and determine third users similar to the second user and a similarity probability value of each third user based on the second user feature and an approximate neighbor search algorithm.

[0053] The second user feature is a user feature of the second user. Based on the second user feature, the similarity probability value between the second user feature and any other user feature can be calculated using the Approximate Nearest Neighbor (ANN) algorithm. The similarity probability value is used to indicate the degree of similarity between the second used feature and any other user feature. The other user features are sorted in descending order according to the similarity probability value. Then, users corresponding to a predetermined number (such as 10, 50 or 100) of similarity probability values ​​are selected from the sorting results as the third user.

[0054] The nearest neighbor algorithm can be used to find a third user who is similar to the second user, thereby improving the accuracy of similar users and further improving the accuracy of the recommended population.

[0055] Further, after determining the third users similar to the second user and the similarity probability value of each third user, the method further includes:

[0056] If a third user is similar to multiple second users at the same time, the average of the similarity probability values ​​in the multiple similarity results is used as the similarity probability value of the third user, and the third user is deduplicated.

[0057] The third users determined by multiple second users may be repeated. For example, a third user may be similar to multiple second users at the same time. In this case, the repeated third users need to be deduplicated so that the third users similar to all second users are not repeated.

[0058] Optionally, determining the second user set according to the third user may be implemented in the following manner:

[0059] The third users are sorted in descending order according to the similarity probability values; and the second user set is determined according to the sorting results of the preset circled number of people and the similarity probability values.

[0060] After all third users are obtained, the third users are sorted in descending order according to the similarity probability values ​​to obtain a sorting result of the similarity probability values. A plurality of third users are selected from the sorting result of the similarity probability values ​​according to the preset circled number of people as the second user set. The preset original number of people can be determined according to demand and is not limited here.

[0061] S104: Determine a recommended group according to the first user set and the second user set.

[0062] All users in the first user set and the second user set may be used as the recommended group. If the group obtained in the above manner is too large, the recommended group may be further reduced in the following manner.

[0063] Optionally, determining the recommended group according to the first user set and the second user set may be implemented in the following manner:

[0064] According to the sorting position and the set weight in the first user set and the second user set, the target score of the first user or the third user is obtained; sorting is performed according to the target score, and the recommended group is determined according to the sorting result of the target score and the preset number of recommended people.

[0065] The first users in the first user set are sorted in descending order according to the similarity scores. The third users in the second user set are sorted in descending order according to the similarity probability values.

[0066] The target score of any one user in the combination of the first user set and the second user set can be calculated by the following formula 2. Then, the ranking is performed according to the target score to obtain a ranking result of the target score.

[0067] score final_market =w old ×e-(old _ of _ index) +w sim ×e -(sim _ of _ index)

[0068] Among them, score final_market represents the target score, w old represents the weight of the first user set, w sim represents the weight of the second user set, old_of_index represents the ranking position of the first user in the first user set, and sim_of_index represents the ranking position of the third user in the second user set.

[0069] S105: Recommend the target product according to the recommendation group.

[0070] After obtaining the recommended group, you want to push the target product information to the recommended group.

[0071] The commodity recommendation method based on the recommended group provided by the present invention determines the similarity score of the first user according to the characteristics of the purchased commodities of the first user and the characteristics of the target commodity of the target commodity, wherein the first user is a user who has not purchased the target commodity; sorts the first users according to the similarity score, and determines the first user set according to the sorting result of the similarity score; obtains a third user similar to the second user according to the user characteristics, and determines the second user set according to the third user, wherein the second user is a user who has purchased the target commodity; determines the recommended group according to the first user set and the second user set; and recommends the target commodity according to the recommended group. Compared with the current recommendation based on the classification of the purchased commodities by the user, the present invention can determine the similarity score according to the characteristics of the purchased commodities of the first user and the characteristics of the target commodity, and obtain the first user set by sorting and selecting the similarity score. According to the user characteristics, obtains the third user similar to the second user who has purchased the target commodity, and obtains the second user set according to the third user. The final recommended group is determined according to the combination of the first user set and the second user, and the combination of the first user set and the second user is determined respectively based on the two dimensions of commodity characteristics and user characteristics, and then the recommended group is jointly determined from the two dimensions of commodity characteristics and user characteristics, so that the characteristics of the recommended group are closer to the target commodity, and the accuracy of the recommended group is improved. Recommending target products based on the recommending population can improve the accuracy of target product recommendations and improve the quality of target product recommendations.

[0072] Figure 2 A flowchart of a product recommendation based on a recommending group provided in an embodiment of the present invention, as a further explanation of the above embodiment, includes:

[0073] Step 201: Determine cosine similarity based on the features of the purchased product and the features of the target product.

[0074] Step 202: Determine a time decay factor according to the purchase time of the purchased product by the first user and the current time.

[0075] The first user is a user who has not purchased the target product. The longer the purchase time is from the current time, the smaller the time decay factor;

[0076] Step 203: Determine the product similarity according to the product of the time decay factor and the cosine similarity;

[0077] Step 204: Calculate the average of the product similarities of all products purchased by the first user, and use the average as the similarity score of the first user.

[0078] Step 205: Sort the first users according to the similarity scores, and determine the first user set according to the sorting results of the similarity scores.

[0079] Step 206: Obtain a second user feature of the second user; and determine third users similar to the second user and a similarity probability value of each third user based on the second user feature and an approximate neighbor search algorithm.

[0080] Step 207: If a third user is similar to multiple second users at the same time, the average of the similarity probability values ​​in the multiple similarity results is used as the similarity probability value of the third user, and the third user is deduplicated.

[0081] Step 208: sort the third users in descending order according to the similarity probability values; and determine the second user set according to the sorting result of the preset circled number of people and the similarity probability values.

[0082] Step 209: Obtain a target score of the first user or the third user according to the sorting position and the set weight in the first user set and the second user set; sort according to the target score, and determine the recommended group according to the sorting result of the target score and the preset number of recommended people.

[0083] Step 210: Recommend the target product according to the recommendation group.

[0084] Figure 3 A structural schematic diagram of a product recommendation device based on a recommendation group provided in an embodiment of the present invention, wherein the product recommendation device based on a recommendation group is used to implement the above-mentioned product recommendation method based on a recommendation group, and comprises: a similarity score determination module 31, a first user set determination module 32, a second user set determination module 33, a recommendation group determination module 34 and a recommendation module 35.

[0085] A similarity score determination module 31, configured to determine a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product;

[0086] A first user set determination module 32, configured to sort the first users according to the similarity scores, and determine the first user set according to the sorting results of the similarity scores;

[0087] A second user set determination module 33, configured to obtain a third user similar to the second user according to user characteristics, and determine a second user set according to the third user, wherein the second user is a user who has purchased the target product;

[0088] A recommended group determination module 34, configured to determine a recommended group according to the first user set and the second user set;

[0089] The recommendation module 35 is used to recommend the target product according to the recommendation group.

[0090] Based on the above implementation, optionally, the similarity score determination module 31 is used to:

[0091] Determine the cosine similarity based on the characteristics of the purchased product and the characteristics of the target product, and determine the product similarity based on the cosine similarity;

[0092] The average of the commodity similarities of all commodities purchased by the first user is calculated, and the average is used as the similarity score of the first user.

[0093] Based on the above implementation, optionally, the similarity score determination module 31 is used to:

[0094] Determine a time decay factor according to the purchase time of the purchased product of the first user and the current time, wherein the longer the purchase time is from the current time, the smaller the time decay factor is;

[0095] The product similarity is determined according to the product of the time decay factor and the cosine similarity.

[0096] Based on the above implementation, optionally, the second user set determination module 33 is used to:

[0097] Acquire a second user characteristic of the second user;

[0098] According to the second user feature and an approximate neighbor search algorithm, third users similar to the second user and a similarity probability value of each third user are determined.

[0099] Based on the above implementation, optionally, the second user set determination module 33 is used to:

[0100] If a third user is similar to multiple second users at the same time, the average of the similarity probability values ​​in the multiple similarity results is used as the similarity probability value of the third user, and the third user is deduplicated.

[0101] Based on the above implementation, optionally, the second user set determination module 33 is used to:

[0102] sorting the third users in descending order according to the similarity probability values;

[0103] The second user set is determined according to the preset number of circled persons and the ranking results of the similarity probability values.

[0104] Based on the above implementation, optionally, the recommended population determination module 34 is used to:

[0105] Obtaining a target score of the first user or the third user according to a sorting position in the first user set and the second user set and a set weight;

[0106] Sorting is performed according to the target scores, and the recommended population is determined according to the sorting results of the target scores and the preset number of recommended people.

[0107] The commodity recommendation device based on the recommended group provided by the present invention comprises a similarity score determination module 31, which is used to determine the similarity score of the first user according to the characteristics of the purchased commodities of the first user and the target commodity characteristics of the target commodity, wherein the first user is a user who has not purchased the target commodity; a first user set determination module 32, which is used to sort the first user according to the similarity score, and determine the first user set according to the sorting result of the similarity score; a second user set determination module 33, which is used to obtain a third user similar to the second user according to the user characteristics, and determine the second user set according to the third user, wherein the second user is a user who has purchased the target commodity; a recommended group determination module 34, which is used to determine the recommended group according to the first user set and the second user set; and a recommendation module 35, which is used to recommend the target commodity according to the recommended group. Compared with the current recommendation based on the classification of the purchased commodities by the user, the present invention can determine the similarity score according to the characteristics of the purchased commodities of the first user and the characteristics of the target commodity, and obtain the first user set by sorting and selecting the similarity score. According to the user characteristics, a third user similar to the second user who has purchased the target commodity is obtained, and the second user set is obtained according to the third user. The final recommended group is determined based on the combination of the first user set and the second user, and the first user set and the second user are determined based on the two dimensions of product characteristics and user characteristics, and then the recommended group is determined from the two dimensions of product characteristics and user characteristics, so that the characteristics of the recommended group are closer to the target product, and the accuracy of the recommended group is improved. Recommending the target product based on the recommended group can improve the recommendation accuracy of the target product and improve the recommendation quality of the target product.

[0108] The commodity recommendation device based on recommending groups provided in the embodiment of the present invention can execute the commodity recommendation method based on recommending groups provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0109] Figure 4 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0110] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a commodity recommendation method based on a recommended population.

[0113] In some embodiments, the commodity recommendation method based on the recommendation group may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the commodity recommendation method based on the recommendation group described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the commodity recommendation method based on the recommendation group in any other appropriate manner (e.g., by means of firmware).

[0114] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] The computer program for implementing the commodity recommendation method based on the recommendation group of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or the block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as an independent software package, partially on the machine and partially on a remote machine, or completely on a remote machine or server.

[0116] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions can be used to enable a processor to execute a commodity recommendation method based on a recommendation group, the method comprising:

[0117] Determining a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product;

[0118] Sorting the first users according to the similarity scores, and determining a first user set according to the sorting results of the similarity scores;

[0119] Acquire a third user similar to the second user according to the user characteristics, and determine a second user set according to the third user, where the second user is a user who has purchased the target product;

[0120] Determining a recommended group of people according to the first user set and the second user set;

[0121] The target product is recommended according to the recommendation group.

[0122] On the basis of the above implementation, determining the similarity score of the first user according to the purchased product features of the first user and the target product features of the target product includes:

[0123] Determine the cosine similarity based on the characteristics of the purchased product and the characteristics of the target product, and determine the product similarity based on the cosine similarity;

[0124] The average of the commodity similarities of all commodities purchased by the first user is calculated, and the average is used as the similarity score of the first user.

[0125] Based on the above implementation, determining the product similarity according to the cosine similarity includes:

[0126] Determine a time decay factor according to the purchase time of the purchased product of the first user and the current time, wherein the longer the purchase time is from the current time, the smaller the time decay factor is;

[0127] The product similarity is determined according to the product of the time decay factor and the cosine similarity.

[0128] On the basis of the above implementation, obtaining a third user similar to the second user according to user characteristics includes:

[0129] Acquire a second user characteristic of the second user;

[0130] According to the second user feature and an approximate neighbor search algorithm, third users similar to the second user and a similarity probability value of each third user are determined.

[0131] On the basis of the above implementation, after determining the third users similar to the second user and the similarity probability value of each third user, the method further includes:

[0132] If a third user is similar to multiple second users at the same time, the average of the similarity probability values ​​in the multiple similarity results is used as the similarity probability value of the third user, and the third user is deduplicated.

[0133] On the basis of the above implementation manner, determining the second user set according to the third user includes:

[0134] sorting the third users in descending order according to the similarity probability values;

[0135] The second user set is determined according to the preset number of circled persons and the ranking results of the similarity probability values.

[0136] On the basis of the above implementation, determining the recommended group according to the first user set and the second user set includes:

[0137] Obtaining a target score of the first user or the third user according to a sorting position in the first user set and the second user set and a set weight;

[0138] Sorting is performed according to the target scores, and the recommended population is determined according to the sorting results of the target scores and the preset number of recommended people.

[0139] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0140] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0141] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0142] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0143] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0144] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A product recommendation method based on recommendation crowd, It is characterized in that include: Determining a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product; Sorting the first users according to the similarity scores, and determining a first user set according to the sorting results of the similarity scores; Acquire a third user similar to the second user according to the user characteristics, and determine a second user set according to the third user, where the second user is a user who has purchased the target product; Determining a recommended group of people according to the first user set and the second user set; The target product is recommended according to the recommendation group.

2. The method according to claim 1, It is characterized in that Determining a similarity score of the first user according to the purchased product feature of the first user and the target product feature of the target product includes: Determine the cosine similarity based on the characteristics of the purchased product and the characteristics of the target product, and determine the product similarity based on the cosine similarity; The average of the commodity similarities of all commodities purchased by the first user is calculated, and the average is used as the similarity score of the first user.

3. The method according to claim 2, It is characterized in that Determine product similarity based on cosine similarity, including: Determine a time decay factor according to the purchase time of the purchased product of the first user and the current time, wherein the longer the purchase time is from the current time, the smaller the time decay factor is; The product similarity is determined according to the product of the time decay factor and the cosine similarity.

4. The method according to claim 1, It is characterized in that Acquiring a third user similar to the second user according to user characteristics includes: Acquire a second user characteristic of the second user; According to the second user feature and an approximate neighbor search algorithm, third users similar to the second user and a similarity probability value of each third user are determined.

5. The method according to claim 4, It is characterized in that After determining third users similar to the second user and the similarity probability value of each third user, the method further includes: If a third user is similar to multiple second users at the same time, the average of the similarity probability values ​​in the multiple similarity results is used as the similarity probability value of the third user, and the third user is deduplicated.

6. The method according to claim 5, It is characterized in that Determining a second user set according to the third user includes: sorting the third users in descending order according to the similarity probability values; The second user set is determined according to the preset number of circled persons and the ranking results of the similarity probability values.

7. The method according to claim 6, It is characterized in that Determining a recommended group of people according to the first user set and the second user set includes: Obtaining a target score of the first user or the third user according to a sorting position in the first user set and the second user set and a set weight; Sorting is performed according to the target scores, and the recommended population is determined according to the sorting results of the target scores and the preset number of recommended people.

8. A commodity recommendation device based on recommendation groups, It is characterized in that include: A similarity score determination module, configured to determine a similarity score of a first user according to characteristics of a purchased product of the first user and characteristics of a target product of the target product, wherein the first user is a user who has not purchased the target product; a first user set determining module, configured to sort the first users according to the similarity scores, and determine the first user set according to the sorting results of the similarity scores; A second user set determination module, configured to obtain a third user similar to the second user according to user characteristics, and determine a second user set according to the third user, wherein the second user is a user who has purchased the target product; A recommended group determination module, used to determine a recommended group according to the first user set and the second user set; The recommendation module is used to recommend the target product according to the recommendation group.

9. An electronic device, It is characterized in that include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the commodity recommendation method based on recommendation population as described in any one of claims 1-7.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the commodity recommendation method based on the recommendation group according to any one of claims 1 to 7 when executed.