Personalized recommendation method, device, electronic device and readable storage medium

By splitting the goods into raw materials and combining them, and generating personalized recommendations based on user historical behavior and characteristics, the problem of insufficient diversity of product combinations in the prior art is solved, the accuracy and diversity of recommendations are improved, and the user experience is improved.

CN110555752BActive Publication Date: 2024-11-19BEIJING SANKUAI ONLINE TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN201910704633.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-07-31
Publication Date
2024-11-19
Estimated Expiration
2039-07-31

AI Technical Summary

Technical Problem

Existing personalized recommendation technologies cannot generate highly diverse product portfolios and lack recommendations for user preferences.

Method used

By splitting the product into raw materials, and then combining to generate more types of target product combinations based on the user's historical behavior records and current characteristics, improving the accuracy and diversity of recommendations.

Benefits of technology

It has achieved the improvement of the diversity and accuracy of the recommended product portfolio, met the diverse needs of users, and enhanced the user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110555752B_ABST
    Figure CN110555752B_ABST
Patent Text Reader

Abstract

The embodiments of the present disclosure provide a personalized recommendation method, device, electronic device and readable storage medium, the method comprising: combining the commodities contained in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination; determining the raw materials of each reference commodity contained in the reference commodity combination for the reference commodity combination, and obtaining the raw material combination corresponding to the reference commodity combination; combining the candidate commodities into a target commodity combination that meets the raw material combination, and recommending the target commodity combination to the target user. The commodities can be split into raw materials, and then the raw materials can be combined to generate more types of target commodity combinations, thereby improving the diversity of the target commodity combinations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of personalized recommendation technology, and in particular, to a personalized recommendation method, device, electronic device, and readable storage medium. Background Art

[0002] In the field of personalized recommendations, multiple product combinations can usually be recommended to users, which can facilitate purchases and help increase the order conversion rate.

[0003] After studying the existing solutions, the inventors found that the recommended products that the existing solutions cannot generate are single and lack diversity. Summary of the invention

[0004] The embodiments of the present disclosure provide a personalized recommendation method, device, electronic device and readable storage medium, which can split commodities into raw materials and then combine the raw materials to generate more types of target commodity combinations, thereby improving the diversity of target commodity combinations.

[0005] Predicting user types through historical feature sequences and current features helps improve prediction accuracy.

[0006] According to a first aspect of an embodiment of the present disclosure, a personalized recommendation method is provided, the method comprising:

[0007] According to the reference type combination, the commodities included in the historical behavior record of the target user are combined into a reference commodity combination corresponding to the target user;

[0008] For the reference commodity combination, determining the raw materials of each reference commodity included in the reference commodity combination, and obtaining a raw material combination corresponding to the reference commodity combination;

[0009] The candidate commodities are combined into a target commodity combination that matches the raw material combination, and the target commodity combination is recommended to the target user.

[0010] According to a second aspect of an embodiment of the present disclosure, a personalized recommendation device is provided, the device comprising:

[0011] A reference product combination generation module is used to combine the products included in the historical behavior record of the target user into a reference product combination corresponding to the target user according to the reference type combination;

[0012] A raw material combination generating module, for determining the raw materials of each reference commodity contained in the reference commodity combination, and obtaining a raw material combination corresponding to the reference commodity combination;

[0013] The recommendation module is used to combine candidate commodities into a target commodity combination that meets the raw material combination, and recommend the target commodity combination to the target user.

[0014] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0015] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the aforementioned personalized recommendation method is implemented when the processor executes the program.

[0016] According to a fourth aspect of an embodiment of the present disclosure, a readable storage medium is provided, characterized in that when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the aforementioned personalized recommendation method.

[0017] The embodiment of the present disclosure provides a personalized recommendation method and device, the method comprising: combining the commodities contained in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination; determining the raw materials of each reference commodity contained in the reference commodity combination for the reference commodity combination, and obtaining the raw material combination corresponding to the reference commodity combination; combining the candidate commodities into a target commodity combination that meets the raw material combination, and recommending the target commodity combination to the target user. The commodities can be split into raw materials, and then the raw materials can be combined to generate more types of target commodity combinations, thereby improving the diversity of the target commodity combinations. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the description of the embodiments of the present disclosure will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0019] Figure 1 A flowchart of the steps of a personalized recommendation method in an embodiment of the present disclosure is shown;

[0020] Figure 2 A flowchart of the steps of a personalized recommendation method in another embodiment of the present disclosure is shown;

[0021] Figure 3 A structural diagram of a personalized recommendation device in an embodiment of the present disclosure is shown;

[0022] Figure 4 A structural diagram of a personalized recommendation device in another embodiment of the present disclosure is shown;

[0023] Figure 5 A structural diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present disclosure.

[0025] Embodiment 1

[0026] Reference Figure 1 , which shows a step flow chart of a personalized recommendation method in an embodiment of the present disclosure, as follows.

[0027] Step 101 : Combining commodities included in the historical behavior record of a target user into a reference commodity combination corresponding to the target user according to a reference type combination.

[0028] The embodiments of the present disclosure may be applicable to application scenarios of recommending combinations of commodities, especially application scenarios of recommending combinations of dishes.

[0029] Among them, the reference type combination is a more reasonable and better quality commodity type combination, which is composed of at least two commodity types. It can be understood that the reference type combination can be set according to experience, rationality, and demand, or it can be selected from the inherent type combination according to certain evaluation indicators. For example, when giving gifts, in order to make the gifts meet the needs of everyone in the family, the commodity types that multiple people like can be combined to obtain a reference type combination, and a reference type combination is obtained: a combination of three types of snacks, cigarettes, and skin care products; for another example, staple foods and side dishes are combined to obtain a reference type combination; for another example, a commodity type combination with a user click rate higher than a certain threshold can also be selected from the user's historical behavior records as a reference type combination. The embodiments of the present disclosure do not limit the generation of reference type combinations.

[0030] The historical behavior records can be obtained from the backend server of a sales platform, and the platform can record the user's behavior, including: click behavior, order behavior, collection behavior, etc. The specific type of behavior in the embodiment of the present disclosure is not limited, as long as the behavior can represent the user's preference for the product.

[0031] It can be understood that, for a reference type combination, the number of commodities included in the generated reference commodity combination is the same as the number of commodity types included in the reference type combination. For example, the reference type combination is type T1-type T2-type T3, and the reference commodity combination generated based on the reference type combination is commodity C1-commodity C2-commodity C3, so that the commodity type of commodity C1 is T1, the commodity type of commodity C2 is T2, and the commodity type of commodity C3 is T3.

[0032] In the embodiments of the present disclosure, the reference type combination is selected based on experience or demand or rationality or evaluation indicators, so that the generated reference product combination has higher rationality, higher flexibility, higher quality, and is also associated with user preferences.

[0033] Step 102: for the reference commodity combination, determine the raw materials of each reference commodity included in the reference commodity combination, and obtain a raw material combination corresponding to the reference commodity combination.

[0034] Specifically, the raw materials of each commodity are manually input into a preset commodity information library, so that the computer can extract the raw materials of the reference commodity from the commodity information library according to the commodity identification.

[0035] Among them, the raw materials of a commodity refer to the basic materials that constitute the commodity. In the embodiments of the present disclosure, the raw materials may be the main raw materials, and the additional materials may not be concerned. For example, for scrambled eggs with fungus, the main raw materials may be fungus and eggs, and the additional materials may also include: onions, garlic, etc. At this time, we only use fungus and eggs as raw materials.

[0036] It can be understood that the raw material combination is composed of the raw materials of all reference commodities in the reference commodity. For example, for the commodity combination: commodity C1-commodity C2-commodity C3, the raw materials of commodity C1 include: M1, M2, the raw material of commodity C2 is M3, and the raw material of commodity C3 is M4, then the raw material combination is: raw material M1-raw material M2-raw material M3-raw material M4.

[0037] Step 103: Combine candidate commodities into a target commodity combination that meets the raw material combination, and recommend the target commodity combination to the target user.

[0038] The candidate products are products to be recommended, which may be all products sold on the sales platform, or products after filtering all products according to the filtering conditions. The filtering conditions may be price conditions, location conditions or other conditions, for example, filtering out products with a price greater than 100 from all products, or filtering out products whose merchant locations are not in a preset area from all products. It is understood that the embodiments of the present disclosure do not limit the filtering conditions.

[0039] It can be understood that the raw materials of all commodities included in the target commodity combination include at least the raw materials included in the raw material combination. For example, the raw material combination is: raw material M1-raw material M2-raw material M3-raw material M4, and the target commodity combination can be: commodity C4-commodity C5, wherein the raw materials of commodity C4 include: M1, M2, M3, and the raw materials of commodity C5 include: M4, M5. It should be noted that the number of commodities included in the generated target commodity combination is not necessarily the same as the number of commodities included in the reference commodity combination, but the number of raw materials included in the target commodity combination is greater than or equal to the number of raw materials included in the reference commodity combination.

[0040] In the embodiments of the present disclosure, combining candidate commodities according to raw material combinations can help improve the diversity of commodity combinations.

[0041] It is understandable that the embodiments of the present disclosure can also combine the user's historical orders on the food delivery platform to recommend ingredients sold on the food delivery platform to the user, which can meet the user's own cooking needs while ensuring that the user purchases all ingredients at one time to avoid missing ingredients. For example, the raw material combination and the target product combination are directly recommended to the user, so that the user can purchase all ingredients from the food delivery platform according to the raw material combination, so that the user can make meals at home, that is, the target product combination.

[0042] In summary, the embodiments of the present disclosure provide a personalized recommendation method, the method comprising: combining the commodities contained in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination; determining the raw materials of each reference commodity contained in the reference commodity combination for the reference commodity combination, and obtaining a raw material combination corresponding to the reference commodity combination; combining the candidate commodities into a target commodity combination that meets the raw material combination, and recommending the target commodity combination to the target user. The commodities can be split into raw materials, and then the raw materials can be combined to generate more types of target commodity combinations, thereby improving the diversity of the target commodity combinations.

[0043] Embodiment 2

[0044] Reference Figure 2 , which shows a specific step flow chart of a personalized recommendation method in another embodiment of the present disclosure, as follows.

[0045] Step 201, obtaining commodities and commodity type combinations on which the target user performs preset operations within a preset historical time from the historical behavior records of the target user, and obtaining reference commodities and reference type combinations, wherein the preset operations include at least one of: an ordering operation and a clicking operation.

[0046] The preset historical time may be pushed from the current time to the previous time, for example, one month before the current time, one week before the current time, and so on.

[0047] In an embodiment of the present disclosure, the reference type combination is a combination of commodity types that the user has clicked on or ordered.

[0048] The embodiments of the present disclosure can directly use the commodity type combinations that the user has clicked or ordered from the historical behavior records as reference type combinations, so that the reference type combinations meet the user preferences.

[0049] Optionally, in an embodiment of the present disclosure, step 201 includes sub-steps A1 to A4:

[0050] Sub-step A1, obtaining the product combination for which the target user performs a preset operation within a preset historical time.

[0051] In the embodiments of the present disclosure, the user may click on or place an order for a product, or the user may select multiple products to form an order, or the user may directly select a product combination from the product combinations provided by the merchant.

[0052] Sub-step A2: clustering the commodity combinations according to the types of each commodity to obtain candidate type combinations.

[0053] Specifically, clustering is to determine the candidate type combination of each product combination, and to classify the product combinations corresponding to the same candidate type combination into the same set.

[0054] Among them, clustering is to determine the candidate type combination of each commodity combination, including: merging the same types of different commodities in the same commodity combination into one type. For example, if the commodity combination is: commodity C6-commodity C7-commodity C7, the type of commodity C6 and the type of commodity C7 are both type T1, and the type of commodity C7 is type T2, so after clustering, the candidate type combination obtained is: type T1-type T2.

[0055] Sub-step A3: for each candidate type combination, according to the number of commodity combinations included and the total number of commodity combinations included in each candidate type combination, calculate a reference score of the candidate type combination.

[0056] Specifically, the reference score can be the ratio of the number of included product combinations to the total number of product combinations included in each candidate type combination, or it can be a further transformed value of the ratio, as long as the following relationship is guaranteed: for a candidate type combination, the more product combinations it includes and the fewer the total number of product combinations included in each candidate type combination, the higher its reference score; if the fewer the number of product combinations it includes and the greater the total number of product combinations included in each candidate type combination, the lower its reference score.

[0057] It can be understood that the reference score of a candidate type combination represents the occurrence frequency of the candidate type combination.

[0058] Sub-step A4, arranging the candidate type combinations in descending order according to the reference scores, and selecting one or more candidate type combinations with the highest ranking to obtain a reference type combination.

[0059] The embodiments of the present disclosure may select candidate type combinations with higher occurrence frequencies as reference type combinations, thereby ensuring that the reference type combinations are more consistent with user preferences.

[0060] Step 202: Filter the reference commodities according to preset conditions, where the preset conditions include at least one of a limiting condition on price and a limiting condition on commodity type.

[0061] Among them, the preset conditions can be set according to actual needs, and the embodiments of the present disclosure do not limit them.

[0062] The price restriction condition may be a price range, thereby filtering out reference products with prices that are too high or too low, a minimum price, thereby filtering out reference products with prices that are too low, or a maximum price, thereby filtering out reference products with prices that are too high.

[0063] The limiting condition for the product type may be a specified product type, thereby filtering out reference products that do not belong to the product type.

[0064] The embodiments of the present disclosure can filter out reference commodities that do not meet preset conditions, thereby using high-quality commodities that meet the conditions as reference commodities.

[0065] Step 203: Expand the reference commodities after filtering using a preset commodity set.

[0066] The embodiments of the present disclosure can select commodities that meet the extension conditions from the preset commodity set and add them to the reference commodities, so that more reference commodities can be provided to improve the diversity of the target commodity combination. The extension conditions can be set according to the actual application scenario, and the embodiments of the present disclosure do not limit them.

[0067] Optionally, in another embodiment of the present disclosure, step 203 includes sub-steps 2031 to 2032:

[0068] Sub-step 2031, for a preset product set, calculate the correlation coefficient between each product and the reference product.

[0069] Specifically, various attribute features of a product may be combined into a feature vector, and the cosine similarity between the feature vectors may be calculated to determine the correlation coefficient between two products.

[0070] Among them, attribute characteristics may include but are not limited to: product type, sales volume, price, rating, etc.

[0071] Sub-step 2032: sort the commodities in descending order according to the correlation coefficients, and select one or more commodities with the highest ranking as expanded reference commodities.

[0072] The embodiments of the present disclosure can expand commodities similar to reference commodities into the reference commodities, so that the expanded reference commodities have higher accuracy.

[0073] Step 204: Combine the reference commodities according to the reference type combination to obtain a reference commodity combination.

[0074] Specifically, the reference commodities are classified according to each commodity type in the reference type combination to obtain a candidate commodity set corresponding to each commodity type, and then a candidate commodity is selected from the candidate commodity set corresponding to each commodity type as a commodity in the reference commodity combination.

[0075] Step 205 : for the reference commodity combination, determine the raw materials of each reference commodity included in the reference commodity combination, and obtain a raw material combination corresponding to the reference commodity combination.

[0076] This step can refer to the detailed description of step 102 and will not be described again here.

[0077] Step 206: Obtain the location of the target user.

[0078] Specifically, the location of the target user includes, but is not limited to: the current location acquired by the positioning system on the mobile terminal, the location set by the user, and the frequently used locations in the historical positioning records.

[0079] Step 207, obtaining merchants whose distance from the location is less than a preset distance threshold to obtain candidate merchants.

[0080] The distance threshold can be set according to the actual application scenario, and the embodiments of the present disclosure do not limit it. For example, it can be set to 500 meters or 1 kilometer. It can be understood that the distance threshold is usually set to a smaller distance to ensure that the candidate merchant is a merchant near the target user.

[0081] Step 208: Acquire the products provided by the candidate merchants to obtain candidate products.

[0082] Specifically, the commodities provided by the candidate merchants may be obtained from the database of the sales platform.

[0083] The embodiments of the present disclosure may select commodities provided by merchants that are closer to the user as candidate commodities, thereby facilitating delivery and reducing delivery costs.

[0084] Step 209 : for each raw material in each of the raw material combinations, select a candidate commodity that matches the raw material from the candidate commodities.

[0085] Specifically, a candidate commodity whose raw materials include the raw materials is selected from the candidate commodities.

[0086] Step 210 , for each raw material in each of the raw material combinations, select a candidate commodity from candidate commodities that match the raw material and add it to the target commodity combination corresponding to the raw material combination.

[0087] Specifically, it is necessary to add corresponding candidate commodities to each raw material in the raw material combination one by one, and before adding, it is necessary to determine whether the candidate commodity already exists in the target commodity combination. For example, if the raw material combination is: raw material M1-raw material M2-raw material M3-raw material M4, then for raw material M1, select commodity C4 containing raw material M1 and add it to the target commodity combination. For raw materials M2 and M3, since commodity C4 containing raw materials M2 and M3 already exists in the target commodity combination, commodity C4 is no longer added to the target commodity combination. For raw material M4, since commodity C5 containing raw material M4 does not exist in the target commodity combination, commodity C5 is added to the target commodity combination; similarly, commodity C5 containing raw material M5 is not added to the target commodity combination again.

[0088] In practical applications, the number of target product combinations can also be limited, so that each candidate product can be matched with the raw material combination first, and the candidate product containing the most raw materials in the raw material combination is first selected and added to the target product combination; then the candidate products containing the remaining raw materials are selected from the remaining candidate products and added to the target product combination; and this cycle is repeated until there are no remaining raw materials, that is, the target product combination contains all the raw materials in the raw material combination.

[0089] Step 211, for each target product combination, obtain the ranking index of the corresponding candidate merchants, wherein the ranking index includes at least one of historical sales volume and evaluation index.

[0090] In the embodiment of the present disclosure, the commodities of the same merchant are combined as candidate commodities to obtain a target commodity combination, and commodities of different merchants are not combined, so as to help reduce the delivery cost. Therefore, all commodities in a target commodity combination belong to the same merchant.

[0091] The historical sales volume may be the total historical sales volume of all commodities of the candidate merchant, which may be obtained from statistics of the merchant's historical orders.

[0092] The evaluation index may be a user's evaluation of a candidate merchant, and may be obtained from an evaluation database.

[0093] Step 212: determine the delivery complexity of the target product combination.

[0094] The delivery complexity is used to represent the cost of delivering the target product combination from the candidate merchant to the target user. The higher the cost, the higher the delivery complexity; the lower the cost, the lower the delivery complexity.

[0095] Optionally, in another embodiment of the present disclosure, step 212 includes sub-step B1:

[0096] Sub-step B1, for each target product combination, determining the delivery complexity according to the distance between the location of the corresponding candidate merchant and the location of the target user.

[0097] In the embodiments of the present disclosure, the delivery complexity can be determined according to the distance between the candidate merchant and the target user. For example, the distance is directly used as the delivery complexity, or the distance is normalized or transformed to obtain a value between 0 and 1 as the delivery complexity. The embodiments of the present disclosure do not limit the specific formula of the delivery complexity, and only need to ensure the following relationship: the greater the distance, the greater the delivery complexity; the smaller the distance, the smaller the delivery complexity.

[0098] The embodiments of the present disclosure can determine the delivery complexity according to the distance. Since the distance is usually determined very accurately, the accuracy of the delivery complexity is relatively high.

[0099] Step 213, recommending the target product combination to the target user in the order of the corresponding ranking index, the corresponding reference score, and the delivery complexity, wherein the reference score is the reference score of the reference type combination corresponding to the target product combination.

[0100] Specifically, target product combinations with higher merchant sales history, higher frequency of corresponding reference type combinations, and shorter delivery distances may be preferentially recommended to target users; or target product combinations with higher merchant evaluations, higher frequency of corresponding reference type combinations, and shorter delivery distances may be preferentially recommended to target users.

[0101] The embodiments of the present disclosure can preferentially recommend target product combinations with higher ranking indexes, higher reference scores, and lower delivery complexity to target users, thereby helping to improve the recommendation success rate.

[0102] In summary, the embodiments of the present disclosure provide a personalized recommendation method based on Embodiment 1. In addition to the beneficial effects of Embodiment 1, the commodity type combination that the user has clicked or ordered can be directly used as a reference type combination from the historical behavior record, so that the reference type combination conforms to the user's preference; the candidate type combination with a higher frequency of occurrence can also be selected as the reference type combination, so as to ensure that the reference type combination is more in line with the user's preference; the reference commodities that do not meet the preset conditions can also be filtered out, so that the qualified high-quality commodities can be used as reference commodities; the commodities similar to the reference commodities can also be expanded to the reference commodities, so that the expanded reference commodities have higher accuracy; the commodities provided by merchants close to the user can also be used as candidate commodities, so as to facilitate the delivery of the target commodity combination and reduce the delivery cost; the target commodity combination with higher sorting index, higher reference score and lower delivery complexity (especially the closer delivery distance) can also be recommended to the target user in priority, which helps to improve the recommendation success rate.

[0103] Embodiment 3

[0104] Reference Figure 3 , which shows a structural diagram of a personalized recommendation device in another embodiment of the present disclosure, as follows.

[0105] The reference commodity combination generating module 301 is used to combine commodities included in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination.

[0106] The raw material combination generating module 302 is used to determine the raw materials of each reference commodity included in the reference commodity combination, and obtain the raw material combination corresponding to the reference commodity combination.

[0107] The recommendation module 303 is used to combine candidate commodities into a target commodity combination that meets the raw material combination, and recommend the target commodity combination to the target user.

[0108] In summary, the embodiment of the present disclosure provides a personalized recommendation device, which includes: combining the commodities contained in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination; determining the raw materials of each reference commodity contained in the reference commodity combination for the reference commodity combination, and obtaining the raw material combination corresponding to the reference commodity combination; combining the candidate commodities into a target commodity combination that meets the raw material combination, and recommending the target commodity combination to the target user. The commodities can be split into raw materials, and then the raw materials can be combined to generate more types of target commodity combinations, thereby improving the diversity of the target commodity combinations.

[0109] Embodiment 3 is a device embodiment corresponding to Embodiment 1. The detailed description may refer to Embodiment 1 and will not be repeated here.

[0110] Embodiment 4

[0111] Reference Figure 4 , which shows a structural diagram of a personalized recommendation device in an embodiment of the present disclosure, as follows.

[0112] The reference product combination generation module 401 is used to combine the products included in the historical behavior record of the target user into a reference product combination corresponding to the target user according to the reference type combination; optionally, in an embodiment of the present disclosure, the reference product combination generation module 401 includes:

[0113] The commodity and type combination extraction submodule 4011 is used to obtain the commodity and commodity type combination on which the target user performs preset operations within a preset historical time from the historical behavior record of the target user, and obtain reference commodities and reference type combinations, wherein the preset operations include at least one of: placing an order and clicking operations.

[0114] The filtering submodule 4012 is used to filter the reference commodities according to preset conditions, wherein the preset conditions include at least one of a limiting condition on price and a limiting condition on commodity type.

[0115] The expansion submodule 4013 is used to expand the reference commodities after filtering by using a preset commodity set.

[0116] The reference commodity combination generating submodule 4014 is used to combine the reference commodities according to the reference type combination to obtain a reference commodity combination.

[0117] The raw material combination generating module 402 is used to determine the raw materials of each reference commodity included in the reference commodity combination, and obtain the raw material combination corresponding to the reference commodity combination.

[0118] The recommendation module 403 is used to combine candidate commodities into a target commodity combination that meets the raw material combination, and recommend the target commodity combination to the target user; optionally, in an embodiment of the present disclosure, the recommendation module 403 includes:

[0119] The location determination submodule 4031 is used to obtain the location of the target user.

[0120] The candidate merchant selection submodule 4032 is used to obtain merchants whose distance from the location is less than a preset distance threshold to obtain candidate merchants.

[0121] The candidate commodity acquisition submodule 4033 is used to acquire commodities provided by the candidate merchants to obtain candidate commodities.

[0122] The raw material matching submodule 4034 is used to select a candidate commodity that matches the raw material from the candidate commodities for each raw material in each of the raw material combinations.

[0123] The candidate commodity adding submodule 4035 is used to select a candidate commodity from the candidate commodities that match the raw material for each raw material in each raw material combination, and add the candidate commodity to the target commodity combination corresponding to the raw material combination.

[0124] The recommendation submodule 4036 is used to recommend the target product combination to the target user.

[0125] Optionally, in another embodiment of the present disclosure, the commodity and type combination extraction submodule 4011 includes:

[0126] The product combination acquisition unit is used to acquire the product combination for which the target user performs a preset operation within a preset historical time.

[0127] The commodity combination clustering unit is used to cluster the commodity combination according to the type of each commodity to obtain candidate type combinations.

[0128] The reference score calculation unit is used to calculate the reference score of each candidate type combination according to the number of commodity combinations included and the total number of commodity combinations included in each candidate type combination.

[0129] A reference type combination selection unit is used to arrange the candidate type combinations in descending order according to the reference scores, and select one or more candidate type combinations with the highest ranking to obtain a reference type combination.

[0130] Optionally, in another embodiment of the present disclosure, the expansion submodule 4013 includes:

[0131] The correlation coefficient calculation unit is used to calculate the correlation coefficient between each commodity and the reference commodity in the preset commodity set.

[0132] The expansion unit is used to arrange the commodities in descending order according to the correlation coefficients, and select one or more commodities with the highest ranking as reference commodities for expansion.

[0133] Optionally, in another embodiment of the present disclosure, the recommendation submodule 4036 includes:

[0134] The ranking index acquisition unit is used to acquire the ranking index of the corresponding candidate merchant for each target product combination, and the ranking index includes at least one of historical sales volume and evaluation index.

[0135] The delivery complexity determination unit is used to determine the delivery complexity of the target product combination.

[0136] The recommendation unit is used to recommend the target product combination to the target user in the order of the corresponding ranking index, the corresponding reference score, and the delivery complexity, wherein the reference score is the reference score of the reference type combination corresponding to the target product combination.

[0137] Optionally, in another embodiment of the present disclosure, the delivery complexity determination unit includes:

[0138] The delivery complexity determination subunit is used to determine the delivery complexity for each target commodity combination according to the distance between the location of the corresponding candidate merchant and the location of the target user.

[0139] In summary, the embodiments of the present disclosure provide a personalized recommendation device based on the third embodiment. In addition to the beneficial effects of the third embodiment, the combination of product types that the user has clicked or ordered can be directly used as a reference type combination from the historical behavior records, so that the reference type combination conforms to the user's preferences; the candidate type combination with a higher frequency of occurrence can also be selected as the reference type combination, so as to ensure that the reference type combination is more in line with the user's preferences; the reference products that do not meet the preset conditions can also be filtered out, so that the qualified high-quality products can be used as reference products; the products similar to the reference products can also be expanded to the reference products, so that the expanded reference products have higher accuracy; the products provided by merchants close to the user can also be used as candidate products, so as to facilitate the delivery of the target product combination and reduce the delivery cost; the target product combination with a higher ranking index, a higher reference score, and a lower delivery complexity (especially a shorter delivery distance) can also be preferentially recommended to the target user, which helps to improve the recommendation success rate.

[0140] Embodiment 4 is a device embodiment corresponding to Embodiment 2. The detailed description may refer to Embodiment 2 and will not be repeated here.

[0141] The embodiment of the present disclosure also provides an electronic device, referring to Figure 5 , including: a processor 501, a memory 502, and a computer program 5021 stored in the memory 502 and executable on the processor, wherein the processor 501 implements the personalized recommendation method of the aforementioned embodiment when executing the program.

[0142] An embodiment of the present disclosure further provides a readable storage medium. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the personalized recommendation method of the aforementioned embodiment.

[0143] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0144] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems may also be used together with the teachings based thereon. According to the above description, it is apparent that the structure required for constructing such systems is. In addition, the embodiments of the present disclosure are not directed to any particular programming language either. It should be understood that the contents of the embodiments of the present disclosure described herein may be realized using various programming languages, and the description of the specific languages ​​above is intended to disclose the best mode of implementation of the embodiments of the present disclosure.

[0145] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the embodiments of the present disclosure can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0146] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present disclosure, the various features of the embodiments of the present disclosure are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following intention: that the claimed embodiments of the present disclosure require more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself as a separate embodiment of the embodiments of the present disclosure.

[0147] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0148] The various component embodiments of the embodiments of the present disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components of the personalized recommendation device according to the embodiments of the present disclosure. The embodiments of the present disclosure may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program implementing an embodiment of the present disclosure may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0149] It should be noted that the above embodiments illustrate rather than limit the embodiments of the present disclosure, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets should not be constructed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The embodiments of the present disclosure may be implemented by means of hardware including several different elements and by means of appropriately programmed computers. In a unit claim that lists several devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names.

[0150] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0151] The above description is only a preferred embodiment of the embodiments of the present disclosure and is not intended to limit the embodiments of the present disclosure. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the embodiments of the present disclosure should be included in the protection scope of the embodiments of the present disclosure.

[0152] The above is only a specific implementation of the embodiment of the present disclosure, but the protection scope of the embodiment of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed by the embodiment of the present disclosure, which should be included in the protection scope of the embodiment of the present disclosure. Therefore, the protection scope of the embodiment of the present disclosure should be based on the protection scope of the claims.

Claims

1. A personalized recommendation method, characterized in that: The method comprises: Combining the commodities contained in the historical behavior record of the target user into a reference commodity combination corresponding to the target user according to the reference type combination, including: Acquire, from the historical behavior record of the target user, the combination of commodities and commodity types on which the target user performs preset operations within a preset historical time, and obtain reference commodities and reference type combinations, wherein the preset operations at least include: an ordering operation and a clicking operation; Combining the reference commodities according to the reference type combination to obtain a reference commodity combination; The step of obtaining, from the historical behavior record of the target user, a combination of commodity types on which the target user has performed a preset operation within a preset historical time period includes: Obtain the product combination that the target user has performed preset operations on within a preset historical time; Clustering the commodity combinations according to the types of each commodity to obtain candidate type combinations; For each candidate type combination, a reference score of the candidate type combination is calculated according to the number of commodity combinations included and the total number of commodity combinations included in each candidate type combination; Arrange the candidate type combinations in descending order according to the reference scores, and select one or more candidate type combinations with the highest ranking to obtain a reference type combination; for the reference commodity combination, determine the raw materials of each reference commodity included in the reference commodity combination to obtain a raw material combination corresponding to the reference commodity combination; The candidate commodities are combined into a target commodity combination that matches the raw material combination, and the target commodity combination is recommended to the target user.

2. The method according to claim 1, characterized in that Before the step of combining the reference commodities according to the reference type combination to obtain a reference commodity combination, the method further includes: Filtering the reference commodities according to preset conditions, wherein the preset conditions include: at least one of a limiting condition for price and a limiting condition for commodity type; The reference commodities after filtering are expanded using a preset commodity set.

3. The method according to claim 2, characterized in that The step of using a preset commodity set to expand the reference commodities after filtering includes: For a preset commodity set, calculating a correlation coefficient between each commodity and the reference commodity; The commodities are arranged in descending order according to the correlation coefficients, and one or more commodities with the highest ranking are selected as extended reference commodities.

4. The method according to any one of claims 1 to 3, characterized in that The step of combining candidate commodities into a target commodity combination that matches the raw material combination and recommending the target commodity combination to the target user includes: For each raw material in each of the raw material combinations, selecting a candidate commodity that matches the raw material from the candidate commodities; For each raw material in each of the raw material combinations, select a candidate commodity from candidate commodities that match the raw material and add it to the target commodity combination corresponding to the raw material combination; The target product combination is recommended to the target user.

5. The method according to claim 4, characterized in that The candidate products are obtained through the following steps: Obtaining the location of the target user; Acquire merchants whose distance from the location is less than a preset distance threshold to obtain candidate merchants; The products provided by the candidate merchants are obtained to obtain candidate products.

6. The method according to claim 5, characterized in that The step of recommending the target product combination to the target user includes: For each target product combination, obtain a ranking index of the corresponding candidate merchants, where the ranking index includes at least one of historical sales volume and evaluation index; Determining the delivery complexity of the target product combination; The target product combination is recommended to the target user in the order of the corresponding ranking index, the corresponding reference score, and the delivery complexity, wherein the reference score is the reference score of the reference type combination corresponding to the target product combination.

7. The method according to claim 6, characterized in that The step of determining the delivery complexity of the target product combination includes: For each target product combination, the delivery complexity is determined according to the distance between the location of the corresponding candidate merchant and the location of the target user.

8. A personalized recommendation device, characterized in that: The device comprises: The reference product combination generation module is used to combine the products included in the historical behavior record of the target user into a reference product combination corresponding to the target user according to the reference type combination, including: Acquire, from the historical behavior record of the target user, the combination of commodities and commodity types on which the target user performs preset operations within a preset historical time, and obtain reference commodities and reference type combinations, wherein the preset operations at least include: an ordering operation and a clicking operation; Combining the reference commodities according to the reference type combination to obtain a reference commodity combination; The step of obtaining, from the historical behavior record of the target user, a combination of commodity types on which the target user has performed a preset operation within a preset historical time period includes: Obtain the product combination that the target user has performed preset operations on within a preset historical time; Clustering the commodity combinations according to the types of each commodity to obtain candidate type combinations; For each candidate type combination, a reference score of the candidate type combination is calculated according to the number of commodity combinations included and the total number of commodity combinations included in each candidate type combination; Arrange the candidate type combinations in descending order according to the reference scores, and select one or more candidate type combinations with the highest ranking to obtain a reference type combination; A raw material combination generating module, for determining the raw materials of each reference commodity contained in the reference commodity combination, and obtaining a raw material combination corresponding to the reference commodity combination; The recommendation module is used to combine candidate commodities into a target commodity combination that meets the raw material combination, and recommend the target commodity combination to the target user.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the personalized recommendation method as described in one or more of claims 1-7 is implemented.

10. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the personalized recommendation method as described in one or more of method claims 1-7.

Citation Information

Patent Citations

  • Goods information feedback method and system

    CN103810612A

  • Method and device for recommending goods information

    CN104881807A

  • Commodity recommendation method and device

    CN105469263A

  • Information recommendation method and device

    CN107093124A

  • Commodity combination method and device, electronic equipment and readable storage medium

    CN109902852A