Recommendation method, device and electronic device

Abstract

The present application provides a recommendation method, device and electronic device, wherein a specific implementation of the method includes: for a target user, determining one or more target merchant sets from a plurality of merchant sets; wherein each of the merchant sets corresponds to a plurality of preset consumer item combinations; the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set; determining a target score for each target consumer item combination corresponding to the target merchant set for the target user; and selecting a portion of the target consumer item combinations for recommendation based on the target score. This implementation enables the recommended content to be expanded to a combination of consumer items from a plurality of merchants, thereby satisfying more user needs, expanding the business of the instant delivery service, and improving the resource utilization of the instant delivery service.

Description

Technical Field

[0001] The present application relates to the field of Internet application technology, and in particular to a recommendation method, device and electronic device. Background Art

[0002] With the continuous development of Internet technology, many Internet-based services have emerged, making people's lives more convenient. Take instant delivery services as an example, people can buy a variety of goods through the client. At present, in related technologies, in order to enable users to find satisfactory goods more quickly, some merchants or items that may be of interest to users are usually recommended. However, this recommendation method is limited to independent merchants and items, so it has certain limitations and is difficult to meet the needs of users. Summary of the invention

[0003] In order to solve one of the above technical problems, the present application provides a recommended method, device and electronic device.

[0004] According to a first aspect of an embodiment of the present application, a recommendation method is provided, including:

[0005] For the target user, one or more target merchant sets from a plurality of merchant sets are determined; wherein each of the merchant sets corresponds to a plurality of preset consumer item combinations; and the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set;

[0006] Determine a target score for each target consumption item combination corresponding to the target merchant set for the target user;

[0007] Based on the target scores, a partial combination of target consumption items is selected for recommendation.

[0008] Optionally, the multiple merchant sets are determined based on a geohash algorithm.

[0009] Optionally, determining a target score for each target consumption item combination corresponding to the target merchant set for the target user includes:

[0010] Determine the historical business data of the target user; the historical business data includes historical browsing data and / or historical consumption data;

[0011] The target score is determined based on the historical business data.

[0012] Optionally, determining the target score based on the historical business data includes:

[0013] Determine, based on the historical business data, a weight vector for each target consumption item corresponding to the target merchant set for the target user;

[0014] Determine a characteristic identifier of each of the target consumption items, wherein the characteristic identifier of any of the target consumption items represents characteristic attribute information of the consumption item;

[0015] The target score is determined based on a weight vector of each target consumption item corresponding to the target merchant set for the target user and a feature identifier of each target consumption item.

[0016] Optionally, the historical business data includes historical browsing data and historical consumption data;

[0017] The determining, based on the historical business data, a weight vector for each target consumption item corresponding to the target merchant set for the target user includes:

[0018] Determine a feature preference vector corresponding to the target user based on the historical business data;

[0019] Determine an adjustment coefficient of the target user for each target consumption item; the adjustment coefficient is a weighted sum of the number of historical browsing times and the number of historical ordering times; wherein the number of historical browsing times is determined based on the historical browsing data, and the number of historical ordering times is determined based on the historical consumption data;

[0020] The weight vector is determined based on the feature preference vector and the adjustment coefficient.

[0021] Optionally, determining the target score based on a weight vector of each target consumption item corresponding to the target merchant set for the target user and a feature identifier of each target consumption item includes:

[0022] Calculate the hash value corresponding to the characteristic identifier of each target consumption item by using a preset hash algorithm; the number of bits of the hash value is the same as the number of dimensions of the weight vector;

[0023] For each of the target consumption items, the positive and negative attributes of the weight vector corresponding to the target consumption item are determined using the hash value corresponding to the characteristic identifier of the target consumption item to obtain a target vector;

[0024] For each target consumption item combination, based on the target vectors corresponding to the target consumption items constituting the target consumption item combination, determine a first code corresponding to the target consumption item combination;

[0025] Determine a second code; the second code is determined based on a combination of historical consumption items corresponding to the target user;

[0026] The target score is determined based on the second code and the first code corresponding to each combination of the target consumption items.

[0027] Optionally, the determining, based on the target vectors corresponding to the target consumption items constituting the target consumption item combination, the first code corresponding to the target consumption item combination includes:

[0028] Superimposing the target vectors corresponding to the target consumption items constituting the target consumption item combination to obtain a combination vector;

[0029] A first code is generated based on the positive and negative attributes of each dimension in the combined vector.

[0030] Optionally, for any target consumption item combination, the target score of the target consumption item combination for the target user is determined based on the second code and the first code corresponding to the target consumption item combination in the following manner:

[0031] Determine the Hamming distance between the second code and the first code corresponding to the target consumption item combination as the initial score corresponding to the target consumption item combination;

[0032] Determine the current business data of the target user and preset correction parameters;

[0033] According to the current business data, the initial score is corrected using the correction parameter to obtain the target score corresponding to the target consumption item combination.

[0034] According to a second aspect of an embodiment of the present application, a recommendation device is provided, including:

[0035] A first determination module is used to determine one or more target merchant sets from a plurality of merchant sets for a target user; wherein each of the merchant sets corresponds to a plurality of preset consumer item combinations; and a consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set;

[0036] A second determination module is used to determine a target score for each target consumption item combination corresponding to the target merchant set for the target user;

[0037] The recommendation module is used to select some target consumption item combinations for recommendation according to the target scores.

[0038] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the first aspects is implemented.

[0039] According to a fourth aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of the first aspects is implemented.

[0040] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0041] The recommendation method and device provided by the embodiment of the present application, by determining one or more target merchant sets from multiple merchant sets for the target user, wherein each merchant set corresponds to multiple preset consumer item combinations, and the consumer item combination corresponding to any merchant set is composed of multiple consumer items corresponding to the merchant set, then determining the target score for each target consumer item combination corresponding to the target merchant set for the target user, and selecting some target consumer item combinations for recommendation based on the above target score. Thus, the recommended content can be expanded to the combination of consumer items of multiple merchants, meeting more user needs, expanding the business of instant delivery services, and improving the resource utilization of instant delivery services.

[0042] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0044] Figure 1 is a flow chart of a recommendation method shown in the present application according to an exemplary embodiment;

[0045] Figure 2 is a flow chart of another recommendation method according to an exemplary embodiment of the present application;

[0046] Figure 3 is a flow chart of another recommendation method according to an exemplary embodiment of the present application;

[0047] Figure 4 is a block diagram of a recommendation device according to an exemplary embodiment of the present application;

[0048] Figure 5 is a block diagram of another recommendation device according to an exemplary embodiment of the present application;

[0049] Figure 6 is a block diagram of another recommendation device according to an exemplary embodiment of the present application;

[0050] Figure 7is a block diagram of another recommendation device according to an exemplary embodiment of the present application;

[0051] Figure 8 It is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0052] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0053] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0054] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0055] like Figure 1 As shown, Figure 1 1 is a flowchart of a recommendation method according to an exemplary embodiment, which can be applied to a server. The method includes the following steps:

[0056] In step 101, one or more target merchant sets from a plurality of merchant sets are determined for a target user.

[0057] In the present embodiment, the service involved may be an instant delivery service, and the merchant involved may be a merchant involved in the instant delivery service. A preset algorithm may be used in advance to divide the merchants involved in the instant delivery service into multiple merchant sets according to geographical locations. Optionally, multiple merchant sets may be determined based on the geohash geographic hashing algorithm. Specifically, the geohash algorithm may be used to number each merchant involved in the instant delivery service, and then the merchants with the same first N digits may be divided into a merchant set (N is a preset value), thereby obtaining multiple merchant sets. Since the merchant set is determined based on the geohash algorithm, the merchants in each merchant set are located similarly, which is more suitable for matching goods in the merchant set. It can be understood that any other reasonable algorithm may also be used to divide multiple merchant sets, and this application is not limited to this aspect.

[0058] In this embodiment, each merchant set may correspond to multiple consumer items, and the consumer items may be goods sold by the merchant (for example, the consumer items may be dishes in a restaurant, desserts in a dessert shop, or items in a supermarket, etc.). For each merchant set, some or all of the consumer items may be selected from the consumer items corresponding to the merchant set in advance, and the selected consumer items may be arbitrarily and reasonably matched based on a preset combination rule to obtain a combination of multiple consumer items corresponding to the merchant set. For example, merchant set A includes merchants a, merchants b, and merchants c. The consumer item m corresponding to merchant a may be matched with the consumer item n corresponding to merchant b to obtain a consumer item combination [m, n], and the consumer item s corresponding to merchant a, the consumer item t corresponding to merchant b, and the consumer item g corresponding to merchant c may be matched to obtain a consumer item combination [s, t, g], etc. Multiple consumer item combinations corresponding to each merchant set may be stored.

[0059] In this embodiment, the target user may be a user who needs to request an instant delivery service, and the target merchant set may be a merchant set that is more suitable for the target user to select from among multiple merchant sets. For example, the target merchant set may be a merchant set that is closer to the target user, or a merchant set that is selected more frequently by the target user, etc. Specifically, the merchant set whose center is closest to the target user may be selected as the target merchant set, or the merchant set whose center is less than a preset distance from the target user may be selected as the target merchant set, or the merchant set whose number of selections is greater than a preset number may be selected as the target merchant set based on the historical data of the target user, etc. It is understood that the target merchant set may also be determined by any other reasonable method, and this application is not limited to this aspect.

[0060] In step 102, a target score for each target consumption item combination corresponding to the target merchant set for the target user is determined.

[0061] In this embodiment, after determining the target merchant set, the target score for each target consumption item combination corresponding to the target merchant set for the target user can be further determined. For any target consumption item combination, the target score for the target user of the target consumption item combination can be used to indicate the degree of match between the target consumption item combination and the target user. For example, the larger the target score, the higher the degree of match between the target consumption item combination and the target user. Specifically, in one implementation, the historical business data of the target user can be determined, and the historical business data may include historical browsing data and / or historical consumption data. Then, a plurality of target consumption item combinations corresponding to the target merchant set are obtained from the pre-stored data, and the target score for each target consumption item combination corresponding to the target merchant set for the target user is determined based on the historical business data.

[0062] In another implementation, the target user's pre-entered label information of product preferences (such as favorite flavors, disliked flavors, favorite cuisines, favorite snacks, etc.) can also be obtained, and the target score for each target consumption item combination corresponding to the target merchant set for the target user is determined based on the target user's label information of product preferences. It is understood that the target score can also be determined by any other reasonable method, and the present application is not limited to this aspect.

[0063] In step 103, a partial combination of target consumption items is selected for recommendation according to the above target scores.

[0064] In this embodiment, a preset number of target consumption item combinations may be selected in descending order of target scores, and the selected target consumption items may be recommended to the target user for selection by the target user.

[0065] The recommendation method provided by the above embodiment of the present application determines one or more target merchant sets from multiple merchant sets for the target user, wherein each merchant set corresponds to multiple preset consumer item combinations, and the consumer item combination corresponding to any merchant set is composed of multiple consumer items corresponding to the merchant set, and then determines the target score for each target consumer item combination corresponding to the target merchant set for the target user, and selects some target consumer item combinations for recommendation based on the above target score. Thus, the recommended content can be expanded to the combination of consumer items of multiple merchants, which meets more user needs, expands the business of instant delivery services, and improves the resource utilization of instant delivery services.

[0066] like Figure 2 As shown, Figure 2 According to a flowchart of another recommendation method shown in an exemplary embodiment, the embodiment describes a process of determining a target score. The method can be applied in a server and includes the following steps:

[0067] In step 201, one or more target merchant sets from a plurality of merchant sets are determined for a target user.

[0068] In this embodiment, each merchant set corresponds to a plurality of preset consumer item combinations, and the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set.

[0069] In step 202, historical service data of the target user is determined, where the historical service data includes historical browsing data and / or historical consumption data.

[0070] In this embodiment, the historical business data of the target user may be determined, and the historical business data may be the business data of the target user in the instant delivery service. For example, the historical business data may include, but is not limited to, the historical browsing data, historical consumption data, and historical search data of the target user. It is understood that the historical business data may also include any other reasonable data, and the present application does not limit the specific content of the historical business data.

[0071] In step 203, a target score for each target consumption item combination corresponding to the target merchant set for the target user is determined based on the historical business data.

[0072] In this embodiment, the target score for the target user for each target consumption item combination corresponding to the target merchant set can be determined based on the historical business data. Specifically, in one implementation, the weight vector for the target user for each target consumption item corresponding to the target merchant set can be determined based on the historical business data. Then, the feature identifier of each target consumption item is determined, wherein the feature identifier of any target consumption item represents the feature information of the consumption item. Finally, based on the weight vector for the target user for each target consumption item corresponding to the target merchant set and the feature identifier of each target consumption item, the target score for the target user for each target consumption item combination corresponding to the target merchant set is determined.

[0073] In another implementation method, the number of times the target user browses and places an order for each target consumption item corresponding to the target merchant set can also be determined based on the historical business data, and the weight of each target consumption item combination corresponding to the target merchant set for the target user can be determined based on the number, and the target score for each target consumption item combination corresponding to the target merchant set for the target user can be determined based on the weight.

[0074] In step 204, a partial combination of target consumption items is selected for recommendation based on the target scores.

[0075] It should be noted that Figure 1The same steps as in the embodiment, in the above Figure 2 The embodiments will not be described in detail. For related contents, please refer to Figure 1 Example.

[0076] The recommendation method provided by the above-mentioned embodiment of the present application determines one or more target merchant sets from multiple merchant sets for the target user, determines the historical business data of the target user, and the historical business data includes historical browsing data and / or historical consumption data. Based on the historical business data, the target score for each target consumption item combination corresponding to the target merchant set for the target user is determined, and some target consumption item combinations are selected for recommendation based on the above-mentioned target scores. Since this embodiment refers to the historical business data of the target user to determine the target score for each target consumption item combination corresponding to the target merchant set for the target user, and the historical business data includes historical browsing data and / or historical consumption data, the rationality of the target score is improved, making the recommendation of the target consumption item combination more accurate.

[0077] like Figure 3 As shown, Figure 3 According to a flowchart of another recommendation method shown in an exemplary embodiment, the embodiment describes in detail the process of determining the target score based on historical business data. The method can be applied to a server and includes the following steps:

[0078] In step 301, one or more target merchant sets from a plurality of merchant sets are determined for a target user.

[0079] In this embodiment, each merchant set corresponds to a plurality of preset consumer item combinations, and the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set.

[0080] In step 302, historical business data of the target user is determined, where the historical business data includes historical browsing data and historical consumption data.

[0081] In step 303, a weight vector for each target consumption item corresponding to the target merchant set for the target user is determined based on the historical business data.

[0082] In this embodiment, the weight vector of each target consumption item corresponding to the target merchant set for the target user can be determined based on the historical business data, wherein each target consumption item corresponding to the target merchant set can be each consumption item corresponding to each merchant in the target merchant set.

[0083] In this embodiment, first, the characteristic preference vector corresponding to the target user can be determined based on the historical business data. Specifically, the dimensions of multiple vectors can be pre-set, and each dimension of the vector can represent a characteristic attribute of the consumption item. For example, the characteristic attributes represented by the dimension of the vector may include but are not limited to sweet, salty, spicy, sour, spicy, Sichuan cuisine, Hunan cuisine, Shandong cuisine, hot pot, buffet, Japanese cuisine, French cuisine, etc. Then, the weight value of the characteristic attribute represented by each dimension of the vector for the target user can be determined based on the historical business data, and the characteristic preference vector corresponding to the target user can be constructed based on the weight value.

[0084] For example, if the characteristic attributes represented by the dimensions of the vector include sweet, spicy, sour, Sichuan cuisine, Hunan cuisine, Shandong cuisine, hot pot, and buffet, and the target user's browsing and ordering times for the above characteristic attributes are 2, 4, 1, 4, 2, 2, 4, 1 respectively, then after normalization, it can be determined that the weight values ​​of the target user's characteristic attributes represented by each dimension of the vector are 0.1, 0.2, 0.05, 0.2, 0.1, 0.1, 0.2, 0.05 respectively, and the characteristic preference vector corresponding to the target user can be constructed based on the above weight values.

[0085] Next, the adjustment coefficient of the target user for each target consumption item is determined, and the adjustment coefficient may be a weighted sum of the historical browsing times and the historical order times, the historical browsing times may be determined based on the historical browsing data, and the historical order times may be determined based on the historical consumption data. The first weight corresponding to the historical browsing times and the second weight corresponding to the historical order times are preset weight values, and the first weight is less than the second weight.

[0086] Finally, the weight vector of each target consumption item corresponding to the target merchant set for the target user can be determined based on the characteristic preference vector corresponding to the target user and the adjustment coefficient. Specifically, for any target consumption item, the adjustment coefficient of the target user for the target consumption item can be multiplied by the characteristic preference vector corresponding to the target user to obtain the weight vector of the target consumption item for the target user.

[0087] In step 304, a characteristic identifier of each target consumption item is determined. The characteristic identifier of any target consumption item represents characteristic attribute information of the consumption item.

[0088] In this embodiment, the corresponding feature identifier can be set in advance according to the feature attributes of each target consumption item. For example, the feature attributes may include but are not limited to sweet, salty, spicy, sour, spicy, Sichuan cuisine, Hunan cuisine, Shandong cuisine, hot pot, buffet, Japanese cuisine, French cuisine, etc. It should be noted that the above-mentioned feature attributes must be consistent with the feature attributes represented by the dimension of the feature preference vector in step 303, and the order of the same feature attributes corresponds. Specifically, for each target consumption item, the weight of each feature attribute of the target consumption item can be determined, and the weight of each feature attribute can be arranged in order to obtain the feature identifier corresponding to the target consumption item.

[0089] In step 305, based on the weight vector of each target consumption item corresponding to the target merchant set for the target user and the feature identifier of each target consumption item, the target score of each target consumption item combination corresponding to the target merchant set for the target user is determined.

[0090] In this embodiment, the target score for each target consumption item combination corresponding to the target merchant set for the target user can be determined based on the weight vector for each target consumption item corresponding to the target merchant set and the feature identifier of each target consumption item.

[0091] In this embodiment, first, a hash value corresponding to the characteristic identifier of each target consumption item can be calculated by a preset hash algorithm, and the hash value can be a binary number, and each digit of the binary number corresponds to a characteristic attribute. The number of digits of the hash value is the same as the number of dimensions of the weight vector of each target consumption item for the target user. Among them, the preset hash algorithm can be any reasonable hash algorithm, and any hash algorithm known in the art and that may appear in the future can be applied to the present application, and the present application is not limited in this respect.

[0092] Next, for each target consumption item, the hash value corresponding to the characteristic identifier of the target consumption item can be used to determine the positive and negative attributes of the weight vector corresponding to the target consumption item, and obtain the target vector. Specifically, since the number of bits of the hash value is the same as the number of dimensions of the weight vector of the target consumption item for the target user, the value corresponding to each dimension of the weight vector in the hash value can be determined. If the value is 1, the corresponding dimension is a positive attribute, and if the value is 0, the corresponding dimension is a negative attribute, thereby obtaining the target vector.

[0093] For example, if the hash value corresponding to the characteristic identifier of the target consumption item is 1001, and the weight vector corresponding to the target consumption item is (a, b, c, d), then the target vector (a, -b, -c, d) can be obtained.

[0094] Then, for each target consumption item combination, based on the target vectors corresponding to the target consumption items constituting the target consumption item combination, the first code corresponding to the target consumption item combination is determined, and the first code is a binary number. Specifically, the target vectors corresponding to the target consumption items constituting the target consumption item combination can be superimposed to obtain a combination vector, and the first code is generated based on the positive and negative attributes of each dimension in the combination vector. For any dimension in the combination vector, if it is positive, it corresponds to the first code 1, and if it is negative, it corresponds to the first code 0.

[0095] For example, if the combination vector corresponding to the target consumption item combination is (-m, n, -j, k), then the first code corresponding to the target consumption item combination may be 0101.

[0096] Next, determine the second code, which is determined based on the combination of historical consumption items corresponding to the target user, and the second code is also a binary number. Specifically, the combination of historical consumption items corresponding to the target user and the historical consumption items that constitute the combination of historical consumption items can be determined in advance. Then, based on the historical business data of the target user, determine the weight vector of each historical consumption item for the target user and the feature identifier of each historical consumption item, determine the target vector corresponding to each historical consumption item, and obtain the combination vector corresponding to each combination of historical consumption items. Superimpose the combination vectors corresponding to all combinations of historical consumption items, and generate the second code based on the positive and negative attributes of each dimension in the superimposed vector (refer to the generation method of the first code).

[0097] Finally, the target score is determined based on the second code and the first code corresponding to each target consumption item combination. Specifically, for any target consumption item combination, the target score of the target consumption item combination for the target user can be determined based on the second code and the first code corresponding to the target consumption item combination in the following manner: First, determine the Hamming distance between the second code and the first code corresponding to the target consumption item combination as the initial score corresponding to the target consumption item combination. Then, determine the current business data of the target user and the preset correction parameters. Finally, according to the current business data, the initial score is corrected using the correction parameters to obtain the target score corresponding to the target consumption item combination. Specifically, the current consumption item browsed by the target user within the preset time period before the current time can be determined, and the corresponding initial score of the target consumption item combination containing the current consumption item is multiplied by the preset correction parameter to obtain the target score corresponding to the target consumption item combination.

[0098] In step 306, a portion of target consumption item combinations are selected for recommendation based on the target scores.

[0099] It should be noted that Figure 1and Figure 2 The same steps as in the embodiment, in the above Figure 3 The embodiments will not be described in detail. For related contents, please refer to Figure 1 and Figure 2 Example.

[0100] The recommendation method provided by the above embodiment of the present application determines the weight vector of each target consumption item corresponding to the target merchant set for the target user based on historical business data, determines the characteristic identification of each target consumption item, and the characteristic identification of any target consumption item represents the characteristic attribute information of the consumption item, and determines the target score of each target consumption item combination corresponding to the target merchant set for the target user based on the weight vector of each target consumption item corresponding to the target merchant set and the characteristic identification of each target consumption item. Thereby, the rationality of the target score is further improved, making the recommendation of the target consumption item combination more accurate.

[0101] It should be noted that although the operations of the method of the present application are described in a specific order in the above embodiments, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted in the flow chart can be changed in the order of execution. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step for execution, and / or one step can be decomposed into multiple steps for execution.

[0102] Corresponding to the aforementioned recommended method embodiment, the present application also provides an embodiment of a recommended device.

[0103] like Figure 4 As shown, Figure 4 4 is a block diagram of a recommendation device according to an exemplary embodiment of the present application. The device may include: a first determination module 401, a second determination module 402 and a recommendation module 403.

[0104] Among them, the first determination module 401 is used to determine one or more target merchant sets from multiple merchant sets for the target user, wherein each merchant set corresponds to multiple preset consumption item combinations, and the consumption item combination corresponding to any merchant set is composed of multiple consumption items corresponding to the merchant set.

[0105] The second determination module 402 is used to determine the target score of each target consumption item combination corresponding to the target merchant set for the target user.

[0106] The recommendation module 403 is used to select some target consumption item combinations for recommendation according to the target scores.

[0107] In some optional implementations, the multiple merchant sets are determined based on a geohash algorithm.

[0108] like Figure 5 As shown, Figure 5 is another recommended device block diagram according to an exemplary embodiment of the present application. Figure 4 On the basis of the illustrated embodiment, the second determining module 402 may include: a first determining submodule 501 and a second determining submodule 502 .

[0109] The first determination submodule 501 is used to determine the historical business data of the target user, where the historical business data includes historical browsing data and / or historical consumption data.

[0110] The second determination submodule 502 is used to determine the target score for the target user for each target consumption item combination corresponding to the target merchant set based on the historical business data.

[0111] like Figure 6 As shown, Figure 6 is another recommended device block diagram according to an exemplary embodiment of the present application. Figure 5 On the basis of the illustrated embodiment, the second determination submodule 502 may include: a vector determination submodule 601 , an information determination submodule 602 and a score determination submodule 603 .

[0112] The vector determination submodule 601 is used to determine the weight vector of each target consumption item corresponding to the target merchant set for the target user based on the above historical business data.

[0113] The information determination submodule 602 is used to determine a characteristic identifier of each target consumption item. The characteristic identifier of any target consumption item represents characteristic attribute information of the consumption item.

[0114] The score determination submodule 603 is used to determine the target score of each target consumption item combination corresponding to the target merchant set for the target user based on the weight vector of each target consumption item corresponding to the target merchant set for the target user and the feature identifier of each target consumption item.

[0115] In some other optional implementations, the historical business data may include historical browsing data and historical consumption data. The vector determination submodule 601 is configured to: determine the characteristic preference vector corresponding to the target user based on the above historical business data, determine the adjustment coefficient of the target user for each target consumption item, the adjustment coefficient is the weighted sum of the historical browsing times and the historical order times, and determine the weight vector based on the above characteristic preference vector and the above adjustment coefficient. The historical browsing times are determined based on the historical browsing data, and the historical order times are determined based on the historical consumption data.

[0116] like Figure 7 As shown, Figure 7 is another recommended device block diagram according to an exemplary embodiment of the present application. Figure 6 Based on the illustrated embodiment, the score determination submodule 603 may include: a calculation submodule 701 , a first acquisition submodule 702 , a first encoding submodule 703 , a second encoding submodule 704 and a second acquisition submodule 705 .

[0117] The calculation submodule 701 is used to calculate the hash value corresponding to the characteristic identifier of each target consumption item through a preset hash algorithm, and the number of bits of the hash value is the same as the number of dimensions of the weight vector.

[0118] The first acquisition submodule 702 is used to determine, for each target consumption item, the positive and negative attributes of the weight vector corresponding to the target consumption item using the hash value corresponding to the feature identifier of the target consumption item, so as to obtain a target vector.

[0119] The first encoding submodule 703 is used to determine, for each target consumption item combination, a first code corresponding to the target consumption item combination based on the target vector corresponding to the target consumption items constituting the target consumption item combination.

[0120] The second coding submodule 704 is used to determine a second code, where the second code is determined based on a combination of historical consumption items corresponding to the target user.

[0121] The second acquisition submodule 705 is used to obtain the target score for the target user for each target consumption item combination corresponding to the first coded target merchant set corresponding to each target consumption item combination based on the second code and the target score for the target user for each target consumption item combination.

[0122] In other optional implementations, the first encoding submodule 703 is configured to: superimpose the target vectors corresponding to the target consumption items constituting the target consumption item combination to obtain a combination vector, and generate a first code based on the positive and negative attributes of each dimension in the combination vector.

[0123] In other optional implementations, for any target consumption item combination, the second acquisition submodule can determine the target score of the target consumption item combination for the target user based on the second code and the first code corresponding to the target consumption item combination in the following manner: determine the Hamming distance between the second code and the first code corresponding to the target consumption item combination as the initial score corresponding to the target consumption item combination. Determine the current business data of the target user and the preset correction parameters, and according to the current business data, use the correction parameters to correct the above initial score to obtain the target score corresponding to the target consumption item combination.

[0124] It should be understood that the above device can be pre-set in the server, or loaded into the server by downloading, etc. The corresponding modules in the above device can cooperate with the modules in the server to implement the recommendation solution.

[0125] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0126] The present application also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figures 1 to 3 Any one of the embodiments provides a recommended method.

[0127] Corresponding to the above-mentioned recommended method, the present application embodiment also proposes Figure 8 The schematic structural diagram of an electronic device according to an exemplary embodiment of the present application is shown in FIG. Figure 8 At the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a recommendation device at the logical level. Of course, in addition to software implementations, this application does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0128] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0129] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A recommendation method, characterized in that: The method comprises: For the target user, one or more target merchant sets from a plurality of merchant sets are determined; wherein each of the merchant sets corresponds to a plurality of preset consumer item combinations; and the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set; Determining a target score for each target consumption item combination corresponding to the target merchant set for the target user includes: Determine the historical business data of the target user; the historical business data includes historical browsing data and / or historical consumption data; Determining the target score based on the historical business data includes determining a weight vector for each target consumption item corresponding to the target merchant set for the target user based on the historical business data; Determine a characteristic identifier of each of the target consumption items, wherein the characteristic identifier of any of the target consumption items represents characteristic attribute information of the consumption item; Determining the target score based on the weight vector of each target consumption item corresponding to the target merchant set for the target user and the feature identifier of each target consumption item includes: Calculate the hash value corresponding to the characteristic identifier of each target consumption item by a preset hash algorithm; the number of bits of the hash value is the same as the number of dimensions of the weight vector; For each of the target consumption items, the positive and negative attributes of the weight vector corresponding to the target consumption item are determined using the hash value corresponding to the characteristic identifier of the target consumption item to obtain a target vector; For each target consumption item combination, based on the target vectors corresponding to the target consumption items constituting the target consumption item combination, determine a first code corresponding to the target consumption item combination; Determine a second code; the second code is determined based on a combination of historical consumption items corresponding to the target user; Determine the target score based on the second code and the first code corresponding to each combination of the target consumption items; Based on the target scores, a partial combination of target consumption items is selected for recommendation.

2. The method according to claim 1, characterized in that The plurality of merchant sets are determined based on a geohash algorithm.

3. The method according to claim 1, characterized in that The historical business data includes historical browsing data and historical consumption data; The determining, based on the historical business data, a weight vector for each target consumption item corresponding to the target merchant set for the target user includes: Determine a feature preference vector corresponding to the target user based on the historical business data; Determine an adjustment coefficient of the target user for each target consumption item; the adjustment coefficient is a weighted sum of the number of historical browsing times and the number of historical ordering times; wherein the number of historical browsing times is determined based on the historical browsing data, and the number of historical ordering times is determined based on the historical consumption data; The weight vector is determined based on the feature preference vector and the adjustment coefficient.

4. The method according to claim 1, characterized in that: The step of determining the first code corresponding to the target consumption item combination based on the target vector corresponding to the target consumption items constituting the target consumption item combination includes: Superimposing the target vectors corresponding to the target consumption items constituting the target consumption item combination to obtain a combination vector; A first code is generated based on the positive and negative attributes of each dimension in the combined vector.

5. The method according to claim 1, characterized in that For any target consumption item combination, the target score of the target consumption item combination for the target user is determined based on the second code and the first code corresponding to the target consumption item combination in the following manner: Determine the Hamming distance between the second code and the first code corresponding to the target consumption item combination as the initial score corresponding to the target consumption item combination; Determine the current business data of the target user and preset correction parameters; According to the current business data, the initial score is corrected using the correction parameter to obtain the target score corresponding to the target consumption item combination.

6. A recommendation device, characterized in that: The device comprises: A first determination module is used to determine one or more target merchant sets from a plurality of merchant sets for a target user; Each of the merchant sets corresponds to a plurality of preset consumer item combinations; the consumer item combination corresponding to any merchant set is composed of a plurality of consumer items corresponding to the merchant set; The second determination module is used to determine the target score of each target consumption item combination corresponding to the target merchant set for the target user, including: Determine the historical business data of the target user; the historical business data includes historical browsing data and / or historical consumption data; Determining the target score based on the historical business data includes determining a weight vector for each target consumption item corresponding to the target merchant set for the target user based on the historical business data; Determine a characteristic identifier of each of the target consumption items, wherein the characteristic identifier of any of the target consumption items represents characteristic attribute information of the consumption item; Determining the target score based on the weight vector of each target consumption item corresponding to the target merchant set for the target user and the feature identifier of each target consumption item includes: Calculate the hash value corresponding to the characteristic identifier of each target consumption item by a preset hash algorithm; the number of bits of the hash value is the same as the number of dimensions of the weight vector; For each of the target consumption items, the positive and negative attributes of the weight vector corresponding to the target consumption item are determined using the hash value corresponding to the characteristic identifier of the target consumption item to obtain a target vector; For each target consumption item combination, based on the target vectors corresponding to the target consumption items constituting the target consumption item combination, determine a first code corresponding to the target consumption item combination; Determine a second code; the second code is determined based on a combination of historical consumption items corresponding to the target user; Determine the target score based on the second code and the first code corresponding to each combination of the target consumption items; The recommendation module is used to select some target consumption item combinations for recommendation according to the target scores.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

Images