An object recommendation method and apparatus, an electronic device, and a storage medium

Abstract

The application discloses an object recommendation method and device, electronic equipment and storage medium, and is used for solving the technical problem that a shopping APP cannot guarantee the activity and stickiness of a user. The application comprises the following steps: in response to a recommendation instruction triggered by a user, it is judged whether the recommendation instruction is triggered for the first time on the day; if the recommendation instruction is triggered for the first time on the day, the cumulative triggering number of times of the recommendation instruction is obtained; according to the cumulative triggering number of times, the user is recommended points; if the recommendation instruction is not triggered for the first time on the day, the triggering number of times of the recommendation instruction on the day is determined; and according to the triggering number of times on the day, the user is recommended goods.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of marketing recommendation, and in particular to an object recommendation method and device, electronic equipment and a storage medium. BACKGROUND

[0002] In actual life, when a user shops using an APP, there is a phenomenon of buying and leaving immediately, which greatly affects the user's activity and stickiness. In order to increase the user's activity and stickiness, a check-in mode can be used to attract the user to check in every day, thereby increasing the user's activity and stickiness.

[0003] However, a single check-in is boring, and if the user is not given enough rewards, it is difficult to make the user check in every day, and the user's activity cannot be guaranteed. Pure check-in cannot make the user develop the habit of shopping using the APP, resulting in insufficient stickiness of the user. SUMMARY

[0004] The present application provides an object recommendation method and device, electronic equipment and a storage medium, which are used to solve the technical problem that a shopping APP cannot guarantee the user's activity and stickiness.

[0005] The present application provides an object recommendation method, wherein the object includes points and goods; the method comprises:

[0006] In response to a recommendation instruction triggered by a user, it is determined whether the recommendation instruction is triggered for the first time on the day;

[0007] If the recommendation instruction is triggered for the first time on the day, the cumulative number of times of triggering the recommendation instruction is obtained;

[0008] According to the cumulative number of times of triggering, the user is recommended points;

[0009] If the recommendation instruction is not triggered for the first time on the day, the number of times of triggering the recommendation instruction on the day is determined;

[0010] According to the number of times of triggering on the day, the user is recommended goods.

[0011] Optionally, the step of determining whether the recommendation instruction is triggered for the first time on the day in response to the recommendation instruction triggered by the user comprises:

[0012] In response to a recommendation instruction triggered by a user, it is determined whether the recommendation instruction is triggered for the first time on the day;

[0013] If not, it is determined that the recommendation instruction is triggered for the first time on the day.

[0014] Optionally, the step of recommending points to the user according to the cumulative number of times of triggering comprises:

[0015] sequentially calculating remainders of the cumulative trigger number and at least one preset value;

[0016] screening the preset value with the remainder of 0 as a target value;

[0017] obtaining a first integral interval corresponding to the target value, and matching a first target integral in the first integral interval;

[0018] recommending the first target integral to the user.

[0019] Optionally, the method further comprises:

[0020] when all remainders are not 0, matching a second target integral in a preset second integral interval, and recommending the second target integral to the user.

[0021] Optionally, the step of obtaining the first integral interval corresponding to the target value, and matching the first target integral in the first integral interval comprises:

[0022] obtaining the first integral interval corresponding to the target value;

[0023] obtaining a priority order of each of the first integral intervals;

[0024] matching the first target integral in the first integral interval with the highest priority.

[0025] Optionally, the step of recommending the goods to the user according to the trigger number of the day comprises:

[0026] obtaining user behavior data, combination price goods data and ordinary goods data of the user;

[0027] generating a first recommendation order by using the user behavior data, the combination price goods data and the ordinary goods data;

[0028] determining a first target recommendation object in the first recommendation order according to the trigger number of the day;

[0029] screening recommended goods from the first target recommendation object, and recommending the recommended goods to the user.

[0030] Optionally, the step of screening the recommended goods from the first target recommendation object, and recommending the recommended goods to the user comprises:

[0031] when the first target recommendation object is user behavior data, screening shopping cart goods data, search goods data and browsing goods data from the user behavior data;

[0032] generate a second recommendation sequence according to the shopping cart commodity data, the search commodity data and the browsing commodity data;

[0033] obtain a current commodity recommendation times of the user behavior data;

[0034] determine a second target recommendation object in the second recommendation sequence according to the current commodity recommendation times;

[0035] screen a recommended commodity from the second target recommendation object and recommend the recommended commodity to the user.

[0036] The application further provides an object recommendation device, the object including points and commodities; the device includes:

[0037] a first-time triggering judgment module for judging whether a recommendation instruction triggered by a user is a first-time triggering of the day in response to the recommendation instruction;

[0038] a cumulative triggering times obtaining module for obtaining a cumulative triggering times of the recommendation instruction if the recommendation instruction is the first-time triggering of the day;

[0039] a point recommendation module for recommending points to the user according to the cumulative triggering times;

[0040] a daily triggering times determining module for determining a daily triggering times of the recommendation instruction if the recommendation instruction is not the first-time triggering of the day;

[0041] a commodity recommendation module for recommending commodities to the user according to the daily triggering times.

[0042] The application further provides an electronic device, the device including a processor and a memory:

[0043] the memory is used for storing program codes and transmitting the program codes to the processor;

[0044] the processor is used for executing the object recommendation method according to the instructions in the program codes.

[0045] The application further provides a computer readable storage medium, the computer readable storage medium is used for storing program codes, the program codes are used for executing the object recommendation method.

[0046] From the above technical solution can be seen, the present application has the following advantages: the present application determines to recommend the user with points or goods by judging whether the user triggered recommendation instruction is the first time triggered in the day. By recommending points to the user, the user can be prompted to log in daily to accumulate points to obtain rewards, thereby enhancing the user's activity. By recommending goods to the user, the user's browsing interest can be aroused, and the user's habit of browsing goods on the current platform can be developed, thereby enhancing the user's stickiness. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0048] Figure 1 A step flowchart of an object recommendation method provided by an embodiment of the present application.

[0049] Figure 2 A step flowchart of an object recommendation method provided by another embodiment of the present application.

[0050] Figure 3 A flowchart of a points recommendation method provided by an embodiment of the present application.

[0051] Figure 4 A structural block diagram of an object recommendation device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0052] The embodiments of the present application provide an object recommendation method, device, electronic equipment and storage medium, which are used to solve the technical problem that a shopping APP cannot guarantee the activity and stickiness of users.

[0053] In order to make the inventive purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the embodiments described below are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0054] Please refer to Figure 1 , Figure 1 A step flowchart of an object recommendation method provided by an embodiment of the present application.

[0055] The object recommendation method provided by the application can be applied to any mobile terminal, electronic device or computer device installed with a shopping APP or a shopping applet, and specifically can include the following steps.

[0056] In step 101, whether the recommendation instruction is triggered for the first time in the day is judged in response to the user-triggered recommendation instruction.

[0057] In the embodiment of the application, the user can trigger the recommendation instruction by clicking, pressing, gesture, shaking (such as "shake" function) and the like after entering the corresponding interface (such as clock-in, draw, recommendation and the like) of the shopping APP, so as to request the shopping APP to recommend an object to the user. In an example, the object recommended by the shopping APP to the user can include points and goods, and each recommendation can be one of points or goods.

[0058] It should be noted that points and goods are only an example of an object provided by the embodiment of the application, and other recommended objects can be selected by those skilled in the art according to actual conditions, including but not limited to coupons, red packets and the like.

[0059] In a specific implementation, whether the recommended object is points or goods can be determined by the number of times the recommendation instruction is triggered. For example, whether the recommendation instruction is triggered for the first time in the day is used as a basis for determination.

[0060] It should be noted that the above determination basis is only an example provided by the embodiment of the application, and does not constitute a constraint on the embodiment of the application, and those skilled in the art can select other numbers of times of triggering as a basis for determination according to actual conditions.

[0061] In step 102, if the recommendation instruction is triggered for the first time in the day, the cumulative number of times of triggering the recommendation instruction is obtained.

[0062] In the embodiment of the application, the cumulative number of times of triggering is the total number of times of triggering the recommendation instruction accumulated after the user registers the shopping APP.

[0063] In a specific implementation, in order to avoid the user accumulating points quickly by repeatedly triggering the recommendation instruction in a single day, resulting in a decrease in the value of the points, the number of times of triggering the point reward in a single day can be set, such as providing the user with a point reward only when the recommendation instruction is triggered for the first time in the day.

[0064] In a specific implementation, the value of the point reward can be determined according to a certain rule each time the point reward is triggered. For example, different upper and lower limits of the point reward are triggered at different cumulative numbers of times of triggering. When the user is tired of small amounts of points, a larger amount of point reward is provided to the user in a timely manner.

[0065] In step 103, points are recommended to the user according to the cumulative number of times of triggering.

[0066] In the embodiment of the present application, different values of points can be recommended to the user according to different cumulative triggering times. For example, small points are continuously inserted in a regular manner, so that the user's enthusiasm can be aroused when the user feels tired of the points.

[0067] In step 104, if the recommendation instruction is not triggered for the first time in the day, the triggering times of the recommendation instruction in the day are determined.

[0068] In step 105, the user is recommended the goods according to the triggering times in the day.

[0069] In the embodiment of the present application, in addition to the points, the user who triggers the recommendation instruction multiple times in a day can also be recommended the goods. According to the user's consumption, search, browsing habits and the like, the user can be recommended the goods of interest, so as to improve the user's stickiness.

[0070] In the specific implementation, the goods of interest of the user can be obtained from different dimensions for recommendation. The dimensions can include but are not limited to the behavior dimensions such as consumption, search and browsing, and can also include other non-behavior dimensions, and the records generated accordingly can all be used as the basis for the goods recommendation. In order to reduce the interference of the dimension data generated by the occasional behavior on the recommendation accuracy, in the embodiment of the present application, the user can be recommended the goods of different dimensions according to different triggering times in the day.

[0071] It should be noted that the recommendation of the goods to the user through the triggering times in the day is only an example of the embodiment of the present application, and the user can be recommended the goods through the triggering times in the month, the triggering times in the year and the like according to the actual needs. The present application does not make specific limitation thereto.

[0072] The present application determines whether the recommendation instruction triggered by the user is triggered for the first time in the day, and then decides whether to recommend the points or the goods to the user. Through the way of recommending the points to the user, the user can be prompted to log in every day to accumulate the points to obtain the rewards, so as to enhance the user's activity. Through the way of recommending the goods to the user, the user's browsing interest can be aroused, and the user's habit of browsing the goods on the current platform can be developed, so as to enhance the user's stickiness.

[0073] Please refer to Figure 2 , Figure 2 The step flowchart of the object recommendation method provided for another embodiment of the present application is further refined and supplemented on the basis of any of the above embodiments. Specifically, the following steps can be included:

[0074] In step 201, in response to the recommendation instruction triggered by the user, it is queried whether the cache information of the recommendation instruction exists in the preset database.

[0075] In step 202, if not, it is determined that the recommendation instruction is triggered for the first time in the day.

[0076] In the embodiment of the present application, whether the recommendation instruction is the first triggered recommendation instruction of the day can be determined by querying the preset database to determine whether the cache information of the recommendation instruction exists.

[0077] In an example, the database can be a Redis (Remote Dictionary Server) database. Redis is an open source Key-Value database written in ANSI C, supporting network, and can be based on memory or persistent log.

[0078] In step 203, if the recommendation instruction is the first triggered recommendation instruction of the day, the cumulative trigger number of the recommendation instruction is obtained.

[0079] In a specific implementation, in order to avoid the user triggering the recommendation instruction repeatedly in a single day to quickly accumulate points, resulting in the value of the points being reduced, the number of times of triggering the point reward in a single day can be set, such as providing the user with a point reward only when the recommendation instruction is triggered for the first time in the day.

[0080] In a specific implementation, the value of the point reward can be determined according to a certain rule each time the point reward is triggered. For example, different upper and lower limits of the point reward are triggered at different cumulative trigger numbers.

[0081] In step 204, the user is recommended points according to the cumulative trigger number.

[0082] In the embodiment of the present application, different values of points can be recommended to the user according to different cumulative trigger numbers. For example, a large value of points is regularly inserted in a continuous small amount of points, so as to mobilize the enthusiasm of the user when the user feels tired of the points.

[0083] In an example, step 204 can include the following sub-steps:

[0084] S41, the remainders of the cumulative trigger number and at least one preset value are calculated in turn;

[0085] S42, the preset value whose remainder is 0 is selected as the target value;

[0086] S43, the first point interval corresponding to the target value is obtained, and the first target point is matched in the first point interval.

[0087] In the embodiments of the present application, different integral intervals can be set to recommend different values of integral for different cumulative trigger times of the recommended instructions. For example, a maximum integral value is set and is divided into 5 first integral intervals, and the capacity of each first integral interval is the same. The 5 first integral intervals can be [1%-20%], [21%-40%], [41%-60%], [61-80%], and [81%-100%] of the maximum integral value, respectively.

[0088] Then, a preset value can be set as a selection basis for different integral intervals. When the remainder between the cumulative trigger time and the preset value is 0, that is, the cumulative trigger time is an integer multiple of the preset value, a random value of integral can be selected from the first integral interval corresponding to the preset value and recommended to the user. For example, if the selected value in [1%-20%] is 5, when the cumulative trigger time is an integer multiple of 5, a random value of integral in [1%-20%] of the maximum integral value is selected as the first target integral recommended to the user.

[0089] In an example, the selection of the preset value can be determined by referring to the churn rate of the shopping APP at different use times. Specifically as follows:

[0090] First, a batch of users who frequently use the shopping APP are selected as a sample set, and the maximum integral value is divided into multiple intervals; the interval with the smallest upper limit of integral is selected as the selection interval for integral recommendation.

[0091] Then, the churn rate of the user after each integral recommendation (i.e., the proportion of users who no longer continuously use the shopping APP after that in the sample set) is recorded.

[0092] After arranging the churn rates in descending order, the recommended times corresponding to the top several churn rates are selected as the preset values.

[0093] Finally, different intervals of the maximum integral value are selected as the first integral interval for different preset values according to the size of the churn rate. For example, the interval with the largest upper limit of integral is selected as the first integral interval for the preset value with the largest churn rate.

[0094] By recommending a higher integral before the user is about to massively churn, the interest of the user in signing in and clocking in through the shopping APP every day can be re-boosted, thereby improving the stickiness of the user in using the shopping APP.

[0095] Considering that there can be more than one preset value that is a common multiple of a cumulative trigger time, in the embodiments of the present application, the step of obtaining the first integral interval corresponding to the target value and matching the first target integral in the first integral interval can include:

[0096] S431, acquire the first integral interval corresponding to the target value;

[0097] S432, acquire the priority order of each first integral interval;

[0098] S433, match the first target integral in the first integral interval with the highest priority.

[0099] In the specific implementation, the target values can be sorted to obtain the priority order, in the case that the remainder of the cumulative trigger times and the plurality of preset values is 0, the first integral interval corresponding to the target value with the highest priority is taken as the target integral interval, and a random integral value in the target integral interval is taken as the first target integral.

[0100] S44, recommend the first target integral to the user;

[0101] S45, when all the remainders are not 0, match the second target integral in the preset second integral interval, and recommend the second target integral to the user.

[0102] In the embodiment of the application, a second integral interval can also be set, when all the preset values are not the factor of the cumulative trigger times, a random integral value in the second integral interval is taken as the second target integral, and the second target integral is recommended to the user.

[0103] For the convenience of understanding, the integral recommendation mode is described through a specific example as follows:

[0104] Please refer to Figure 3 , Figure 3 The flow chart of the integral recommendation mode provided by the embodiment of the application.

[0105] As shown in Figure 3 , when the user opens the shopping APP to enter the integral check-in page and triggers the "shake" operation, first, it is inquired whether the trigger information of the "shake" instruction is cached in the redis, if not, it is judged as the first trigger of the day, and the integral recommendation flow is entered:

[0106] First, three preset values are set, which are 17, 7 and 5, then it is inquired whether the "shake" times of the user is the multiple of at least one of the values 17, 7 and 5, if yes, the following enumerated integral logic is recommended, if not hit, the normal integral result is inquired and the cumulative trigger times is not cleared.

[0107] 1, the cumulative trigger times of the user is the multiple of 17, then a random integral value in the [90%-100%] interval of the maximum integral value is matched;

[0108] 2. If the cumulative number of triggers of the user is a multiple of 7, then randomly match any one of the integral values in the [60%-80%] interval of the maximum integral value (if it conflicts with the first rule, then follow this rule);

[0109] 3. If the cumulative number of triggers of the user is a multiple of 5, then randomly match any one of the integral values in the [30%-60%] interval of the maximum integral value (if it conflicts with the first or second rule, then follow this rule);

[0110] 4. If none of the above conditions are met, then randomly match any one of the integral values in the [1%-20%] interval of the maximum integral value.

[0111] Step 205, if the recommendation instruction is not triggered for the first time on the day, then determine the number of triggers of the recommendation instruction on the day;

[0112] Step 206, according to the number of triggers on the day, recommend goods to the user.

[0113] In the embodiments of the present application, in addition to the integral, the user who triggers the recommendation instruction multiple times a day can also be recommended goods, and according to the user's consumption, search, browsing habits, etc., the user is recommended goods of interest to improve the user's stickiness.

[0114] In one example, step 206 can include the following sub-steps:

[0115] S61, obtain user behavior data, combination price commodity data and ordinary commodity data of the user;

[0116] S62, generate a first recommendation order using the user behavior data, the combination price commodity data and the ordinary commodity data;

[0117] In a specific implementation, the first recommendation order can be determined in combination with the user behavior data, the combination price commodity data and the ordinary commodity data on the attractiveness of the user. For example, the user behavior data reflects the user's purchase tendency to some extent, so the user can be recommended the goods corresponding to the user behavior data first. The combination price commodity has an advantage in price compared to a single commodity, so it has greater attractiveness to the user than the ordinary commodity; therefore, the recommendation order of the combination price commodity data can be arranged after the user behavior data and before the ordinary commodity data, thereby obtaining the first recommendation order.

[0118] The above first recommendation order reflects the user's purchase tendency to some extent, and the user can be recommended the goods of interest to the user at the beginning, so that the user has high interest in using the shopping APP, and avoids the user losing interest in the shopping APP due to the recommended products not meeting the expectations at the beginning, resulting in a decrease in activity.

[0119] S63, determining the first target recommendation object in the first recommendation order according to the daily trigger number;

[0120] The first target recommendation object refers to one of the user behavior data, the combination price commodity data and the ordinary commodity data, and can be obtained by matching the daily trigger number and the first recommendation order.

[0121] For example, assuming that the first recommendation order is the user behavior data, the combination price commodity data and the ordinary commodity data, the user behavior data, the combination price commodity data and the ordinary commodity data can be determined as the first target recommendation object according to the daily trigger number.

[0122] S64, screening the recommended commodity from the first target recommendation object and recommending the recommended commodity to the user.

[0123] In the embodiment of the application, the user interested commodity can be obtained from different dimensions for recommendation. The dimensions can include but are not limited to the user behavior dimensions such as consumption, search and browsing, and the actual commodity conditions such as the combination price commodity and the original price commodity, and the records generated by the non-user behavior dimensions can be used as the basis for commodity recommendation. In order to reduce the interference of the dimension data generated by the occasional behavior on the recommendation accuracy, in the embodiment of the application, different dimensions of commodities can be recommended to the user according to different daily trigger numbers.

[0124] In the specific implementation, the user behavior data, the combination price commodity data and the ordinary commodity data can be arranged in a certain order to obtain the first recommendation order, and then the commodities can be recommended in the first recommendation order, for example, the current recommended commodity is obtained from the user behavior data, and the next time the commodity can be selected from the combination price commodity data for recommendation.

[0125] In actual application, in order to ensure the diversification of commodities and avoid the user's interest in the recommended commodities being weakened due to frequent viewing of repeated commodities, a plurality of commodities can be provided in each first target recommendation object for screening. Different commodity screening methods can be set for different first target recommendation objects.

[0126] In one example, when the first target recommendation object is the user behavior data, the purchase tendency of the user can be found according to the search number, the browsing number and the like of the user, and the commodity can be recommended according to the purchase tendency.

[0127] In another example, when the first target recommendation object is a combination price commodity, the combination price commodities can be sorted according to one or more of the factors such as discount strength, preferential time, whether in season, user purchase tendency, actual sold quantity, user praise degree, etc., and the combination price commodities are recommended in the order of the sorting. When multiple factors are involved, different weights can be set for each factor, the weighted scores of the combination price commodities are calculated in combination with the weights, and the combination price commodities are sorted according to the weighted scores.

[0128] In one example, when the first target recommendation object is a normal commodity, the weighted scores of the normal commodities can also be calculated in combination with the weights of different factors to sort all the normal commodities, and the normal commodities are recommended according to the sorting.

[0129] Considering that different types of commodities can be included in each first target recommendation object, the commodities in the first target recommendation object can be clustered according to commodity categories before the first target recommendation object is filtered; and the weight of each cluster is calculated, and the commodities are preferentially filtered in the cluster with a high weight. The weight of the cluster can be determined according to the user's purchase habits, search habits, browsing habits, etc.

[0130] In the embodiments of the present application, first, the user behavior data, combination price commodity data, normal commodity data and other data of different dimensions are sorted according to the user purchase tendency to obtain a first recommendation order, and the first target recommendation object is determined according to the number of daily triggers in the first recommendation order, which can reduce the number of commodity filtering when the commodities are recommended, thereby reducing the data processing amount; then, the commodities in the first target recommendation object are sorted according to different factors (such as the number of search times, the number of browsing times, etc.) according to the different first target recommendation objects, so as to filter the recommended commodities for the user according to the sorting, which can recommend the commodities of interest to the user as much as possible, improve the accuracy of commodity recommendation, thereby increasing the frequency of the user using the shopping APP, and further improving the user stickiness of the shopping APP.

[0131] It should be noted that when there is no relevant data in a certain dimension (such as the user behavior dimension, non-user behavior dimension, etc. mentioned above), the dimension can not participate in the sorting of the first recommendation order.

[0132] Through the cross-recommendation of commodities of different dimensions, the decrease of user interest degree caused by accidental commodity data or non-user behavior related commodities can be reduced, thereby bringing higher user stickiness.

[0133] In one example, the step of filtering the recommended commodities from the first target recommendation object and recommending the recommended commodities to the user can include:

[0134] S641, when the first target recommendation object is user behavior data, screening the shopping cart commodity data, search commodity data and browsing commodity data from the user behavior data;

[0135] S642, generating a second recommendation order by using the shopping cart commodity data, search commodity data and browsing commodity data;

[0136] S643, obtaining a current commodity recommendation frequency of the user behavior data;

[0137] S644, determining a second target recommendation object in the second recommendation order according to the current commodity recommendation frequency;

[0138] The second target recommendation object is one of the shopping cart commodity data, search commodity data and browsing commodity data, and can be obtained by matching the current commodity recommendation frequency and the second recommendation order.

[0139] The current commodity recommendation frequency refers to the number of times that the user behavior data is determined as the first target recommendation object.

[0140] In actual application, assuming that the second recommendation order is the shopping cart commodity data, search commodity data and browsing commodity data in sequence, the daily trigger frequency is 10 times, and the first recommendation target corresponding to the 2nd, 5th and 8th trigger frequency is the user behavior data, the current commodity recommendation frequency is 3. According to the second recommendation order, the recommended commodities corresponding to the 2nd, 5th and 8th daily trigger frequency can be the commodities in the shopping cart commodity data, search commodity data and browsing commodity data respectively.

[0141] S645, screening the recommended commodities from the second target recommendation object, and recommending the recommended commodities to the user.

[0142] In the embodiment of the application, the user behavior data can include the shopping cart commodity data, search commodity data and browsing commodity data, and when the commodity is recommended, the second recommendation order can be generated by using the above-mentioned user behavior data, so as to sequentially recommend according to the second recommendation order. For example, when the current recommendation is the shopping cart commodity data, the next time the user behavior data needs to be recommended, the search commodity data can be recommended.

[0143] In actual application, in order to ensure the diversification of commodities and avoid the user's interest in the recommended commodities being weakened due to frequently seeing repeated commodities, a plurality of commodities can be provided in each second target recommendation object for screening. Different commodity screening methods can be set for different second target recommendation objects.

[0144] In one example, when the second target recommendation object is shopping cart commodity data, the purchase possibility of each commodity in the shopping cart can be analyzed according to the adding time of the commodity in the shopping cart, the page browsing time, etc., so as to recommend the shopping cart commodities according to the purchase possibility.

[0145] In another example, when the second target recommendation object is search commodity data, the recommendation priority of each search commodity can be set according to the search times, the search frequency, and the recently searched commodities of the search commodity, so as to recommend the search commodities according to the recommendation priority.

[0146] In another example, when the second target recommendation object is browsing commodity data, the recommendation priority of each browsing commodity can be set according to the browsing times, the browsing frequency, and the recently browsed commodities of the browsing commodity, so as to recommend the browsing commodities according to the recommendation priority.

[0147] For the convenience of understanding, the commodity recommendation process is described through specific examples as follows:

[0148] When the user opens the shopping APP to enter the integral check-in page and triggers the "shake" operation, first, it is determined whether the trigger information of the "shake" instruction is cached in the redis. If yes, it is determined that the "shake" operation is not the first trigger of the day, and then the commodity recommendation process is entered.

[0149] First, the cycle period is set to 1 day, and then with the increase of the number of triggers in the day, the repeated recommendation is performed in the following order of 1, 2, and 3. When repeated to the first one, the repeated recommendation is performed in the order of 1A, 1B, and 1C.

[0150] 1. User behavior data related recommendation:

[0151] 1A. Shopping cart commodity data recommendation: all the 50 random commodities under the fine category in the shopping cart are obtained, and one of them is randomly selected for recommendation;

[0152] 1B. Search commodity data recommendation: the latest 50 random commodities with results in the search record are obtained, and one of them is randomly selected for recommendation;

[0153] 1C. Browsing commodity data recommendation: the latest 50 random commodities under the fine category in the browsing commodity are obtained, and one of them is randomly selected for recommendation;

[0154] 2. Combination price commodity data recommendation;

[0155] 3. Normal commodity data recommendation.

[0156] For example, assuming that the above data types all have relevant commodities, then the commodity recommendation is performed in the order of 1A-2-3-1B-2-3-1C-2-3 in a loop. Assuming that 1A above does not have relevant commodities, then the commodity recommendation is performed in the order of 1B-2-3-1C-2-3 in a loop, and so on.

[0157] It should be noted that the above process is only an example of recommendation selecting user behavior data, combining price commodity data and general commodity data as three dimensions, and those skilled in the art can select other dimensional data for commodity recommendation according to actual conditions, and the embodiments of the application do not make specific limitations.

[0158] The application determines whether to recommend points or commodities to the user by judging whether the recommendation instruction triggered by the user is the first time triggered in the day. The uncertainty of point recommendation is enriched by setting different point recommendation forms in different point intervals, so as to fully call the enthusiasm of the user for logging in every day to draw points, thereby enhancing the activity of the user. By setting different dimensions of commodity recommendation channels, the recommended commodities can be more diversified under the condition of meeting the user's preferences, thereby arousing the user's interest in browsing, forming the user's habit of browsing commodities on the current platform, and further enhancing the user's stickiness.

[0159] Please refer to Figure 4 , Figure 4 The structure block diagram of an object recommendation device provided by the embodiments of the application is shown in FIG. 1.

[0160] The embodiments of the application provide an object recommendation device, the objects including points and commodities; the device includes:

[0161] A first-time-in-the-day triggering judgment module 401 is configured to respond to a recommendation instruction triggered by a user and judge whether the recommendation instruction is the first time triggered in the day.

[0162] A cumulative triggering number acquisition module 402 is configured to acquire a cumulative triggering number of the recommendation instruction if the recommendation instruction is the first time triggered in the day.

[0163] A point recommendation module 403 is configured to recommend points to the user according to the cumulative triggering number.

[0164] A triggering number in the day determination module 404 is configured to determine a triggering number in the day of the recommendation instruction if the recommendation instruction is not the first time triggered in the day.

[0165] A commodity recommendation module 405 is configured to recommend commodities to the user according to the triggering number in the day.

[0166] In the embodiments of the application, the first-time-in-the-day triggering judgment module 401 includes:

[0167] The cache information query submodule is configured to query whether the cache information of the recommendation instruction exists in the preset database in response to the recommendation instruction triggered by the user.

[0168] The first trigger judgment submodule of the day is configured to determine that the recommendation instruction is the first trigger of the day if the answer is no.

[0169] In the embodiment of the present application, the integral recommendation module 403 comprises:

[0170] The remainder calculation submodule is configured to calculate the remainders of the cumulative trigger number and at least one preset value in sequence.

[0171] The target value screening submodule is configured to screen the preset value with a remainder of 0 as a target value.

[0172] The first target integral matching submodule is configured to obtain a first integral interval corresponding to the target value and match a first target integral in the first integral interval.

[0173] The first target integral recommendation submodule is configured to recommend the first target integral to the user.

[0174] In the embodiment of the present application, the integral recommendation module 403 further comprises:

[0175] The second target integral recommendation submodule is configured to match a second target integral in a preset second integral interval and recommend the second target integral to the user when all the remainders are not 0.

[0176] In the embodiment of the present application, the first target integral matching submodule comprises:

[0177] The first integral interval obtaining unit is configured to obtain a first integral interval corresponding to the target value.

[0178] The priority order obtaining unit is configured to obtain the priority order of each first integral interval.

[0179] The first target integral matching unit is configured to match the first target integral in the first integral interval with the highest priority.

[0180] In the embodiment of the present application, the commodity recommendation module 405 comprises:

[0181] The user behavior data, combination price commodity data and ordinary commodity data obtaining submodule is configured to obtain the user behavior data, combination price commodity data and ordinary commodity data of the user.

[0182] The first recommendation order generating submodule is configured to generate a first recommendation order by using the user behavior data, combination price commodity data and ordinary commodity data.

[0183] The first target recommendation object determination sub-module is configured to determine a first target recommendation object in a first recommendation sequence according to a daily triggering number.

[0184] The commodity recommendation sub-module is configured to filter a recommended commodity from the first target recommendation object and recommend the recommended commodity to the user.

[0185] In the embodiment of the present application, the commodity recommendation sub-module comprises:

[0186] The filtering unit is configured to filter shopping cart commodity data, search commodity data and browsed commodity data from the user behavior data when the first target recommendation object is the user behavior data.

[0187] The second recommendation sequence generation unit is configured to generate a second recommendation sequence by using the shopping cart commodity data, the search commodity data and the browsed commodity data.

[0188] The current commodity recommendation number acquisition unit is configured to acquire a current commodity recommendation number of the user behavior data.

[0189] The second target recommendation object determination unit is configured to determine a second target recommendation object in the second recommendation sequence according to the current commodity recommendation number.

[0190] The commodity recommendation unit is configured to filter a recommended commodity from the second target recommendation object and recommend the recommended commodity to the user.

[0191] The embodiment of the present application further provides an electronic device, which comprises a processor and a memory:

[0192] The memory is configured to store program code and transmit the program code to the processor.

[0193] The processor is configured to execute the object recommendation method according to the instructions in the program code.

[0194] The embodiment of the present application further provides a computer readable storage medium, which is configured to store program code, and the program code is configured to execute the object recommendation method.

[0195] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, device and unit can refer to the corresponding process in the foregoing method embodiments, and will not be described here.

[0196] Each embodiment in the specification adopts a progressive manner for description, and each embodiment focuses on the difference from other embodiments, and the same and similar parts between each embodiment can be referred to.

[0197] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, apparatus, or computer program product. Accordingly, embodiments of the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, and the like) embodying computer program instructions.

[0198] Embodiments of the present application are described herein with reference to the drawings, which are as follows: Figure 1 one or more processes and / or blocks Figure 1 means for performing the functions specified in the one or more processes and / or blocks

[0199] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 one or more processes and / or blocks Figure 1 means for performing the functions specified in the one or more processes and / or blocks

[0200] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the Figure 1 one or more processes and / or blocks Figure 1 means for performing the functions specified in the one or more processes and / or blocks

[0201] While preferred embodiments of the present application have been described, additional variations and modifications can be made to these embodiments by those skilled in the art once they have the benefit of the foregoing description. Therefore, the appended claims are intended to encompass within their scope all such variations and modifications as are included within the scope of the present application.

[0202] Finally, it needs to be pointed out that in this document, relational terms such as first and second and the like can only be used to differentiate one entity or action from another entity or action, without necessarily requiring or implying that there is any such actual relationship or order between these entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, so that a process, method, article, or terminal device including a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or terminal device. Without more limitations, an element defined by the statement "comprising a" does not exclude the existence of additional identical elements in the process, method, article, or terminal device including the element.

[0203] The above description and the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An object recommendation method characterized by comprising: The objects include points and commodities; the method includes: In response to a user triggering a recommendation instruction, determining whether the recommendation instruction is triggered for the first time on the day; If the recommendation instruction is triggered for the first time on the day, obtaining the cumulative triggering number of the recommendation instruction; According to the cumulative triggering number, recommending points to the user; If the recommendation instruction is not triggered for the first time on the day, determining the triggering number of the recommendation instruction on the day; According to the triggering number on the day, recommending commodities to the user; The step of recommending commodities to the user according to the triggering number on the day includes: Obtaining user behavior data, combined price commodity data and ordinary commodity data of the user; Using the user behavior data, the combined price commodity data and the ordinary commodity data to generate a first recommendation sequence; According to the triggering number on the day, determining a first target recommendation object in the first recommendation sequence; Filtering recommended commodities from the first target recommendation object and recommending the recommended commodities to the user; The step of filtering recommended commodities from the first target recommendation object and recommending the recommended commodities to the user includes: When the first target recommendation object is user behavior data, filtering shopping cart commodity data, search commodity data and browsing commodity data from the user behavior data; Using the shopping cart commodity data, the search commodity data and the browsing commodity data to generate a second recommendation sequence; Obtaining the current commodity recommendation number of the user behavior data; According to the current commodity recommendation number, determining a second target recommendation object in the second recommendation sequence; Filtering recommended commodities from the second target recommendation object and recommending the recommended commodities to the user; The step of recommending points to the user according to the cumulative triggering number includes: In turn, calculate the remainder of the cumulative triggering number and at least one preset value; the preset value is determined according to the churn rate of the shopping APP under different numbers; Filtering the preset value with a remainder of 0 as a target value; Obtaining a first point interval corresponding to the target value, and matching a first target point in the first point interval; Recommending the first target point to the user; The step of obtaining a first point interval corresponding to the target value and matching a first target point in the first point interval includes: Obtaining a first point interval corresponding to the target value; Obtaining the priority order of each first point interval; Matching a first target point in the first point interval with the highest priority.

2. The method of claim 1, wherein, The step of determining whether the recommendation instruction is triggered for the first time on the day in response to the user triggering the recommendation instruction includes: In response to a user triggering a recommendation instruction, querying a preset database to determine whether there is cache information of the recommendation instruction; If not, it is determined that the recommendation instruction is triggered for the first time on the day.

3. The method of claim 1, wherein, Also includes: When all remainders are not 0, matching a second target point in a preset second point interval and recommending the second target point to the user.

4. An object recommendation device, characterized by, The objects include points and commodities; the device includes: The first triggering judgment module is triggered on the same day, and is configured to determine whether the recommendation instruction is triggered for the first time on the same day in response to the user triggering the recommendation instruction; The cumulative triggering number acquisition module is configured to acquire a cumulative triggering number of the recommendation instruction if the recommendation instruction is triggered for the first time on the same day; The integral recommendation module is configured to recommend an integral to the user according to the cumulative triggering number; The triggering number determination module is configured to determine a triggering number of the recommendation instruction on the same day if the recommendation instruction is not triggered for the first time on the same day; The commodity recommendation module is configured to recommend a commodity to the user according to the triggering number on the same day; The commodity recommendation module comprises: The user behavior data, the combined price commodity data and the ordinary commodity data acquisition submodule is configured to acquire user behavior data, combined price commodity data and ordinary commodity data of the user; The first recommendation sequence generation submodule is configured to generate a first recommendation sequence by using the user behavior data, the combined price commodity data and the ordinary commodity data; The first target recommendation object determination submodule is configured to determine a first target recommendation object in the first recommendation sequence according to the triggering number on the same day; The commodity recommendation submodule is configured to filter a recommended commodity from the first target recommendation object and recommend the recommended commodity to the user; The commodity recommendation submodule comprises: The filtering unit is configured to filter cart commodity data, search commodity data and browsing commodity data from the user behavior data when the first target recommendation object is the user behavior data; The second recommendation sequence generation unit is configured to generate a second recommendation sequence by using the cart commodity data, the search commodity data and the browsing commodity data; The current commodity recommendation number acquisition unit is configured to acquire a current commodity recommendation number of the user behavior data; The second target recommendation object determination unit is configured to determine a second target recommendation object in the second recommendation sequence according to the current commodity recommendation number; The commodity recommendation unit is configured to filter a recommended commodity from the second target recommendation object and recommend the recommended commodity to the user; The integral recommendation module comprises: The remainder calculation submodule is configured to calculate remainders of the cumulative triggering number and at least one preset value in sequence; the preset value is determined according to a churn rate of a shopping APP at different times; The target value filtering submodule is configured to filter a preset value with a remainder of 0 as a target value; The first target integral matching submodule is configured to acquire a first integral interval corresponding to the target value and match a first target integral in the first integral interval; The first target integral recommendation submodule is configured to recommend the first target integral to the user; The first target integral matching submodule comprises: The first integral interval acquisition unit is configured to acquire a first integral interval corresponding to the target value; The priority order acquisition unit is configured to acquire a priority order of each first integral interval; The first target integral matching unit is configured to match the first target integral in the first integral interval with the highest priority order.

5. An electronic device, comprising: The device comprises a processor and a memory: The memory is configured to store program code and transmit the program code to the processor; The processor is configured to execute the object recommendation method according to instructions in the program code.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium is configured to store program code for performing the object recommendation method according to any one of claims 1-3.

Images