CN116957750A

CN116957750A - Commodity package generation method, commodity package generation system, storage medium and electronic equipment

Info

Publication number: CN116957750A
Application number: CN202311212615.6A
Authority: CN
Inventors: 潘春霞; 姜凤龙; 朱亚辉
Original assignee: Suzhou Jiyi Technology Co ltd
Current assignee: Suzhou Jiyi Technology Co ltd
Priority date: 2023-09-20
Filing date: 2023-09-20
Publication date: 2023-10-27
Anticipated expiration: 2043-09-20
Also published as: CN116957750B

Abstract

The embodiment of the application discloses a commodity package generation method, a system, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring a first commodity paid by a user and a second commodity to be paid; determining a first recommended good associated with the first good, the first good having a time period of separation greater than 0 between a start time of use of the first good and an estimated start time of use of the first recommended good; determining a first association degree of each first recommended commodity and the first commodity; according to the interval duration and the first association degree, combining the first recommended commodities to generate a first commodity package; and according to the second commodity, adjusting a first recommended commodity in the first commodity package to obtain a target commodity package. By adopting the embodiment of the application, the commodity package which fits the purchasing requirement of the user can be generated.

Description

Commodity package generation method, commodity package generation system, storage medium and electronic equipment

Technical Field

The application relates to the technical field of network information mining, in particular to a commodity package generation method, a commodity package generation system, a storage medium and electronic equipment.

Background

With the vigorous development of the mobile internet and the popularization of electronic payment means, an e-commerce platform has become a preferred way for online shopping for wide consumers. In order to improve the online shopping experience and platform sales performance of clients, the e-commerce platform can actively push commodity packages to users.

Existing commodity packages are generated based primarily on user history purchases. For example, a user purchases an infant bath product 1 year ago, and the e-commerce platform generates a commodity package containing infant bath commodities according to the record. However, the child of the user may now have entered the childhood stage, and such baby merchandise is no longer needed, but is instead needed.

As can be seen, the existing solution can generate a large number of invalid commodity packages, which wastes the pushing resources of the platform.

Disclosure of Invention

The application provides a commodity package generation method, a commodity package generation system, a storage medium and electronic equipment, which can generate a commodity package which meets the purchase demands of users.

In a first aspect of the present application, the present application provides a method for generating a commodity package, including:

acquiring a first commodity paid by a user and a second commodity to be paid;

determining a first recommended good associated with the first good, the first good having a time period of separation greater than 0 between a start time of use of the first good and an estimated start time of use of the first recommended good;

Determining a first association degree of each first recommended commodity and the first commodity;

according to the interval duration and the first association degree, combining the first recommended commodities to generate a first commodity package;

and according to the second commodity, adjusting a first recommended commodity in the first commodity package to obtain a target commodity package.

By adopting the technical scheme, the commodity paid and to-be-paid by the user is obtained, and the current and future purchasing demands of the user can be comprehensively reflected. And determining the first recommended commodity with continuous use time according to the first commodity, so as to realize the continuity of the recommended result. Further calculating the multidimensional association degree of the first recommended commodity and the first commodity, comprehensively evaluating the matching degree, and enabling the recommendation to be more accurate. And optimizing and selecting the recommended commodity with the highest association degree through a combination algorithm to form a first commodity package with the highest matching degree. And finally, the first commodity package is adjusted according to the second commodity to be paid, so that the recommended dynamic update is realized, and the latest purchasing intention of the user is met. The commodity package conforming to the purchasing demands of the user can be generated by combining the purchasing behaviors of the user.

Optionally, the combining each of the first recommended commodities according to each of the interval durations and each of the first relevancy, and generating a first commodity package includes:

And combining the first recommended commodities according to the interval duration and the first association degree of the first recommended commodities by adopting a knapsack algorithm to obtain the first commodity package, wherein the knapsack capacity of the knapsack algorithm is a preset duration, and the knapsack price of the knapsack algorithm is the first association degree of the first recommended commodities in the first commodity package.

By adopting the technical scheme, the preset total commodity package duration is taken as the knapsack capacity, the commodity association degree is taken as the commodity value, and the commodity use time interval is taken as the commodity weight, so that the specific problem is abstracted into a standard knapsack problem model. And then solving by applying a knapsack algorithm, and selecting a commodity bag which can maximize the total relevance value sum in the knapsack under the constraint of not exceeding the total duration.

Optionally, the step of combining each first recommended commodity to obtain the first commodity package according to the interval duration and the first association degree of each first recommended commodity by using a knapsack algorithm includes:

substituting the interval duration and the first association degree of each first recommended commodity into a state transfer equation of the knapsack algorithm to obtain a target recommended commodity;

wherein, the state transition equation is:

F(i，t)=max{F(i-1，t)，F(i-1，t-Δt[i])+R[i]}；

Wherein F (i, t) represents that the sum of the first association degrees of the first i first recommended commodities is maximum within a preset time period, F (i-1, t) represents that the sum of the first association degrees of the first i-1 first recommended commodities is maximum within a preset time period when the first i first recommended commodities are not included, deltat [ i ] represents the interval time period of the first i first recommended commodities, and Ri represents the first association degree of the first i first recommended commodities;

and according to the state transition equation and a dynamic programming algorithm, combining the target recommended commodities to obtain a first commodity package, wherein the sum of the interval durations of the first recommended commodities in the first commodity package is smaller than or equal to the preset duration, and the sum of the first relevancy of the first recommended commodities is maximum.

By adopting the technical scheme, the dynamic programming algorithm is applied to solve the state transfer equation, so that the optimal commodity combination with the maximum association degree can be obtained. Compared with the knapsack algorithm for directly solving, the method introduces a mathematical model and dynamic programming to enable the problem description and solving to be more accurate. The state transition equation intuitively represents the conditions that require optimization and constraints for recommended merchandise selection. The dynamic programming algorithm can efficiently and quickly solve the equation. And obtaining the optimal solution, and then backtracking and deducing to obtain the target commodity package with the maximum association degree weight.

Optionally, the adjusting the recommended commodity in the first commodity package according to the second commodity to obtain a target commodity package includes:

determining whether an interval duration exists between the starting use time of the second commodity and the estimated starting use time of the first recommended commodity, and adding the second commodity without the interval duration into the first commodity package;

determining a differential time length of the first commodity package;

if a second commodity with the interval duration being less than or equal to the difference duration exists, adding the second commodity with the interval duration being less than or equal to the difference duration into the first commodity package;

if a second commodity with the interval time longer than the difference time exists, replacing a first recommended commodity with the lowest first association degree in the first commodity package with the second commodity with the interval time longer than the difference time;

and determining the adjusted first commodity package as the target commodity package.

By adopting the technical scheme, whether the service time of the second commodity is continuous with that of the recommended commodity is judged, and the second commodity without intervals is directly added, so that time faults are avoided. And then calculating the residual time capacity of the first commodity package, and directly adding second commodities with intervals smaller than the capacity in the capacity range. And when the second commodity interval exceeds the capacity, performing a replacement operation, and replacing the recommended commodity with the second commodity with the weakest association degree. On the premise of ensuring time continuity, the second commodity is fused, so that the dominant effect of the first commodity package is exerted, and the latest influence of the second commodity is reflected. The dynamic adjustment and optimization of the commodity package are realized, so that the time consistency is ensured, and the association degree update according to the second commodity is also embodied.

Optionally, the determining the first association degree between each of the first recommended commodities and the first commodity includes:

acquiring user information, and determining the user characteristic association degree of the user and each first recommended commodity according to the user information;

determining the cost performance association degree of the first commodity and each first recommended commodity according to the price information and the evaluation information of the first commodity;

multiplying the user characteristic association degree of each first recommended commodity by a first weight to obtain a first association value;

multiplying the cost performance relevance by a second weight to obtain a second relevance value;

and determining the sum of the first association value and the second association value of each first recommended commodity as the first association degree.

By adopting the technical scheme, the user characteristic association degree is calculated, the personalized matching degree of the commodity can be estimated from the user characteristic dimension, then the cost performance association degree is calculated, the matching degree of the commodity can be estimated from the cost performance dimension, and the commodity with similar cost performance can be recommended. For reasonably determining the weight of the two relevancy degrees, weight coefficients are respectively set for the two relevancy degrees. And finally, calculating the sum of weighted association values as a first association degree, comprehensively considering two dimensions of the user characteristics and the cost performance, and calculating the first association degree capable of comprehensively reflecting the matching degree of the commodity and the user.

Optionally, the adjusting the first recommended commodity in the first commodity package according to the second commodity, after obtaining the target commodity package, further includes:

the target commodity package is sent to a user side;

receiving first feedback information fed back by the user side according to the target commodity package;

if the first feedback information is satisfied, adding the target commodity package into the user information of the user terminal;

if the first feedback information is unsatisfactory, generating a second commodity packet according to unsatisfactory commodities in the first feedback information, sending the second commodity packet to the user side, and receiving second feedback information fed back by the user side according to the second commodity packet;

and if the second feedback information fed back by the user side according to the second commodity packet is determined to be unsatisfactory, the second feedback information is taken as first feedback information, the step of generating a second commodity packet according to the first feedback information is re-executed, and the second commodity packet is sent to the user side until the second feedback information is satisfactory.

By adopting the technical scheme, the target commodity package is pushed to the user, and feedback comments of the user are collected. If the feedback is satisfactory, the commodity package is saved for immediate future use. If the feedback is unsatisfactory, a new commodity package is generated according to the unsatisfactory commodity adjustment of the user, and the feedback is acquired again. If the new packet feedback is still unsatisfactory, the iterative adjustment process described above is repeated until the user is satisfied. Through multi-round interactive feedback with the user, the real demand of the user can be continuously approximated, and the recommendation effect of the commodity package is gradually improved. Meanwhile, the closed-loop iteration mechanism can train the system to automatically adjust the capability of the recommended result.

Optionally, the generating a second commodity package according to the unsatisfactory commodity in the first feedback information includes:

adjusting a first weight and a second weight based on the unsatisfactory merchandise;

deleting the unsatisfactory commodity in the target commodity package, and re-executing the step of determining a first recommended commodity associated with the first commodity to obtain a target commodity package;

and taking the target commodity package as the second commodity package.

By adopting the technical scheme, the weight of the unsatisfactory commodity and related commodity is reduced according to the fed-back unsatisfactory commodity. And deleting unsatisfactory commodities, and recalculating the association degree to generate a new commodity package. At the time of regeneration, the new result will automatically avoid the unsatisfactory merchandise due to the reduced weight of the unsatisfactory merchandise. And taking the newly generated commodity package as a second commodity package. The targeted adjustment is performed through user feedback, the original algorithm framework is reserved, and user opinion and historical data are combined. The new commodity package still shows the consistency of personalized recommendation on the basis of eliminating unsatisfied commodities.

In a second aspect of the present application, there is provided a commodity package generating system comprising:

the commodity information acquisition module is used for acquiring a first commodity paid by a user and a second commodity to be paid;

A first recommended article determination module configured to determine a first recommended article associated with the first article, the interval time between a start-use time of the first article and an estimated start-use time of the first recommended article being greater than 0;

the first association degree calculation module is used for determining a first association degree of each first recommended commodity and the first commodity;

the first commodity package generating module is used for combining the first recommended commodities according to the interval duration and the first association degree to generate a first commodity package;

and the target commodity package generating module is used for adjusting the first recommended commodity in the first commodity package according to the second commodity to obtain a target commodity package.

In a third aspect the application provides a computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the above-described method steps.

In a fourth aspect of the application there is provided an electronic device comprising: a processor, a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the above-mentioned method steps.

In summary, one or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:

by adopting the technical scheme of the application, the commodity paid and to-be-paid by the user is obtained, and the current and future purchasing demands of the user can be comprehensively reflected. And determining the first recommended commodity with continuous use time according to the first commodity, so as to realize the continuity of the recommended result. Further calculating the multidimensional association degree of the first recommended commodity and the first commodity, comprehensively evaluating the matching degree, and enabling the recommendation to be more accurate. And optimizing and selecting the recommended commodity with the highest association degree through a combination algorithm to form a first commodity package with the highest matching degree. And finally, the first commodity package is adjusted according to the second commodity to be paid, so that the recommended dynamic update is realized, and the latest purchasing intention of the user is met. The commodity package conforming to the purchasing demands of the user can be generated by combining the purchasing behaviors of the user.

Drawings

Fig. 1 is a flow chart of a method for generating a commodity package according to an embodiment of the present application;

fig. 2 is a schematic structural diagram of a commodity package generating system according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of an electronic device according to the disclosure.

Reference numerals illustrate: 300. an electronic device; 301. a processor; 305. a memory; 303. a user interface; 304. a network interface; 302. a communication bus.

Detailed Description

In order that those skilled in the art will better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments.

In describing embodiments of the present application, words such as "for example" or "for example" are used to mean serving as examples, illustrations, or descriptions. Any embodiment or design described herein as "such as" or "for example" in embodiments of the application should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "or" for example "is intended to present related concepts in a concrete fashion.

In the description of embodiments of the application, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of screen terminals means two or more screen terminals. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating an indicated technical feature. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

In an embodiment, please refer to fig. 1, fig. 1 is a flow chart of a commodity package generating method according to an embodiment of the present application, where the method may be implemented by a computer program, may be implemented by a single chip microcomputer, or may be run on a commodity package generating system based on von neumann system. The computer program may be integrated in the application or may run as a stand-alone tool class application. Specifically, the method may include the steps of:

step 101: the method comprises the steps of obtaining a first commodity paid by a user and a second commodity to be paid.

The paid first commodity can be understood as a commodity which is purchased or placed by a user on an e-commerce platform, and reflects the current purchase requirement of the user. The second item to be paid may be understood as an item that the user has not purchased on the e-commerce platform, but may be in the shopping cart, the collection bar, the state to be paid, and the wish list, reflecting the user's future purchase plan. The acquisition of the first merchandise and the second merchandise may accurately reflect the current and future purchase needs of the user.

In one possible implementation, after the user confirms the purchase of the order at the e-commerce platform, the historical purchase record of the user at the platform is obtained immediately to determine the first and second merchandise.

In another implementation, all purchase records of the user over a period of time may be periodically extracted to determine the first and second merchandise.

By combining the historical purchasing behavior and the current purchasing intention of the user, the purchasing requirement of the user can be comprehensively and accurately reflected, and an important reference is provided for the generation of the follow-up personalized commodity package. The real-time updating and the periodic extraction of the acquisition mode also realize the combination of the effectiveness and the comprehensiveness.

Step 102: a first recommended good associated with the first good is determined, the interval between the time of onset of use of the first good and the estimated time of onset of use of the first recommended good being greater than 0.

The first recommended commodity refers to a commodity with the connection with the service time of the first commodity. And the use time with connectivity may include: continuity of the use stage, overlapping of the use age groups, continuity of the use period, matching degree of the repeated purchase period and the like, and the interval between the start use time of the first commodity and the first recommended commodity and the estimated start use time of the first recommended commodity is longer than 0. Examples are as follows:

(1) Continuity of the usage phase;

if the use period of the first commodity is continuous with the use period of the first recommended commodity, for example, the first commodity is infant milk powder of 0-1 year old and the first recommended commodity is infant milk powder of 1-3 year old, the use periods of the first commodity and the first recommended commodity are continuous, the time engagement is considered to exist, and the interval time of the first recommended commodity is 1-2 years long.

(2) Overlap of age groups used;

if there is an overlap in age groups of two products, for example, the first product has an applicable age of 0-2 years, the first recommended product has an applicable age of 1-3 years, and the two product has an overlap of 1-2 years, then the time engagement is considered to exist, and the first recommended product has an interval duration of 1 year.

(3) Continuity of the usage period;

if the end of the period of use of the first commodity is close to the beginning of use of the first recommended commodity, for example within 1-3 months of each other, then it may be considered that there is a link in the period of use, the first recommended commodity being spaced apart by 1-3 months.

(4) Repeating the matching degree of the purchase period;

if the first recommended merchandise matches the repeat purchase period of the first merchandise, the time to purchase the first recommended merchandise may be linked to the first merchandise use time, and the first recommended merchandise interval period is 1 year.

Specifically, the engagement of the use time can better meet the continuous demand evolution of the user, after the first commodity is determined, candidate recommended commodities with which the use time features are engaged can be searched in the commodity database, the interval between the commodities and the use time of the first commodity is calculated, and the commodity with the interval larger than 0 is selected as the first recommended commodity.

It should be noted that the above examples are only illustrative, and the use time engagement can be determined from various aspects, and is not limited to these standards. Reasonable time engagement judgment standards are selected, so that the continuous change of the user demands can be met, and recommended commodities which meet the actual demands better are provided.

Step 103: and determining a first association degree between each first recommended commodity and the first commodity.

The first association degree refers to the degree that the first recommended commodity and the first commodity can meet the related requirements of the user. The above process determines a plurality of first recommended products that are satisfactory from the standpoint of time engagement. This process is to calculate a first degree of association of the first recommended good with the first good from the more dimensions. The purpose of calculating the first degree of association is to generate a commodity package matching the user's needs according to the first degree of association. If considered only from the dimension of the time of use, there may be multiple recommended merchandise candidates. And adding association degree calculation of other dimensions, and finding recommended commodities which are more strongly associated with the first commodity in more aspects to generate a commodity package with higher matching degree.

On the basis of the above embodiment, as an alternative embodiment, step 103 above: the step of determining the first association degree between each first recommended commodity and the first commodity may specifically further include the following steps:

Step 201: and acquiring user information, and determining the user characteristic association degree of the user and each first recommended commodity according to the user information.

The user information is acquired, the user characteristic association degree is calculated, the personalized matching degree of the commodity can be estimated from the dimension of the user characteristics, the user characteristics are used as important references for commodity recommendation decisions, the adaptation degree of recommendation results to specific users is improved, and personalized recommendation in the true sense is realized.

Specifically, the user characteristics including the basic information, interest and hobbies, purchasing preference, life stage of the user and the like can be extracted from the registered information, browsing records, purchasing history, user labels and other multifaceted data of the user on the e-commerce platform.

Further, for each first recommended commodity, evaluating which features of the user can be matched, wherein the more matched user features, the higher the association degree of the first recommended commodity and the user characteristics of the user is. For example, for a female user in the period of baby care, the commodity such as the baby bath product can be matched with the characteristics of her baby care stage, female identity and the like, so that the commodity has higher association with the user, while the commodity of the child amusement facility can only be associated with the female identity, and the association of the user characteristics is lower.

By the method, the matching degree of the commodity and the specific user can be quantized from the angle of the user characteristics, the matching degree is used as an important judgment dimension of commodity selection, and the matching degree of the finally formed commodity package to the personalized demands of the user is improved by integrating the commodity package generation and screening process, so that the truly personalized and accurate commodity recommendation service is provided.

Step 202: and determining the cost performance association degree of the first commodity and each first recommended commodity according to the price information and the evaluation information of the first commodity.

The price and evaluation information of the first commodity are obtained to calculate the correlation degree of the cost performance, the matching degree between the commodities can be estimated from the dimension of the cost performance, the characteristic of the cost performance is used as an important judgment standard for commodity selection, and the commodity with the cost performance similar to or better than that of the purchased commodity is recommended to a user.

Specifically, the system may acquire information such as sales price, historical price, user evaluation, platform evaluation, etc. of the first commodity, and perform quantitative processing on the information. And then the system can acquire similar price and evaluation information of each commodity in the first recommended commodity pool, compare the same price and evaluation value with the price and evaluation value of the first commodity, and determine the difference between the price level and the evaluation level, namely determine the cost performance association degree. The commodity with high cost performance correlation degree shows that the cost performance characteristic of the commodity is closer to or better than that of the first commodity.

For example, if the first commodity is a flat milk powder, its price is near the median, and the score is 4.5 points. The first recommended commodity also belongs to a flat price milk powder section, the price is similar, the score is 4.7, and the cost performance is not inferior and slightly superior to that of the first commodity, so that the relevance of the cost performance is higher. If another recommended commodity is milk powder with higher end, the price is obviously higher, and the cost performance correlation degree is lower.

The cost performance is used as a judgment standard for selecting recommended commodities, so that the commodity with overlarge difference between the cost performance and the cost performance of the purchased commodity is prevented from being recommended, the recommendation result is ensured to be more matched with the expectation and preference of the user, the satisfaction degree of the user on the recommendation is improved, and the whole personalized recommendation scheme is perfected.

Step 203: and multiplying the user characteristic association degree of each first recommended commodity by a first weight to obtain a first association value.

Specifically, in order to comprehensively consider the influence of two dimensions, namely, the user characteristic association degree and the cost performance association degree, on the selection of the first recommended commodity, different weights are required to be given to each association degree index, and the weighted comprehensive association degree is calculated.

The first weight is given to the association degree of the user characteristics, so that the importance degree of the judgment dimension of the user characteristics in the final calculation of the association degree of the recommended commodity can be determined. Specifically, the influence of the user characteristics on the user satisfaction degree and the purchase conversion rate can be determined according to the historical data of the platform, and the value of the first weight is set according to the influence. Generally, the degree of personalized matching can be directly reflected by the relevance of the user characteristics, and the weight value of the user characteristics is higher.

And then calculating a first association value of each first recommended commodity in the dimension of the user characteristic association degree according to the product of the preset first weight value and the user characteristic association degree of each first recommended commodity. For example, if the user characteristic association degree of a certain commodity is 0.8 and the first weight is set to 0.6, the first association value of the commodity in the user characteristic dimension is 0.48.

Step 204: multiplying the cost performance relevance by a second weight to obtain a second relevance value.

The cost performance relevance is given a second weight, the judgment dimension of the cost performance can be determined, and the influence weight occupied when the final relevance of the commodity is calculated. Specifically, the value of the second weight may be set according to the magnitude of the influence of the deterministic cost ratio on the user decision according to the data of the platform. Because the cost performance relevance is relatively objective, the weight of the cost performance relevance may be lower than the relevance of the user characteristics reflecting individualization.

After the second weight is determined, a second association value of each commodity on the dimension of the cost performance association degree can be calculated according to the product of the preset second weight and the cost performance association degree of each first recommended commodity. For example, the cost performance relevance of a commodity is 0.7, and the second weight is set to 0.3, then the second relevance of the commodity in the cost performance dimension is 0.21.

The function of the cost performance in the overall relevance judgment can be reasonably reflected through weighted calculation, the cost performance is quantized and then integrated into a relevance index system, support is provided for subsequent commodity recommendation, commodities similar to or better in cost performance than purchased commodities of users are recommended, satisfaction of the users is improved, and an overall personalized recommendation scheme is perfected.

Step 205: and determining the sum of the first association value and the second association value of each first recommended commodity as a first association degree.

After the association degree of each first recommended commodity in the two dimensions is calculated respectively, the two association values are combined to obtain the first association degree capable of comprehensively reflecting the matching degree of the first recommended commodity and the first commodity. The method provides index support for the subsequent combined commodity package according to the first association degree, and can recommend commodities which have strong association with the purchased commodities of the user in two key aspects of user characteristics and cost performance.

Step 104: and combining the first recommended commodities according to the interval duration and the first association degree to generate a first commodity package.

Specifically, the generation of the first commodity package needs to take into account the usage time engagement of the first recommended commodity and the degree of association with the first commodity, and the commodity package is formed by combining the recommended commodities satisfying both conditions to recommend the commodity which is both continuous and relevant.

In a possible implementation manner, a knapsack algorithm can be adopted, and according to the interval duration and the first association degree of each first recommended commodity, each first recommended commodity is combined to obtain a first commodity package, wherein the knapsack capacity of the knapsack algorithm is a preset duration, and the knapsack price value of the knapsack algorithm is the first association degree of each first recommended commodity in the first commodity package.

The combination optimization problem of the recommended commodities can be mathematically and accurately solved by adopting a knapsack algorithm, so that a first commodity package with optimal association degree is generated.

Specifically, the total duration of the generation of the preset commodity package is taken as the 'knapsack capacity' of a knapsack algorithm, and the first association degree of each recommended commodity is taken as the 'commodity value'. The time interval of use of the recommended merchandise is then considered as the "weight of the item" and the recommended merchandise is placed in turn into the backpack in order to maximize the sum of the value of the merchandise within the backpack without exceeding the total capacity of the backpack.

For example, the backpack has a capacity of 1 year, a commodity A association degree of 0.8, an interval of 2 months, and a commodity B association degree of 0.6, and an interval of 6 months. A is placed in the order from high to low, then the space remains for 10 months, and B cannot be placed any more. Then there is only a in the final backpack, with a total value of 0.8.

The recommended commodity combination with the highest association weight can be obtained through accurate calculation of the mathematical model, so that the method meets the requirement of using time continuity and provides the commodity with the highest association. Compared with visual combination, the calculation result is better, and the method is quicker and more efficient. The knapsack algorithm can effectively solve the problem of recommending combination optimization, and provides a personalized and high-quality first commodity package.

Based on the above embodiment, as an alternative embodiment, step 104: according to the interval duration and the first association degree, combining the first recommended commodities to generate a first commodity package, and specifically, the method further comprises the following steps:

step 301: substituting the interval duration and the first association degree of each first recommended commodity into a state transition equation to obtain a target recommended commodity.

The state transfer equation of the knapsack algorithm is as follows:

F(i，t)=max{F(i-1，t)，F(i-1，t-Δt[i])+R[i]}；

wherein F (i, t) represents that the sum of the first relevancy of the first i first recommended commodities is maximum within a preset time period, F (i-1, t) represents that the sum of the first relevancy of the first i-1 first recommended commodities is maximum within a preset time period when the first i first recommended commodities are not included, deltat [ i ] represents the interval time period of the first i first recommended commodities, and Ri represents the first relevancy of the first i first recommended commodities.

The state transition equation is a core mathematical model for solving the recommended commodity combination optimization problem by using a knapsack algorithm. The objective of introducing the state transition equation is to provide a mathematical framework of accurate calculations that can explicitly describe the optimization objectives and included variable conditions of recommended commodity selections. The method aims at maximizing the association degree sum of the selected commodities of the previous i pieces within a preset duration range. The decision variables comprise the sum of the relevancy of the previous i commodities, the sum of the relevancy when the i commodity is not selected, the time interval of the i commodity and the relevancy.

The time interval and the association degree of each recommended commodity can be substituted into the state transition equation in turn for calculation, and the maximum association degree sum which can be achieved under the current condition can be obtained by comparing the sum of the association degrees under the two conditions of selecting the commodity and not selecting the commodity. And recursively carrying out the calculation process to finally obtain the recommended commodity combination with the maximum association degree weight sum.

Compared with direct combination, the state transfer equation provides a mathematical optimization framework, and can quickly and accurately solve the combination optimization problem. The mathematical model is used instead of manual judgment, and the calculation result is more objective and reliable. Therefore, the introduction of the state transition equation provides effective support for generating the recommended commodity combination with the strongest association degree.

Step 302: according to a state transition equation and a dynamic planning algorithm, combining the target recommended commodities to obtain a first commodity package, wherein the sum of the interval durations of the first recommended commodities in the first commodity package is smaller than or equal to the preset duration, and the sum of the first relevancy of the first recommended commodities is maximum.

The dynamic programming algorithm is introduced to effectively solve the state transition equation so as to obtain the optimal recommended commodity combination with the maximum sum of the association degrees. And solving an optimal combination scheme of recommended commodities by using an optimization algorithm in a mathematical optimization framework provided by the state transfer equation. The dynamic programming algorithm can efficiently calculate the optimal solution of the state transition equation.

Specifically, the time intervals and the associated data of all recommended commodities are input into a dynamic programming algorithm, and for each recommended commodity, the values of the target state transition equations under the two decisions of 'selecting the commodity' and 'not selecting the commodity' are compared, and if the commodity is selected to make the target value larger, the target value is added into the solution set. And sequentially and circularly traversing each commodity to perform decision optimization, and finally obtaining an optimal recommended commodity set with the maximum correlation sum and the total using time meeting the preset duration limit as the generated first commodity package.

Compared with an exhaustion method, the dynamic planning utilizes the characteristic of overlapping sub-problems of the problem, so that the calculated amount is greatly reduced, and the optimization problem can be quickly and efficiently solved. The first commodity package is obtained by dynamic programming, and the calculation result is more accurate and reliable.

This process will be illustrated below:

assume that the following 7 first recommended commodities are: commodity 1, wherein the first association degree is 60, and the interval duration is 1 hour; commodity 2, wherein the first association degree is 100, and the interval duration is 2 hours; commodity 3, wherein the first association degree is 120, and the interval duration is 3 hours; commodity 4, wherein the first association degree is 50, and the interval duration is 5 hours; commodity 5, wherein the first association degree is 80, and the interval duration is 2 hours; commodity 6, wherein the first association degree is 70, and the interval duration is 3 hours; the first degree of association of commodity 7 is 90, and the interval duration is 4 hours.

Assuming that the preset time length is 8 hours, generating a commodity package according to 7 first recommended commodities.

First, a 2-dimensional array needs to be created, where F [ i ] [ t ] represents the maximum degree of correlation that the first i items can produce at a total time of t. And through the state transition equation: f (i, t) =max { F (i-1, t), F (i-1, t- Δtj) +Rj ] }; the 2-dimensional array is populated to yield an array as described in table 1.

F(t,i)	0	1	2	3	4	5	6	7	8
										0	0	0	0	0	0	0	0	0	0
1	0	60	60	60	60	60	60	60	60
										2	0	60	100	160	160	220	280	280	340
3	0	60	100	160	220	280	340	400	460
										4	0	60	100	160	220	280	340	400	460
5	0	60	100	180	240	300	360	420	480
										6	0	60	100	180	250	310	370	430	490
7	0	60	100	180	250	340	400	400	520

TABLE 1

In order to find a specific commodity contained in the commodity package with the highest correlation, the forming path of the solution needs to be traced back and gradually deduced from the obtained optimal solution, wherein the rows in table 1 represent the total time (t): from 0 to 8 hours, each number represents one hour. The rows of table 1 represent the first i commodities: from 0 to 7, each number represents a commodity.

Specifically, starting from the calculated maximum degree of correlation F7 8, it can be inferred that item 7 is included in the optimal solution because F7 8 is not equal to F6 8. Then looking at the solution F6 8-Deltat 7 = F6 4 when no item 7 is included, it can be inferred that item 6 is not in the optimal solution because F6 4 equals F5 4. Then, checking F5 4, since F5 4 is not equal to F4, it can be inferred that the 5 th commodity is included in the optimal solution. Continuing to look at F4 [ 4-Deltat [5] = F4 [2], because F4 [2] is not equal to F3 [2], it can be inferred that item 4 is also included in the optimal solution. Then examining F3 ] [ 2-Deltat [4] = F3 ] [0], since F3 ] [0] equals F2 ] [0], it can be inferred that the 3 rd commodity is not in the optimal solution. Finally, F2 0 equals F1 0, so it can be inferred that the 2 nd and 1 st products are not in the optimal solution.

By backtracking and deriving the path of formation of the optimal solution step by step, it can be determined that the optimal package of commodities is a combination of commodity 7, commodity 5, and commodity 4, with a maximum degree of association of 520.

Step 105: and according to the second commodity, adjusting the first recommended commodity in the first commodity package to obtain the target commodity package.

The second commodity is a commodity which is not paid by the user, and the first commodity package is adjusted according to the second commodity, so that possible purchasing behavior feedback of the user can be integrated into a recommendation result, and the recommendation is more personalized and dynamic.

Specifically, the association degree of the second commodity and each recommended commodity in the first commodity package can be searched, and if the related commodity exists, the weight or the arrangement of the related commodity in the commodity package can be improved, so that more recommended exposure can be obtained for the commodity which is more matched with the new purchase.

Meanwhile, if the first commodity package contains commodities which are highly repeated with the second commodity, filtering and removing the commodities can be considered, so that the commodities which have been purchased by the recommending user can be avoided. In addition, new recommended commodities can be selected from the alternative commodity library in a supplementary mode according to the characteristics of the second commodity, and the constitution of the commodity package is further optimized.

Through adjustment and optimization of the first commodity package, the generated target commodity package can not only consider historical purchasing preference of the user, but also reflect the latest decision of the user, and realize the dynamization of the recommendation result. The method can improve the individuation degree of recommendation, enhance the continuous purchase wish of the user and enable the commodity package generation scheme to be more intelligent.

Based on the above embodiment, as an alternative embodiment, in step 105: according to the second commodity, the first recommended commodity in the first commodity package is adjusted to obtain a target commodity package, and the step can specifically further comprise the following steps:

step 401: determining whether an interval duration exists between the starting use time of the second commodity and the estimated starting use time of the first recommended commodity, and adding the second commodity without the interval duration into the first commodity package.

Specifically, the estimated start use time of each recommended article in the second article and the first article package may be first acquired.

And then judging whether an interval exists between the second commodity and the starting use time of each recommended commodity in sequence. If the second commodity has no interval from the start of use of a recommended commodity, i.e. their use times can be closely linked, the second commodity can be added directly to the commodity package.

Through the step, the second commodity can be newly added into the commodity package without damaging time continuity, so that commodity selection is enriched, the purchasing intention of a user is attached, and dynamic optimization of the commodity package is realized. Meanwhile, space is created for subsequent second commodity adjustment, and time conflict is prevented.

Step 402: a differential time period for the first commodity package is determined.

Specifically, a preset time length of the first commodity package is obtained, and then the sum of time intervals of all recommended commodities in the commodity package is subtracted, wherein the obtained allowance is the difference time length of the commodity package. The deficit duration may reflect a remaining capacity of the first commodity package.

Step 403: and if the second commodity with the interval duration being smaller than or equal to the difference duration exists, adding the second commodity with the interval duration being smaller than or equal to the difference duration into the first commodity package.

Specifically, the usage time interval of the second commodity and the difference time length of the first commodity package obtained through calculation are firstly obtained. And then judging whether the interval duration of each second commodity is smaller than or equal to the difference duration one by one. And for the second commodity with the interval duration meeting the requirement, directly adding the second commodity into the first commodity package.

For example, if the first commodity package is 6 months in difference, the second commodity a is 3 months apart, and the second commodity B is 5 months apart, both a and B may be added directly to the commodity package to form a combination package containing more related commodities.

On the premise of ensuring continuous time connection, dynamic expansion of the commodity package is realized, so that not only is the vacant space of time fully utilized, but also the new purchase intention of the user is effectively integrated into commodity package generation. The recommendation is richer and more personalized, and the user can obtain more flexible commodity selection.

Step 404: and if the second commodity with the interval time longer than the difference time exists, replacing the first recommended commodity with the lowest first association degree in the first commodity package with the second commodity with the interval time longer than the difference time.

Specifically, it is first determined whether the interval duration of each second commodity is greater than the difference duration of the first commodity package. For the second commodity with the interval time exceeding the difference time, the second commodity cannot be directly added into the commodity package, otherwise, the continuity of time is destroyed. At this time, the system identifies the recommended product with the lowest association in the first product package, and then replaces the recommended product with the second product.

By the replacement mode, time continuity of the commodity package is guaranteed, the second commodity with higher association degree is used for replacing the original recommended commodity with lower weight, and association degree and quality of the commodity package are effectively improved.

Step 405: and determining the adjusted first commodity package as a target commodity package.

The above embodiment has described the generation process of the target commodity package, and on the basis of the above embodiment, the process of pushing the target commodity package to the user side will be described below. The process may specifically further comprise the steps of:

Step 501: and sending the target commodity package to the user side.

Step 502: and receiving first feedback information fed back by the user side according to the target commodity package.

Step 503: if the first feedback information is satisfied, adding the target commodity package into the user information of the user side.

Specifically, the generated target commodity package is sent to a user side for display. The user can browse the recommended goods information in the goods package. And then, the user side generates first feedback information according to browsing and evaluation of the user, and reflects the satisfaction degree of the user on the commodity package.

The system receives and analyzes the first feedback information, and if the feedback is satisfied, the target commodity package is directly added into the user information of the user for storage. By collecting the active feedback of the user, the generation effect of the commodity package can be continuously improved, and the accumulation and iteration of the commodity package scheme are realized, so that the commodity package generation mechanism becomes more intelligent.

Step 504: if the first feedback information is unsatisfactory, generating a second commodity packet according to unsatisfactory commodities in the first feedback information, sending the second commodity packet to the user side, and receiving second feedback information fed back by the user side according to the second commodity packet.

Step 505: if the second feedback information fed back by the user side according to the second commodity packet is determined to be unsatisfactory, the second feedback information is taken as the first feedback information, the step of generating the second commodity packet according to the first feedback information is re-executed, and the second commodity packet is sent to the user side until the second feedback information is satisfactory.

When the system receives that the first feedback information fed back by the user side is unsatisfactory, specific goods which are unsatisfactory by the user in the feedback information can be analyzed, then a new second goods package is generated according to the unsatisfactory goods, and meanwhile the goods which are satisfactory by the user are reserved to enter the second goods package. And sending the adjusted new commodity package to a user for re-browsing evaluation.

If the second feedback information of the user aiming at the second commodity packet is still unsatisfactory, the system continues to receive the second feedback information and uses the second feedback information as new first feedback information, and the adjustment optimization flow for generating the second commodity packet and acquiring user feedback is repeated. The system will continue to iteratively update the commodity package until the user is satisfied with the newly generated commodity package feedback.

The system can continuously approach to the real demands of users through the polling feedback collection and commodity package adjustment, and the personalized recommendation effect of the commodity package is gradually optimized through a plurality of loops until the users are satisfied. And meanwhile, the capability of the system for adjusting the recommended result is trained and enhanced.

Based on the above embodiment, as an alternative embodiment, in step 504 above: generating a second commodity package according to the unsatisfactory commodity in the first feedback information, wherein the method specifically comprises the following steps:

step 601: the first weight and the second weight are adjusted based on the unsatisfactory merchandise.

By adjusting the weight, the selection probability of the unsatisfied commodity in the commodity package generated later can be reduced, and the repeated recommendation of the unsatisfied commodity is avoided.

Specifically, after the unsatisfactory commodities fed back by the user are obtained, the system can search the association degree between the unsatisfactory commodities and other commodities, namely the first weight. For other products with high relevance to unsatisfactory products, the first weight of the products can be properly reduced, and the influence of the products is reduced.

Meanwhile, the second weight value of the unsatisfied commodity can be directly reduced, so that the unsatisfied commodity is more difficult to select to enter the generated new commodity package. When the related commodity of the unsatisfied commodity and the unsatisfied commodity are regenerated into commodity packages, the calculated result of the degree of correlation is reduced, and the selection probability is also reduced.

The dissatisfaction of the user is fed back by adjusting the weight, so that the commodity package algorithm can be guided to avoid dissatisfaction commodities and related commodities, and a new commodity package can be more in line with the requirements of the user.

Step 602: and deleting unsatisfactory commodities in the target commodity package, and re-executing the step of determining the first recommended commodity associated with the first commodity to obtain the target commodity package.

Specifically, the system first deletes the commodity that the user feedback is unsatisfactory from the originally generated target commodity package. The system re-performs the commodity relevance calculation and optimization selection process, but during this regeneration process, the selection priorities of these unsatisfactory commodities and their related commodities are greatly reduced due to the previous weight adjustment. And finally, a new target commodity package is generated, meanwhile, unsatisfactory commodities of the user are not included, and recommendation of related commodities is avoided.

Step 603: and taking the target commodity package as a second commodity package.

The following are system embodiments of the present application that may be used to perform method embodiments of the present application. For details not disclosed in the system embodiments of the present application, please refer to the method embodiments of the present application.

Referring to fig. 3, a system for generating a commodity packet according to an embodiment of the present application may include: the system comprises a commodity information acquisition module, a first recommended commodity determination module, a first association degree calculation module, a first commodity package generation module and a target commodity package generation module, wherein:

On the basis of the foregoing embodiment, as an optional embodiment, the first commodity packet generating module may further include:

the first commodity package generating unit is used for combining the first recommended commodities according to the interval duration and the first association degree of the first recommended commodities by adopting a knapsack algorithm to obtain the first commodity package, wherein the knapsack capacity of the knapsack algorithm is a preset duration, and the knapsack price of the knapsack algorithm is the first association degree of the first recommended commodities in the first commodity package.

On the basis of the above embodiment, as an optional embodiment, the first commodity packet generating unit may further include: a target recommended commodity determining subunit and a first commodity package generating subunit, wherein:

the target recommended commodity determining subunit is used for substituting the interval duration and the first association degree of each first recommended commodity into a state transition equation of the knapsack algorithm to obtain a target recommended commodity;

wherein, the state transition equation is:

F(i，t)=max{F(i-1，t)，F(i-1，t-Δt[i])+R[i]}；

wherein F (i, t) represents that the sum of the first association degrees of the first i first recommended commodities is maximum within a preset time period, F (i-1, t) represents that the sum of the first association degrees of the first i-1 first recommended commodities can be obtained the maximum within the preset time period when the first i first recommended commodities are not included, deltat [ i ] represents the interval time period of the first i first recommended commodities, and Ri represents the first association degree of the first i first recommended commodities;

the first commodity package generating subunit is configured to combine the target recommended commodities according to the state transition equation and a dynamic planning algorithm to obtain a first commodity package, where the sum of the interval durations of the first recommended commodities in the first commodity package is less than or equal to the preset duration, and the sum of the first relevancy of the first recommended commodities is the largest.

On the basis of the above embodiment, as an optional embodiment, the target commodity packet generating module may further include: the system comprises a second commodity adding unit, a difference duration determining unit, a second commodity adding unit, a second commodity replacing unit and a target commodity package determining unit, wherein:

a second commodity adding unit, configured to determine whether an interval duration exists between a start use time of the second commodity and an estimated start use time of the first recommended commodity, and add the second commodity without the interval duration to the first commodity package;

a difference time length determining unit, configured to determine a difference time length of the first commodity packet;

a second commodity adding unit, configured to add, if there is a second commodity with an interval duration less than or equal to the difference duration, the second commodity with an interval duration less than or equal to the difference duration to the first commodity package;

the second commodity replacing unit is used for replacing the first recommended commodity with the lowest first association degree in the first commodity package with the second commodity with the interval time longer than the difference time length if the second commodity with the interval time longer than the difference time length exists;

and the target commodity package determining unit is used for determining the adjusted first commodity package as the target commodity package.

On the basis of the foregoing embodiment, as an optional embodiment, the first association degree calculating module may further include: the system comprises a user characteristic association degree determining unit, a cost performance association degree determining unit, a first association value determining unit, a second association value determining unit and a first association degree determining unit, wherein:

the user characteristic association degree determining unit is used for obtaining user information and determining the user characteristic association degree of the user and each first recommended commodity according to the user information;

the cost performance association degree determining unit is used for determining the cost performance association degree of the first commodity and each first recommended commodity according to the price information and the evaluation information of the first commodity;

the first association value determining unit is used for multiplying the user characteristic association degree of each first recommended commodity by a first weight to obtain a first association value;

the second association value determining unit is used for multiplying the cost performance association degree by a second weight to obtain a second association value;

and the first association degree determining unit is used for determining the sum of the first association value and the second association value of each first recommended commodity as the first association degree.

On the basis of the above embodiment, as an optional embodiment, the commodity package generating system may further include: the system comprises a target commodity package sending module, a first feedback information receiving module, a target commodity package adding module and a second commodity package generating module, wherein:

the target commodity package sending module is used for sending the target commodity package to a user side;

the first feedback information receiving module is used for receiving first feedback information fed back by the user side according to the target commodity package;

the target commodity package adding module is used for adding the target commodity package to the user information of the user side if the first feedback information is satisfied;

the second commodity package generating module is used for generating a second commodity package according to unsatisfactory commodities in the first feedback information if the first feedback information is unsatisfactory, sending the second commodity package to the user side, and receiving second feedback information fed back by the user side according to the second commodity package;

and the second commodity packet generating module is used for taking the second feedback information as the first feedback information if the second feedback information fed back by the user side according to the second commodity packet is determined to be unsatisfactory, re-executing the step of generating the second commodity packet according to the first feedback information, and sending the second commodity packet to the user side until the second feedback information is satisfactory.

On the basis of the foregoing embodiment, as an optional embodiment, the second commodity packet generating module may further include: the system comprises a weight adjustment unit, a target commodity packet generation unit and a second commodity packet determination unit, wherein:

a weight adjustment unit for adjusting a first weight and a second weight based on the unsatisfactory commodity;

a target commodity package generating unit, configured to delete the unsatisfactory commodity in the target commodity package, and re-execute the step of determining a first recommended commodity associated with the first commodity, to obtain a target commodity package;

and the second commodity package determining unit is used for taking the target commodity package as the second commodity package.

It should be noted that: in the device provided in the above embodiment, when implementing the functions thereof, only the division of the above functional modules is used as an example, in practical application, the above functional allocation may be implemented by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to implement all or part of the functions described above. In addition, the embodiments of the apparatus and the method provided in the foregoing embodiments belong to the same concept, and specific implementation processes of the embodiments of the method are detailed in the method embodiments, which are not repeated herein.

The application also discloses electronic equipment. Referring to fig. 3, fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, at least one communication bus 302.

Wherein the communication bus 302 is used to enable connected communication between these components.

The user interface 303 may include a Display screen (Display), a Camera (Camera), and the optional user interface 303 may further include a standard wired interface, and a wireless interface.

The network interface 304 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), among others.

Wherein the processor 301 may include one or more processing cores. The processor 301 utilizes various interfaces and lines to connect various portions of the overall server, perform various functions of the server and process data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and invoking data stored in the memory 305. Alternatively, the processor 301 may be implemented in hardware in at least one of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programmable Logic Array, PLA). The processor 301 may integrate one or a combination of several of a central processing unit (Central Processing Unit, CPU), an image processor (Graphics Processing Unit, GPU), and a modem etc. The CPU mainly processes an operating system, a user interface diagram, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 301 and may be implemented by a single chip.

The Memory 305 may include a random access Memory (Random Access Memory, RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). Memory 305 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 305 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-described respective method embodiments, etc.; the storage data area may store data or the like involved in the above respective method embodiments. Memory 305 may also optionally be at least one storage device located remotely from the aforementioned processor 301. Referring to fig. 3, an operating system, a network communication module, a user interface module, and an application program of a commodity package generating method may be included in the memory 305 as a computer storage medium.

In the electronic device 300 shown in fig. 3, the user interface 303 is mainly used for providing an input interface for a user, and acquiring data input by the user; and the processor 301 may be configured to invoke an application program in the memory 305 that stores a method of generating a commodity package, which when executed by the one or more processors 301, causes the electronic device 300 to perform the method as described in one or more of the embodiments above. It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all of the preferred embodiments, and that the acts and modules referred to are not necessarily required for the present application.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to related descriptions of other embodiments.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, such as a division of units, merely a division of logic functions, and there may be additional divisions in actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some service interface, device or unit indirect coupling or communication connection, electrical or otherwise.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable memory. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in whole or in part in the form of a software product stored in a memory, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present application. And the aforementioned memory includes: various media capable of storing program codes, such as a U disk, a mobile hard disk, a magnetic disk or an optical disk.

The foregoing is merely exemplary embodiments of the present disclosure and is not intended to limit the scope of the present disclosure. That is, equivalent changes and modifications are contemplated by the teachings of this disclosure, which fall within the scope of the present disclosure. Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure.

This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a scope and spirit of the disclosure being indicated by the claims.

Claims

1. A method of generating a commodity package, comprising:

acquiring a first commodity paid by a user and a second commodity to be paid;

2. The method of claim 1, wherein the combining each of the first recommended products according to each of the interval durations and each of the first associations to generate the first product package comprises:

3. The method of claim 2, wherein the combining each of the first recommended products to obtain the first product package by using a knapsack algorithm according to the interval duration and the first association degree of each of the first recommended products comprises:

wherein, the state transition equation is:

F(i，t)=max{F(i-1，t)，F(i-1，t-Δt[i])+R[i]}；

4. The method for generating a commodity package according to claim 1, wherein said adjusting the recommended commodity in the first commodity package according to the second commodity to obtain the target commodity package includes:

determining a differential time length of the first commodity package;

5. The method of claim 1, wherein determining a first association degree between each of the first recommended products and the first product comprises:

6. The method for generating a commodity package according to claim 1, wherein said adjusting a first recommended commodity in the first commodity package according to the second commodity, after obtaining a target commodity package, further comprises:

the target commodity package is sent to a user side;

7. The method of claim 6, wherein the generating a second package of goods based on the unsatisfactory goods in the first feedback information comprises:

And taking the target commodity package as the second commodity package.

8. A system for generating a commodity package, the system comprising:

9. An electronic device comprising a processor, a memory, a user interface, and a network interface, the memory for storing instructions, the user interface and the network interface for communicating to other devices, the processor for executing the instructions stored in the memory to cause the electronic device to perform the method of any of claims 1-7.

10. A computer readable storage medium storing instructions which, when executed, perform the method of any one of claims 1-7.