Personalized distribution method and system

Abstract

The invention discloses a personalized distribution method and system, and the method comprises the steps: receiving a purchase operation triggered by a user for a preset commodity recommendation list, and generating a first order; extracting first order data from the first order; determining corresponding target data according to the order commodity information and the intention features; sending an order placing instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and obtaining a second order generated by the supply platform based on the order placing instruction; and sending the second order to the client of the user. The method can achieve the deep mining of the actual dynamic demands of the user, meets the dynamic change demands of the user, improves the satisfaction, conversion rate, sales rate and user experience of the customer, and meets the market demands.

Description

Technical Field

[0001] The present invention relates to the technical field of commodity distribution, and in particular to a personalized distribution method and system. Background Art

[0002] Distribution refers to the establishment of sales channels, that is, the sale of products to consumers through certain channels. From this perspective, any sales method can be called distribution. In other words, distribution is the process of moving products from the production location to the sales location. Products must go through a certain distribution method to reach consumers.

[0003] However, with the rapid development of e-commerce and the diversified and dynamic changes in consumer demand, the distribution methods of traditional e-commerce platforms lack in-depth exploration of users' actual dynamic needs, and are unable to further improve customer satisfaction and conversion rates, resulting in low sales rates and poor user experience, making it difficult to meet market demand. Summary of the invention

[0004] In order to solve at least one of the above-mentioned technical problems, the present invention provides a personalized distribution method and system.

[0005] In a first aspect, the present invention provides a personalized distribution method, the method comprising:

[0006] Receiving a purchase operation triggered by a user on a preset product recommendation list, and generating a first order;

[0007] Extracting first order data from the first order; wherein the first order data includes order product information, order address, and intention characteristics;

[0008] Determine corresponding target data according to the order product information and the intention characteristics; wherein the target data includes a target product and a target supplier and a target supply platform corresponding to the target product;

[0009] issuing an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and obtaining a second order generated by the supply platform based on the order instruction;

[0010] sending the second order to a client of the user;

[0011] The receiving of a purchase operation triggered by a user on a preset product recommendation list includes:

[0012] When the user has a history of purchasing operations;

[0013] Generate a preliminary product recommendation list based on the multi-dimensional data of the user;

[0014] Acquire the operation data of the user's last purchase operation, and determine the intention feature corresponding to the operation data according to the operation data;

[0015] The preliminary commodity recommendation list is updated according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list;

[0016] The preset commodity recommendation list is pushed to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

[0017] Preferably, the receiving of a purchase operation triggered by a user on a preset commodity recommendation list further includes:

[0018] When the user has no history of purchase operations;

[0019] Acquire user information, and determine similar user groups based on the user information; wherein the user information includes the user's gender, age, and location;

[0020] Obtain the most recent preset product recommendation list for similar user groups to obtain a preset product recommendation list set;

[0021] Determining a repetition rate of commodities according to the preset commodity recommendation list set;

[0022] Determining the preset commodity recommendation list according to the repetition rate of the commodities;

[0023] The preset product recommendation list is pushed to the user's client, and the preset product recommendation list is dynamically updated in real time according to the user's real-time behavior data to receive purchase operations triggered by the user on the preset product recommendation list or the updated preset product recommendation list.

[0024] Preferably, determining corresponding target data according to the order product information and the intention characteristics includes:

[0025] Determine a number of related product information according to the order product information;

[0026] Respectively determine the associated supply platforms and associated suppliers corresponding to the associated product information;

[0027] According to the intention characteristics, determining a target supply platform and a target supplier from the corresponding associated supply platforms and associated suppliers;

[0028] Determine a target product according to the target supplier, the target supply platform and the intended characteristics;

[0029] Target data is determined according to the target commodity, the target supply platform and the target supplier.

[0030] Preferably, generating a preliminary product recommendation list according to the multi-dimensional data of the user includes:

[0031] Obtaining user behavior data, as well as product data and time data associated with the user behavior data, to obtain multi-dimensional data of the user;

[0032] Calculating user similarity based on the multi-dimensional data by using a preset user-user similarity algorithm;

[0033] Constructing a user-user similarity matrix based on the user similarities, and determining similar users according to the user-user similarity matrix;

[0034] Determine the target user's predicted rating for the product based on the historical ratings of similar users, sort the products based on the predicted ratings, and generate a preliminary product recommendation list.

[0035] Preferably, the calculating user similarity based on the multi-dimensional data by using a preset user-user similarity algorithm includes:

[0036]

[0037] In the formula, p ik and p jk is the rating or behavior data of user i and user j on product k;

[0038] N indicates how many products there are;

[0039] ReLU(x) is the rectified linear unit activation function, which is used to deal with noise and extreme values ​​in the data;

[0040] w k Represents the weight coefficient of item k, which is used to adjust the influence of different items in similarity calculation;

[0041] log(1+p ik ), log(1+p jk ) are the scoring data p ik 、p jk Logarithmic transformation can alleviate the problem of inconsistent rating scales of different users. ik +1, p jk +1 is to avoid undefined logarithm calculation when the score is zero;

[0042] is the time decay factor used to reduce the weight of stale interaction data, λ is the time decay coefficient, Tt ikis the time interval of scoring. As time increases, the weight gradually decreases. T is the current timestamp, which is used for the calculation of time decay. ik The timestamp when user i rates or interacts with item k, which is used to calculate the timeliness of the rating;

[0043] μ is the regularization coefficient, which is used to control the weight of the regularization term;

[0044] ‖p ik ‖ is the score p ik The modulus of is used to represent the regularization term;

[0045] and are the rating normalization items for user i and user j, respectively.

[0046] Preferably, the obtaining of the operation data of the user's last purchase operation and determining the intention feature corresponding to the operation data according to the operation data includes:

[0047] Obtain the user's last order data and the previous order data;

[0048] Determine a target time period according to the timestamp of the last order data and the timestamp of the last order data;

[0049] Obtain all recommended product lists within the target time period to obtain a recommended list set;

[0050] Determine, from the set of recommendation lists, a number of related recommendation lists according to the behavior data generated by the user based on the purchased product information of the last order data; wherein the relevance between the related recommendation lists and the purchased product information is greater than a preset relevance;

[0051] Obtaining browsing information of the user on the related recommendation list to obtain a browsing data set; wherein the browsing information includes the purchased product information, a plurality of similar product information, a first browsing time, and a second browsing time, the similarity between the similar product information and the purchased product information is greater than a preset similarity, the first browsing time is the browsing time of the related recommendation list, and the second browsing time is the browsing time of the purchased product information or the similar product information under the same related recommendation list;

[0052] Determining a first target recommendation list from the plurality of related recommendation lists according to a purchase operation of a previous order;

[0053] Determining one or more second target recommendation lists from the plurality of related recommendation lists according to the purchased product information, the first browsing time, and the second browsing time;

[0054] determining, according to the first target recommendation list and the second target recommendation list, an intention feature corresponding to the operation data;

[0055] The occurrence frequency of the intention feature is determined based on the historical order data, so as to update the preliminary commodity recommendation list based on the occurrence frequency of the intention feature, thereby obtaining the preset commodity recommendation list.

[0056] Preferably, determining the intention feature corresponding to the operation data according to the first target recommendation list and the second target recommendation list includes:

[0057] Obtaining sorting data of various preset feature tags of purchased products from the first target recommendation list and the second target recommendation list respectively; wherein the preset feature tags include one or more of product attributes, purchasing habits, logistics preferences, after-sales service, promotion sensitivity, social and sharing, and personalized content;

[0058] Determine a target feature tag from each of the preset feature tags according to the sorting data;

[0059] The user's intention feature is determined according to the target feature tag.

[0060] Preferably, determining a second target recommendation list from a plurality of related recommendation lists according to the purchased product information, the first browsing time, and the second browsing time includes:

[0061] Determining from the plurality of related recommendation lists that there is a related recommendation list of purchased goods, and obtaining the second target recommendation list;

[0062] Determining whether the relevant recommendation list meets a preset condition according to the first browsing time and the second browsing time;

[0063] If satisfied, the related recommendation list is determined as the second target recommendation list.

[0064] Preferably, the preset conditions include:

[0065] The first browsing time is shorter than a first preset time, and the second browsing time is longer than a second preset time; wherein the first preset time is shorter than the second preset time.

[0066] In a second aspect, the present invention further provides a personalized distribution system, the system comprising:

[0067] An order generation module, used to receive a purchase operation triggered by a user on a preset product recommendation list and generate a first order;

[0068] An extraction module, configured to extract first order data from the first order; wherein the first order data includes order product information, order address, and intention features;

[0069] A determination module, configured to determine corresponding target data according to the order product information and the intention characteristics; wherein the target data includes a target product and a target supplier and a target supply platform corresponding to the target product;

[0070] An order acquisition module, configured to issue an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and to acquire a second order generated by the supply platform based on the order instruction;

[0071] An order sending module, used for sending the second order to the client of the user;

[0072] The receiving of a purchase operation triggered by a user on a preset product recommendation list includes:

[0073] When the user has a history of purchasing operations;

[0074] Generate a preliminary product recommendation list based on the multi-dimensional data of the user;

[0075] Acquire the operation data of the user's last purchase operation, and determine the intention feature corresponding to the operation data according to the operation data;

[0076] The preliminary commodity recommendation list is updated according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list;

[0077] The preset commodity recommendation list is pushed to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] 1) The personalized distribution method provided by the present invention first generates a preliminary product recommendation list according to the user's multi-dimensional data when the user has a historical purchase operation; then obtains the operation data of the user's last purchase operation, determines the intention feature corresponding to the operation data according to the operation data, and after determining the intention feature corresponding to the operation data, updates the preliminary product recommendation list according to the intention feature corresponding to the operation data to obtain a preset product recommendation list; pushes the preset product recommendation list to the user's client to receive the purchase operation triggered by the user on the preset product recommendation list; when the user has a historical purchase operation, the present invention can meet the user's diverse and comprehensive needs by generating a preliminary product recommendation list according to the user's multi-dimensional data, and determines the corresponding intention feature according to the operation data of the user's last purchase operation, and updates the preliminary product recommendation list according to the intention feature corresponding to the operation data to obtain the preset product recommendation list, thereby realizing in-depth mining of the user's actual dynamic needs, meeting the user's dynamically changing needs, and thereby improving customer satisfaction and conversion rate, as well as sales rate and user experience, and meeting market demand.

[0080] 2) extracting the first order data from the first order; wherein the first order data includes order product information, order address and intention characteristics; determining corresponding target data according to the order product information and the intention characteristics; the target data includes the target product and the target supplier and target supply platform corresponding to the target product; the present invention extracts data from the first order, and determines the target product and the target supplier and target supply platform corresponding to the target product according to the extracted data, thereby quickly matching the corresponding supplier and supply platform for the user, determining the target product that meets the user's needs, improving data processing efficiency and realizing the most optimized matching method.

[0081] 3) Send an order instruction to the supply platform to purchase the target product based on the first order data and the target data, and obtain a second order generated by the supply platform based on the order instruction; send the second order to the user's client. By combining the order data and the target data, the optimal order delivery and purchase plan can be quickly and accurately matched, thereby improving the user experience, and ultimately greatly improving customer satisfaction and conversion rate.

[0082] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

[0084] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0085] Figure 1 A flowchart of a personalized distribution method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0086] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0088] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0089] In addition, in order to better illustrate the present invention, numerous specific details are provided in the following specific embodiments. It should be understood by those skilled in the art that the present invention can be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present invention.

[0090] See also Figure 1 , Figure 1 A flowchart of a personalized distribution method provided by an embodiment of the present invention. Figure 1 As shown, a personalized distribution method comprises the following steps:

[0091] S10, receiving a purchase operation triggered by a user on a preset product recommendation list, and generating a first order;

[0092] In this embodiment, by receiving the purchase operation triggered by the user on the preset product recommendation list and analyzing the user's historical behavior, interests, purchase preferences and other multi-dimensional data, the intelligent algorithm can accurately recommend the product that best meets the user's needs. This personalized recommendation not only saves the user's shopping time, but also improves the user's satisfaction and user conversion rate.

[0093] The receiving of a purchase operation triggered by a user on a preset product recommendation list includes:

[0094] S110, when the user has a history of purchase operations;

[0095] S111, generating a preliminary product recommendation list according to the multi-dimensional data of the user; specifically, including:

[0096] S111-1. Obtain user behavior data, as well as product data and time data associated with the user behavior data, to obtain multi-dimensional data of the user; wherein the user behavior data includes:

[0097] Browsing history: product categories and specific products that users have browsed in the past.

[0098] Search history: Keywords that users have searched for on the platform.

[0099] Purchase history: the user’s past purchase records, including the type and brand of goods purchased.

[0100] Purchase habits: the user's past purchase habits, including purchase frequency, purchase amount, purchase time and repurchase behavior. Purchase frequency: the frequency of user purchases. Purchase time: the time period when the user most often shops. Purchase amount: the number of purchases made by the user in a single purchase. Repurchase behavior: whether the user will repeatedly purchase the same or similar products.

[0101] Collection / Follow: Products, stores, etc. that the user has collected or followed.

[0102] Evaluation and feedback: User evaluation and feedback on purchased products.

[0103] User behavior log: click rate, dwell time, etc.

[0104] S111-2, calculating user similarity based on the multi-dimensional data by using a preset user-user similarity algorithm;

[0105] Specifically, the similarity between user i and user j is calculated as follows:

[0106]

[0107] In the formula, p ik and p jk is the rating or behavior data of user i and user j on product k, where it should be noted that the rating refers to the user's explicit rating of the product (e.g., a rating from 1 to 5), and the behavior data refers to the user's implicit behavior (e.g., number of clicks, browsing time, etc.);

[0108] N represents the number of features (or products); the first recommendation result is a collection of products that are liked by users with similar behavior to the user.

[0109] ReLU(x) is the Rectified Linear Unit activation function, defined as max(0,x). It is used to process noise and extreme values ​​in data;

[0110] w k Represents the weight coefficient of item k, which is used to adjust the influence of different items in similarity calculation, such as dynamic calculation based on entropy, item category, user preference and other factors;

[0111] log(1+p ik ) is the score data p ik Logarithmic transformation can alleviate the problem of inconsistent rating scales among different users. jk ) is the score data p jk Logarithmic transformation can alleviate the problem of inconsistent rating scales of different users. ik +1, p jk +1 is to avoid undefined logarithm calculation when the score is zero,

[0112] is the time decay factor used to reduce the weight of stale interaction data, λ is the time decay coefficient, Tt ik is the time interval of scoring. As time increases, the weight gradually decreases. T is the current timestamp, which is used for the calculation of time decay. It can be the time of the prediction moment or the current time of the system. ik The timestamp when user i rates or interacts with item k, which is used to calculate the timeliness of the rating;

[0113] μ is the regularization coefficient, which is used to control the weight of the regularization term to prevent the model from overfitting.

[0114] ‖p ik ‖ is the score p ik The modulus of is used to represent the regularization term. It can be a simple absolute value or a complex function based on the score distribution. The regularization term ‖p ik ‖ is the penalty factor of the score, which is used to avoid excessive deviation of the score data;

[0115] and are the rating normalization terms for user i and user j, respectively. They are used to normalize the ratings to ensure that the rating scales of different users are consistent.

[0116] In this embodiment, by adopting a more complex and accurate recommendation algorithm, the user's behavior and needs are deeply analyzed, so as to calculate a preliminary commodity recommendation list, thereby improving user satisfaction and purchase rate.

[0117] S111-3, constructing a user-user similarity matrix based on the user similarities, and determining similar users according to the user-user similarity matrix;

[0118] S111-4. Determine the target user's predicted rating for the product based on the historical ratings of similar users, sort the products based on the predicted ratings, and generate a preliminary product recommendation list.

[0119] In this embodiment, by obtaining the user rating matrix, item weights, timestamps, time decay coefficients, and regularization coefficients, and performing logarithmic transformation and ReLU activation on the rating data, the time decay weight is calculated, and the similarity between users is calculated based on the weighted ratings, time decay, and regularization terms, and a personalized recommendation list is generated based on the user similarity matrix, a more personalized and dynamic recommendation is achieved. When dealing with problems such as data sparsity and adjustment of historical data weights, the system can better adapt to complex user behavior patterns, thereby providing more accurate and timely recommendation results.

[0120] Among them, through the time decay factor, newer ratings will get greater weights. For example, product 2 has a higher recent rating, so its influence is greater than historical products 1 and 3. It ensures that the user's real-time interest changes can be tracked. Through logarithmic transformation and ReLU activation function, the scale problem of ratings is solved, and the extreme impact of extreme ratings (such as 0 or 5) on similarity calculations is avoided, thereby enhancing the stability of the model. And by introducing regularization terms, the model is prevented from being too dependent on a small number of ratings, which can improve the generalization ability of recommendation results when data is sparse. Through dynamic adjustment of weights, entropy weights can capture users' preferences for different categories of products. For categories that users often buy, the system will give higher weights to enhance the accuracy of recommendations. For example, both user i and user j prefer product 2, so this product will be recommended first. The matrix can better handle items that users have not rated, and predict the preference differences between similar users through potential features.

[0121] As needed, it also includes: constructing a user portrait using a machine learning algorithm based on the multi-dimensional data;

[0122] It should be noted that multi-dimensional data also includes user information, such as age, gender and region. By building a user portrait based on multi-dimensional data, we can accurately obtain the user's specific needs and preferences, so as to make accurate recommendations to the user.

[0123] Among them, user portraits can include demographic characteristics, behavioral characteristics, interest characteristics, consumption capacity and delivery characteristics and corresponding weights, so as to achieve accurate recommendation effects; specifically, demographic characteristics include age, gender, geographic location, etc. Behavioral characteristics include browsing, clicking, searching, and purchasing behaviors. Interest characteristics include extracting the categories and brands of goods that users are interested in based on user behavior and social data. Consumption capacity includes evaluating the user's consumption capacity based on purchase history and amount. Delivery characteristics include the user's preferred delivery speed (such as fast, slow, moderate), delivery packaging requirements, etc. User portrait construction can be achieved through clustering analysis (such as K-means), classification algorithms (such as logistic regression, decision trees), and other methods.

[0124] S112, obtaining the operation data of the user's last purchase operation, and determining the intention feature corresponding to the operation data according to the operation data;

[0125] In this embodiment, since the user's demand changes have diverse and dynamic trend change characteristics, the user's intention characteristics are determined based on the operation data of the user's last purchase behavior, so as to dynamically meet the user's trend change characteristics needs in real time and achieve the effect of accurate push to the user.

[0126] In one embodiment, the obtaining operation data of the user's last purchase operation and determining the intention feature corresponding to the operation data according to the operation data includes:

[0127] S112-1, obtaining the user's last order data and the previous order data;

[0128] S112-2, determining a target time period according to the timestamp of the previous order data and the timestamp of the previous previous order data;

[0129] S112-3, obtaining all recommended lists of commodities within the target time period to obtain a recommended list set;

[0130] It should be noted that the operation data of the user's last purchase operation should include the operation data between the last order and the previous order, where the operation data includes start, close, click, share, comment, forward, view, search term input, search and other data, so as to obtain a list of all product recommendations within the target time period based on the operation data corresponding to the target time period, so as to facilitate the accurate analysis of the intention characteristics corresponding to the operation data, thereby improving the accuracy of recommendations.

[0131] S112-4, determining a plurality of related recommendation lists from the recommendation list set according to the behavior data generated by the user based on the purchased product information of the last order data; wherein the correlation between the related recommendation lists and the purchased product information is greater than a preset correlation;

[0132] S112-5. Obtain browsing information of the user on the related recommendation list to obtain a browsing data set; wherein the browsing information includes the purchased product information, a plurality of similar product information, a first browsing time, and a second browsing time, the similarity between the similar product information and the purchased product information is greater than a preset similarity, the first browsing time is the browsing time of the related recommendation list, and the second browsing time is the browsing time of the purchased product information or the similar product information under the same related recommendation list;

[0133] S112-6, determining a first target recommendation list from the plurality of related recommendation lists according to the purchase operation of the last order;

[0134] S112-7, determining one or more second target recommendation lists from the plurality of related recommendation lists according to the purchased product information, the first browsing time, and the second browsing time; specifically, comprising:

[0135] S112-71, determining from the plurality of related recommendation lists that there is a related recommendation list of purchased goods, and obtaining the second target recommendation list;

[0136] S112-72. Determine whether the related recommendation list meets a preset condition based on the first browsing time and the second browsing time; wherein the preset condition includes: the first browsing time is less than a first preset time, and the second browsing time is greater than a second preset time; wherein the first preset time is less than the second preset time.

[0137] S112-73. If satisfied, determine the related recommendation list as the second target recommendation list.

[0138] S112-8, determining the intention feature corresponding to the operation data according to the first target recommendation list and the second target recommendation list;

[0139] In one embodiment, determining the intention feature corresponding to the operation data according to the first target recommendation list and the second target recommendation list includes:

[0140] S112-81, respectively obtaining the sorting data of each preset feature tag of the purchased goods from the first target recommendation list and the second target recommendation list; wherein the preset feature tag includes one or more of product attributes, logistics preference, after-sales service, promotion sensitivity, social and sharing, and personalized content; specifically,

[0141] Product attributes include category, brand, function and features. Category: the category of products that users are more inclined to buy (such as electronic products, clothing, books, etc.). Brand: the brand preferred by users. Price range: the price range that users are willing to accept. Color and style: the color and style that users like. Function and feature: the product features that users value, such as waterproof performance, material, etc.

[0142] Logistics preferences include delivery speed, delivery method and packaging requirements, among which, delivery speed: the user's requirements for delivery time (such as fast, slow and moderate). Delivery method: the user's preferred delivery method, such as express delivery, self-pickup, etc. Packaging requirements: whether the user has special packaging requirements.

[0143] After-sales service includes return and exchange policy and customer service. Return and exchange policy: the user's attitude and expectations towards returns and exchanges. Customer service: the user's opinion on the customer service response speed and service quality.

[0144] Social and sharing include willingness to share and participation in the community. Willingness to share: whether the user is willing to share shopping experiences with friends or social media. Participation in the community: whether the user actively participates in community activities such as product evaluation and Q&A.

[0145] Promotion sensitivity includes discount sensitivity and holiday promotions. Discount sensitivity refers to the user's reaction to discounts, coupons, and other activities. Holiday promotion refers to whether the user is easily affected by specific holiday promotions.

[0146] Personalized content includes recommendation preference and content preference. Recommendation preference refers to the user’s interest in personalized recommendations. Content preference refers to the type of marketing content that the user prefers, such as pictures, texts, videos, live broadcasts, etc.

[0147] S112-82, determining a target feature tag from each of the preset feature tags according to the sorting data;

[0148] S112-83. Determine the user's intention feature based on the target feature tag.

[0149] S112-9, and determining the frequency of occurrence of the intention feature based on the historical order data, so as to update the preliminary product recommendation list based on the frequency of occurrence of the intention feature, thereby obtaining the preset product recommendation list.

[0150] S113, updating the preliminary commodity recommendation list according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list;

[0151] S114: Push the preset commodity recommendation list to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

[0152] In this embodiment, the preliminary product recommendation list is updated according to the intention characteristics to obtain a preset product recommendation list, thereby achieving accurate recommendations for users, achieving personalized recommendation effects, improving user satisfaction and experience, and thereby improving the overall business sales rate and customer conversion rate.

[0153] In one embodiment, the receiving a purchase operation triggered by a user on a preset product recommendation list further includes:

[0154] S120: When the user has no history of purchase operations;

[0155] S121, obtaining user information, and determining a similar user group according to the user information; wherein the user information includes the user's gender, age, and location;

[0156] S122, obtaining the most recent preset product recommendation list for a similar user group, and obtaining a preset product recommendation list set;

[0157] S123, determining a repetition rate of commodities according to the preset commodity recommendation list set;

[0158] S124, determining the preset commodity recommendation list according to the repetition rate of the commodities;

[0159] S125. Push the preset product recommendation list to the user's client, and dynamically update the preset product recommendation list in real time according to the user's real-time behavior data, so as to receive purchase operations triggered by the user on the preset product recommendation list or the updated preset product recommendation list.

[0160] In this embodiment, when the user has no history of purchasing operations, similar user groups are determined based on user information, and a preset product recommendation list set is obtained based on the most recent preset product recommendation list of the similar user groups, and the repetition rate of the products is determined based on the preset product recommendation list set, and a preset product recommendation list is determined based on the repetition rate of the products, thereby providing the user with recommendations for the most popular and hottest products corresponding to the user information, thereby increasing the purchasing desire of first-time registered users.

[0161] S20, extracting first order data from the first order; wherein the first order data includes product information, order address, and intention characteristics;

[0162] In this embodiment, the first order data is extracted from the first order, so as to determine the corresponding target data according to the commodity information selected by the user and the user's order address and intention characteristics, so as to select a suitable supplier and supply platform for the user to place an order, and distribute the order to the corresponding supplier to reduce the supply pressure of the sales point, thereby reducing the investment in human resources, and such a statistical method is directly linked to the supplier, which can reduce the dwell time of the order, handle the order delivery in time, shorten the order processing cycle, and greatly improve the consumer's user experience. Furthermore, since the corresponding supplier is selected for delivery by analyzing the order data, such a processing method can, to a certain extent, ensure the supply management and recording of the goods by the outlets, improve the management accuracy of the goods, and further shorten the processing time of the order.

[0163] S30: Determine corresponding target data according to the order product information and the intention characteristics, specifically including:

[0164] S31, determining a number of related product information according to the order product information;

[0165] In this embodiment, the corresponding associated product information is quickly determined based on the order product information, so that the corresponding supply platform and supplier can be locked based on the order product information, thereby achieving a fast and accurate ordering effect.

[0166] The order product information includes a first part of information and a second part of information. The first part of information is the product information set by the distributor, and the second part of information is the product information set by the supplier.

[0167] It should be noted that due to the commodity distribution system of the existing e-commerce platform, the data between distributors and suppliers is not in a unified mode, wherein distributors usually modify the commodity information provided by suppliers according to business needs to form their own personalized commodity information. For the same commodity, there may be multiple supply platforms and multiple suppliers for supply, and at the same time, distributors usually sell multiple different types of commodities. At this time, multiple personalized commodity information is stored in the commodity distribution system, resulting in a large amount of storage space occupied. In addition, when the supplier updates the commodity information, it is impossible to synchronously update the commodity data of all distributors, resulting in a large amount of updated data. Therefore, when the present invention receives a user's purchase operation for a commodity, the order commodity information is determined according to the commodity information set by the distributor and the commodity information set by the supplier, so as to quickly determine the supply information of the commodity, and facilitate unified management of the commodity, so as to achieve the effect of quickly and synchronously updating the commodity information.

[0168] S32, respectively determining the associated supply platforms and associated suppliers corresponding to the associated product information;

[0169] S33, determining a target supply platform and a target supplier from the corresponding associated supply platforms and associated suppliers according to the intention characteristics;

[0170] In this embodiment, after the associated product information is determined, the associated supply platform and associated supplier corresponding to the associated product information are obtained, and the corresponding target supply platform and target supplier are quickly locked according to the intention characteristics, thereby improving data processing efficiency.

[0171] S34, determining a target product according to the target supplier, the target supply platform and the intended characteristics;

[0172] S35, determining target data according to the target commodity, the target supply platform and the target supplier;

[0173] In this embodiment, target data is determined based on the target commodity, target supply platform and target supplier, so that an order instruction for purchasing the target commodity can be subsequently issued to the supply platform and supplier based on the target data and the first order data.

[0174] S40, issuing an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and obtaining a second order generated by the supply platform based on the order instruction;

[0175] It should be noted that the first order quantity also includes the business type, material and quantity of the order goods, so that the orders can be distributed to the corresponding suppliers to reduce the supply pressure at the sales point, thereby reducing the investment in human resources. In addition, this statistical method is directly linked to the supplier, which can reduce the residence time of the order, handle the order delivery in time, shorten the order processing cycle, and greatly improve the consumer experience.

[0176] S50, sending the second order to the client of the user;

[0177] In summary, the personalized distribution method provided by the present invention can at least achieve the following effects:

[0178] 1) The personalized distribution method provided by the present invention first generates a preliminary product recommendation list according to the user's multi-dimensional data when the user has a historical purchase operation; then obtains the operation data of the user's last purchase operation, determines the intention feature corresponding to the operation data according to the operation data, and after determining the intention feature corresponding to the operation data, updates the preliminary product recommendation list according to the intention feature corresponding to the operation data to obtain a preset product recommendation list; pushes the preset product recommendation list to the user's client to receive the purchase operation triggered by the user on the preset product recommendation list; when the user has a historical purchase operation, the present invention can meet the user's diverse and comprehensive needs by generating a preliminary product recommendation list according to the user's multi-dimensional data, and determines the corresponding intention feature according to the operation data of the user's last purchase operation, and updates the preliminary product recommendation list according to the intention feature corresponding to the operation data to obtain the preset product recommendation list, thereby realizing in-depth mining of the user's actual dynamic needs, meeting the user's dynamically changing needs, and thereby improving customer satisfaction and conversion rate, as well as sales rate and user experience, and meeting market demand.

[0179] 2) extracting the first order data from the first order; wherein the first order data includes order product information, order address and intention characteristics; determining corresponding target data according to the order product information and the intention characteristics; the target data includes the target product and the target supplier and target supply platform corresponding to the target product; the present invention extracts data from the first order, and determines the target product and the target supplier and target supply platform corresponding to the target product according to the extracted data, thereby quickly matching the corresponding supplier and supply platform for the user, determining the target product that meets the user's needs, improving data processing efficiency and realizing the most optimized matching method.

[0180] 3) Send an order instruction to the supply platform to purchase the target product based on the first order data and the target data, and obtain a second order generated by the supply platform based on the order instruction; send the second order to the user's client. By combining the order data and the target data, the optimal order delivery and purchase plan can be quickly and accurately matched, thereby improving the user experience, and ultimately greatly improving customer satisfaction and conversion rate.

[0181] In some embodiments, the present invention further provides a personalized distribution system, the system comprising:

[0182] An order generation module, used to receive a purchase operation triggered by a user on a preset product recommendation list and generate a first order;

[0183] An extraction module, configured to extract first order data from the first order; wherein the first order data includes order product information, order address, and intention features;

[0184] A determination module, configured to determine corresponding target data according to the order product information and the intention characteristics; wherein the target data includes a target product and a target supplier and a target supply platform corresponding to the target product;

[0185] An order acquisition module, configured to issue an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and to acquire a second order generated by the supply platform based on the order instruction;

[0186] An order sending module, used for sending the second order to the client of the user;

[0187] The receiving of a purchase operation triggered by a user on a preset product recommendation list includes:

[0188] When the user has a history of purchasing operations;

[0189] Generate a preliminary product recommendation list based on the multi-dimensional data of the user;

[0190] Acquire the operation data of the user's last purchase operation, and determine the intention feature corresponding to the operation data according to the operation data;

[0191] The preliminary commodity recommendation list is updated according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list;

[0192] The preset commodity recommendation list is pushed to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

[0193] Wherein, the order generation module includes:

[0194] A list generation submodule, used to generate a preliminary product recommendation list based on the multi-dimensional data of the user;

[0195] A determination submodule, used to obtain the operation data of the user's last purchase operation, and determine the update direction according to the operation data;

[0196] An updating submodule, used to update the preliminary commodity recommendation list according to the updating direction to obtain the preset commodity recommendation list;

[0197] The push submodule is used to push the preset commodity recommendation list to the user's client to receive the purchase operation triggered by the user on the preset commodity recommendation list.

[0198] It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0199] The present invention also provides an electronic device, including a processor and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any possible implementation manner.

[0200] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method as described in any possible implementation manner.

[0201] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0202] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. Those skilled in the art can also clearly understand that the descriptions of the various embodiments of the present invention have different focuses. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, for parts not described or not described in detail in a certain embodiment, refer to the records of other embodiments.

[0203] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0204] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple 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.

[0205] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0206] A person skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by a computer program to instruct the relevant hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The aforementioned storage medium includes: a read-only memory (ROM) or a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

Claims

1. A personalized distribution method, characterized in that: The method comprises: Receiving a purchase operation triggered by a user on a preset product recommendation list, and generating a first order; Extracting first order data from the first order; wherein the first order data includes order product information, order address, and intention characteristics; Determine corresponding target data according to the order product information and the intention characteristics; wherein the target data includes a target product and a target supplier and a target supply platform corresponding to the target product; issuing an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and obtaining a second order generated by the supply platform based on the order instruction; sending the second order to a client of the user; The receiving of a purchase operation triggered by a user on a preset product recommendation list includes: When the user has a history of purchasing operations; Generate a preliminary product recommendation list based on the multi-dimensional data of the user; Acquire the operation data of the user's last purchase operation, and determine the intention feature corresponding to the operation data according to the operation data; The preliminary commodity recommendation list is updated according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list; The preset commodity recommendation list is pushed to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

2. The personalized distribution method according to claim 1, characterized in that: The receiving of a purchase operation triggered by a user on a preset commodity recommendation list further includes: When the user has no history of purchase operations; Acquire user information, and determine similar user groups based on the user information; wherein the user information includes the user's gender, age, and location; Obtain the most recent preset product recommendation list for similar user groups to obtain a preset product recommendation list set; Determining a repetition rate of commodities according to the preset commodity recommendation list set; Determining the preset commodity recommendation list according to the repetition rate of the commodities; The preset product recommendation list is pushed to the user's client, and the preset product recommendation list is dynamically updated in real time according to the user's real-time behavior data to receive purchase operations triggered by the user on the preset product recommendation list or the updated preset product recommendation list.

3. The personalized distribution method according to claim 1, characterized in that: Determining corresponding target data according to the order product information and the intention characteristics includes: Determine a number of related product information according to the order product information; Respectively determine the associated supply platforms and associated suppliers corresponding to the associated product information; According to the intention characteristics, determining a target supply platform and a target supplier from the corresponding associated supply platforms and associated suppliers; Determine a target product according to the target supplier, the target supply platform and the intended characteristics; Target data is determined according to the target commodity, the target supply platform and the target supplier.

4. The personalized distribution method according to claim 1, characterized in that: Generating a preliminary product recommendation list according to the multi-dimensional data of the user includes: Obtaining user behavior data, as well as product data and time data associated with the user behavior data, to obtain multi-dimensional data of the user; Calculating user similarity based on the multi-dimensional data by using a preset user-user similarity algorithm; Building a user-user similarity matrix based on the user similarities, and determining similar users according to the user-user similarity matrix; Determine the target user's predicted rating for the product based on the historical ratings of similar users, sort the products based on the predicted ratings, and generate a preliminary product recommendation list.

5. The personalized distribution method according to claim 4, characterized in that: The calculating the user similarity based on the multi-dimensional data by using a preset user-user similarity algorithm includes: In the formula, p ik and p jk is the rating or behavior data of user i and user j on product k; N indicates how many products there are; ReLU(x) is the rectified linear unit activation function, which is used to deal with noise and extreme values ​​in the data; w k Represents the weight coefficient of item k, which is used to adjust the influence of different items in similarity calculation; log(1+p ik ), log(1+p jk ) are the scoring data p ik 、p jk Logarithmic transformation can alleviate the problem of inconsistent rating scales of different users. ik +1, p jk +1 is to avoid undefined logarithm calculation when the score is zero; is the time decay factor used to reduce the weight of stale interaction data, λ is the time decay coefficient, Tt ik is the time interval of scoring. As time increases, the weight gradually decreases. T is the current timestamp, which is used for the calculation of time decay. ik The timestamp when user i rates or interacts with item k, which is used to calculate the timeliness of the rating; μ is the regularization coefficient, which is used to control the weight of the regularization term; ||p ik ‖ is the score p ik The modulus of is used to represent the regularization term; and are the rating normalization items for user i and user j, respectively.

6. The personalized distribution method according to claim 1, characterized in that: The obtaining the operation data of the user's last purchase operation and determining the intention feature corresponding to the operation data according to the operation data includes: Obtain the user's last order data and the previous order data; Determine a target time period according to the timestamp of the last order data and the timestamp of the last order data; Obtain all recommended product lists within the target time period to obtain a recommended list set; Determine, from the set of recommendation lists, a number of related recommendation lists according to the behavior data generated by the user based on the purchased product information of the last order data; wherein the relevance between the related recommendation lists and the purchased product information is greater than a preset relevance; Obtaining browsing information of the user on the related recommendation list to obtain a browsing data set; wherein the browsing information includes the purchased product information, a plurality of similar product information, a first browsing time, and a second browsing time, the similarity between the similar product information and the purchased product information is greater than a preset similarity, the first browsing time is the browsing time of the related recommendation list, and the second browsing time is the browsing time of the purchased product information or the similar product information under the same related recommendation list; Determining a first target recommendation list from the plurality of related recommendation lists according to a purchase operation of a previous order; Determining one or more second target recommendation lists from the plurality of related recommendation lists according to the purchased product information, the first browsing time, and the second browsing time; determining, according to the first target recommendation list and the second target recommendation list, an intention feature corresponding to the operation data; The occurrence frequency of the intention feature is determined based on the historical order data, so as to update the preliminary commodity recommendation list based on the occurrence frequency of the intention feature, thereby obtaining the preset commodity recommendation list.

7. The personalized distribution method according to claim 6, characterized in that: The determining, according to the first target recommendation list and the second target recommendation list, the intention feature corresponding to the operation data includes: Obtaining sorting data of various preset feature tags of purchased products from the first target recommendation list and the second target recommendation list respectively; wherein the preset feature tags include one or more of product attributes, purchasing habits, logistics preferences, after-sales service, promotion sensitivity, social and sharing, and personalized content; Determine a target feature tag from each of the preset feature tags according to the sorting data; The user's intention feature is determined according to the target feature tag.

8. The personalized distribution method according to claim 6, characterized in that: The determining a second target recommendation list from the plurality of related recommendation lists according to the purchased product information, the first browsing time, and the second browsing time includes: Determining from the plurality of related recommendation lists that there is a related recommendation list of purchased goods, and obtaining the second target recommendation list; Determining whether the relevant recommendation list meets a preset condition according to the first browsing time and the second browsing time; If satisfied, the related recommendation list is determined as the second target recommendation list.

9. The personalized distribution method according to claim 8, characterized in that: The preset conditions include: The first browsing time is shorter than a first preset time, and the second browsing time is longer than a second preset time; wherein the first preset time is shorter than the second preset time.

10. A personalized distribution system, characterized in that: The system comprises: An order generation module, used to receive a purchase operation triggered by a user on a preset product recommendation list and generate a first order; An extraction module, configured to extract first order data from the first order; wherein the first order data includes order product information, order address, and intention features; A determination module, configured to determine corresponding target data according to the order product information and the intention characteristics; wherein the target data includes a target product and a target supplier and a target supply platform corresponding to the target product; An order acquisition module, configured to issue an order instruction for purchasing the target commodity to the supply platform and the supplier according to the first order data and the target data, and to acquire a second order generated by the supply platform based on the order instruction; An order sending module, used for sending the second order to the client of the user; The receiving of a purchase operation triggered by a user on a preset product recommendation list includes: When the user has a history of purchasing operations; Generate a preliminary product recommendation list based on the multi-dimensional data of the user; Acquire the operation data of the user's last purchase operation, and determine the intention feature corresponding to the operation data according to the operation data; The preliminary commodity recommendation list is updated according to the intention feature corresponding to the operation data to obtain the preset commodity recommendation list; The preset commodity recommendation list is pushed to the client of the user to receive a purchase operation triggered by the user on the preset commodity recommendation list.

Images