







技术领域technical field
本说明书涉及信息处理领域,尤其涉及一种基于支付行为的信息推荐方法和系统。This specification relates to the field of information processing, in particular to an information recommendation method and system based on payment behavior.
背景技术Background technique
用户在使用终端支付的过程中,支付平台(如支付宝)为了增加用户的支付意愿、促进支付成功、或是为了增加支付平台中第三方服务平台的引流,可以向该用户的终端推荐权益等信息。比如,向用户终端推荐“本单支付可得现金0.8元”、“本单支付可返1元话费红包”,等等。而如何向目标用户推荐权益,才能提高用户的支付转化率,使用户最大概率上成功完成对当前交易订单的支付,是一个亟需解决的问题。During the payment process of the user using the terminal, the payment platform (such as Alipay) can recommend information such as rights and interests to the user's terminal in order to increase the user's willingness to pay, promote payment success, or increase the drainage of third-party service platforms in the payment platform. . For example, recommend to the user terminal that "you can get 0.8 yuan in cash when you pay this order", "you can get a red packet of 1 yuan when you pay this order", and so on. How to recommend rights and interests to target users in order to improve the user's payment conversion rate and enable users to successfully complete the payment of the current transaction order with the highest probability is an urgent problem that needs to be solved.
因此,需要提供一种基于支付行为的信息推荐方法,以提高用户的支付转化率。Therefore, it is necessary to provide an information recommendation method based on payment behavior to improve the payment conversion rate of users.
发明内容Contents of the invention
本说明书提供的基于支付行为的信息推荐方法和系统,可以提高用户的支付转化率。The payment behavior-based information recommendation method and system provided in this manual can improve the payment conversion rate of users.
第一方面,本说明书提供一种基于支付行为的信息推荐方法,应用于支付平台,所述方法包括:接收目标用户通过目标终端发起的支付请求;确定与所述目标用户的用户画像匹配的推荐信息集合,所述推荐信息集合包括至少一个配置平台下的推荐信息,所述推荐信息表征完成所述支付请求对应的支付行为可获得的权益类型及其对应的权益值,所述用户画像包括所述目标用户的多个特征标签,其中,所述权益值是以总权益值约束为约束条件、以总支付转化率最大为约束目标计算得到的;以及从所述推荐信息集合中确定目标推荐信息并发送给所述目标终端。In the first aspect, this specification provides an information recommendation method based on payment behavior, which is applied to a payment platform. The method includes: receiving a payment request initiated by a target user through a target terminal; determining a recommendation that matches the user portrait of the target user information set, the recommendation information set includes at least one recommendation information under the configuration platform, the recommendation information represents the type of rights and interests that can be obtained by completing the payment behavior corresponding to the payment request and the corresponding rights and interests value, and the user portrait includes the A plurality of feature tags of the target user, wherein the equity value is calculated with the constraint of the total equity value as the constraint condition and the maximum total payment conversion rate as the constraint objective; and determining the target recommendation information from the recommendation information set and sent to the target terminal.
在一些实施例中,所述总权益值约束包括第一约束,包括对所述至少一个配置平台中的每个配置平台:与当前配置平台对应的权益类型集合相匹配的用户集合的权益值的和小于第一总预算,所述用户集合包括所述目标用户。In some embodiments, the total equity value constraint includes a first constraint, including for each configuration platform in the at least one configuration platform: the equity value of the user set that matches the equity type set corresponding to the current configuration platform and are less than the first total budget, the user set includes the target user.
在一些实施例中,所述总权益值约束包括第二约束,包括对所述权益类型集合中的每个权益类型:与当前权益类型相匹配的子用户集合的权益值的和小于第二总预算,所述用户集合包括所述子用户集合。In some embodiments, the total equity value constraint includes a second constraint, including for each equity type in the equity type set: the sum of the equity values of the sub-user sets matching the current equity type is less than the second total budget, the user set includes the sub-user set.
在一些实施例中,所述总支付转化率包括对所述每个配置平台:所述用户集合的权益值对应的支付转化率的和。In some embodiments, the total payment conversion rate includes a sum of payment conversion rates corresponding to each configuration platform: the equity value of the user set.
在一些实施例中,所述权益值是通过以下方法得到的:确定与所述用户集合中的每个用户匹配的至少一个权益类型,及所述至少一个权益类型中每个权益类型对应的多个候选权益值;以所述总权益值约束为约束条件,以所述总支付转化率最大为约束目标,对所述每个用户的所述每个权益类型,从所述多个候选权益值中确定一个为其对应的权益值。In some embodiments, the equity value is obtained by determining at least one equity type that matches each user in the user set, and the multiple candidate equity values; with the total equity value constraint as the constraint condition and the maximum total payment conversion rate as the constraint target, for the each equity type of each user, from the plurality of candidate equity values Determine one of its corresponding equity values.
在一些实施例中,确定所述多个候选权益值包括:获得所述至少一个权益类型中每个权益类型对应的多个权益值区间,所述权益值区间包括多个基础权益值,所述多个权益值区间之间按照所述基础权益值的大小顺序排列;以及从所述多个权益值区间的每个权益值区间内确定一个所述候选权益值,以得到所述多个候选权益值。In some embodiments, determining the plurality of candidate equity values includes: obtaining multiple equity value intervals corresponding to each equity type in the at least one equity type, the equity value intervals including a plurality of basic equity values, the A plurality of equity value intervals are arranged according to the order of the size of the basic equity value; and one of the candidate equity values is determined from each equity value interval of the plurality of equity value intervals to obtain the plurality of candidate equity values value.
在一些实施例中,在所述每个权益值区间内,所述候选权益值对应的边际效益大于除所述候选权益值之外的任一非候选权益值对应的边际效益,所述边际效益是所述每个配置平台每增加或减少一单位基础权益值所增加或减少的收益。In some embodiments, within each equity value interval, the marginal benefit corresponding to the candidate equity value is greater than the marginal benefit corresponding to any non-candidate equity value other than the candidate equity value, and the marginal benefit It is the income increased or decreased by each unit of basic equity value increased or decreased by each configuration platform.
在一些实施例中,所述总支付转化率通过以下方法得到:确定所述每个用户的所述每个权益类型对应的支付转化率,包括:确定当前权益类型的所述多个候选权益值中的每个候选权益值对应的候选支付转化率;确定所述当前权益类型的所述权益值对应的候选支付转化率为其对应的支付转化率;将所述用户集合对应的支付转化率的和作为所述总支付转化率。In some embodiments, the total payment conversion rate is obtained by the following method: determining the payment conversion rate corresponding to each equity type of each user includes: determining the multiple candidate equity values of the current equity type The candidate payment conversion rate corresponding to each candidate equity value; determine the candidate payment conversion rate corresponding to the equity value of the current equity type; the payment conversion rate corresponding to the user set and as the total payment conversion rate.
在一些实施例中,所述候选支付转化率通过以下方法得到:确定所述候选支付转化率的初始概率值;基于所述初始概率值与所述候选支付转化率的目标概率值之间的映射关系对所述初始概率值进行预估校准,得到所述目标概率值;以及In some embodiments, the candidate payment conversion rate is obtained through the following methods: determining the initial probability value of the candidate payment conversion rate; based on the mapping between the initial probability value and the target probability value of the candidate payment conversion rate Predicting and calibrating the initial probability value to obtain the target probability value; and
将所述目标概率值确定为所述候选支付转化率。The target probability value is determined as the candidate payment conversion rate.
在一些实施例中,所述推荐信息集合是从原始信息集合中召回得到的,所述原始信息集合与所述用户画像匹配。In some embodiments, the recommended information set is recalled from an original information set, and the original information set matches the user profile.
在一些实施例中,所述召回包括按照如下召回方式中的至少一种召回:召回与召回平台的第一召回条件匹配的第一子推荐信息,所述至少一个配置平台包括所述召回平台;召回所述目标用户的第一支付转化率大于预设转化率阈值的第二子推荐信息;召回包含所述目标用户的偏好权益类型的第三子推荐信息,所述偏好权益类型是基于所述目标用户在第一历史行为数据确定的偏好值大于预设偏好阈值的权益类型;以及召回与第二召回条件匹配的第四子推荐信息,所述第二召回条件基于专家经验数据确定,其中,所述推荐信息集合包括第一子推荐信息、第二子推荐信息、第三子推荐信息以及第四子推荐信息中的至少一种。In some embodiments, the recall includes at least one of the following recall methods: recalling the first sub-recommendation information that matches the first recall condition of the recall platform, and the at least one configuration platform includes the recall platform; Recalling the second sub-recommendation information whose first payment conversion rate of the target user is greater than the preset conversion rate threshold; recalling the third sub-recommendation information containing the target user's preferred equity type, the preferred equity type is based on the The target user's preference value determined in the first historical behavior data is greater than the preset preference threshold; and recalling the fourth sub-recommendation information matching the second recall condition, the second recall condition is determined based on expert experience data, wherein, The recommendation information set includes at least one of the first sub-recommendation information, the second sub-recommendation information, the third sub-recommendation information and the fourth sub-recommendation information.
在一些实施例中,所述推荐信息集合按照所述目标用户对所述推荐信息集合中推荐信息的第一支付转化率的大小顺序进行了排序。In some embodiments, the recommendation information set is sorted according to the order of the target user's first payment conversion rate of the recommendation information in the recommendation information set.
在一些实施例中,所述从所述推荐信息集合中确定目标推荐信息包括:基于所述目标用户的属性数据确定所述目标用户对所述推荐信息集合中推荐信息的第二支付转化率;按照所述第二支付转化率的大小顺序,对所述推荐信息集合中的推荐信息进行排序,得到目标信息集合;以及在所述目标信息集合中确定所述目标推荐信息。In some embodiments, the determining the target recommendation information from the recommendation information set includes: determining the target user's second payment conversion rate for the recommendation information in the recommendation information set based on the attribute data of the target user; Sorting the recommendation information in the recommendation information set according to the order of the second payment conversion rate to obtain a target information set; and determining the target recommendation information in the target information set.
在一些实施例中,所述属性数据包括如下至少一种:所述目标用户第二历史行为数据、所述目标终端所处的地理位置、当前的天气状况、以及所述至少一个配置平台的当前预算。In some embodiments, the attribute data includes at least one of the following: the second historical behavior data of the target user, the geographic location of the target terminal, the current weather condition, and the current status of the at least one configuration platform. Budget.
在一些实施例中,所述目标推荐信息的数量为一条或多条。In some embodiments, the number of the target recommendation information is one or more.
在一些实施例中,所述权益类型包括现金、积分、话费红包、能量、饲料以及道具中的至少一种类型。In some embodiments, the benefit type includes at least one type of cash, points, red envelopes, energy, feed and props.
在一些实施例中,所述发送给所述目标终端包括:指示所述目标终端在支付页面上显示所述目标推荐信息,其中,所述支付页面包括支付前置页、支付首页、退出挽回页、失败挽回页、支付结果页、支付后推荐页中的至少一种。In some embodiments, the sending to the target terminal includes: instructing the target terminal to display the target recommendation information on a payment page, wherein the payment page includes a payment front page, a payment home page, and an exit recovery page. , failure recovery page, payment result page, and at least one of post-payment recommendation pages.
在一些实施例中,所述指示所述目标终端在支付页面上显示所述目标推荐信息包括,指示所述目标终端:在所述支付页面上显示所述目标推荐信息的获取方式,其中,所述获取方式包括支付立减和支付返中的至少一种。In some embodiments, the instructing the target terminal to display the target recommendation information on the payment page includes instructing the target terminal to: display the acquisition method of the target recommendation information on the payment page, wherein the The acquisition method includes at least one of payment instant discount and payment rebate.
第二方面,本说明书还提供一种基于支付行为的信息推荐系统,应用于支付平台,包括至少一个存储介质,存储有至少一组指令集用于实现所述信息推荐;以及至少一个处理器,同所述至少一个存储介质通信连接,其中当所述信息推荐系统运行时,所述至少一个处理器读取所述至少一个指令集并实施第一方面所述的信息推荐方法。In the second aspect, this specification also provides an information recommendation system based on payment behavior, applied to a payment platform, including at least one storage medium storing at least one set of instructions for implementing the information recommendation; and at least one processor, Communicatively connected with the at least one storage medium, wherein when the information recommendation system is running, the at least one processor reads the at least one instruction set and implements the information recommendation method described in the first aspect.
由以上技术方案可知,本说明书提供的基于支付行为的信息推荐方法,支付平台接收目标用户在目标终端上发起的支付请求后,为目标用户确定匹配的推荐信息集合。在该推荐信息集合中,推荐信息的权益值是以总权益约束为约束条件、以总支付转化率最大为约束目标计算得到,也即是,支付平台在权益定价时给出了最低成本控制下转化率最大的定价,因此基于该推荐信息集合为目标用户确定的目标推荐信息,能够提升目标用户的支付转化率,降低陪跑成本。It can be seen from the above technical solutions that in the payment behavior-based information recommendation method provided in this specification, the payment platform determines a matching recommendation information set for the target user after receiving the payment request initiated by the target user on the target terminal. In this recommended information set, the equity value of the recommended information is calculated based on the constraint of the total equity and the maximum total payment conversion rate as the constraint objective. The pricing with the highest conversion rate, so the target recommendation information determined for the target user based on the recommendation information set can improve the payment conversion rate of the target user and reduce the cost of accompanying running.
本说明书提供的基于支付行为的信息推荐方法和系统的其他功能将在以下说明中部分列出。根据描述,以下数字和示例介绍的内容将对那些本领域的普通技术人员显而易见。本说明书提供的基于支付行为的信息推荐方法和系统的创造性方面可以通过实践或使用下面详细示例中所述的方法、装置和组合得到充分解释。The payment behavior-based information recommendation method and other functions of the system provided in this manual will be partially listed in the following description. The following figures and example presentations will be apparent to those of ordinary skill in the art from the description. The inventive aspects of the payment behavior-based information recommendation method and system provided in this specification can be fully explained by practicing or using the methods, devices and combinations described in the following detailed examples.
附图说明Description of drawings
为了更清楚地说明本说明书实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本说明书的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of this specification. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort.
图1示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐方法的应用场景示意图;Fig. 1 shows a schematic diagram of an application scenario of an information recommendation method based on payment behavior provided according to some embodiments of this specification;
图2示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐系统的示意图;Fig. 2 shows a schematic diagram of an information recommendation system based on payment behavior provided according to some embodiments of this specification;
图3示出了根据本说明书的一些实施例提供的一种计算设备的硬件结构图;Fig. 3 shows a hardware structure diagram of a computing device provided according to some embodiments of this specification;
图4示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐方法的流程图;Fig. 4 shows a flow chart of an information recommendation method based on payment behavior provided according to some embodiments of this specification;
图5示出了根据本说明书的一些实施例提供的一种支付前置页的示意图;Fig. 5 shows a schematic diagram of a payment front page provided according to some embodiments of this specification;
图6示出了根据本说明书的一些实施例提供的一种权益定价方法的流程图;Fig. 6 shows a flowchart of an equity pricing method provided according to some embodiments of this specification;
图7示出了根据本说明书的一些实施例提供的在支付首页上显示目标推荐信息的示意图;以及Fig. 7 shows a schematic diagram of displaying target recommendation information on the payment homepage according to some embodiments of this specification; and
图8示出了根据本说明书的一些实施例提供的一种信息推荐的示意框图。Fig. 8 shows a schematic block diagram of information recommendation provided according to some embodiments of this specification.
具体实施方式Detailed ways
以下描述提供了本说明书的特定应用场景和要求,目的是使本领域技术人员能够制造和使用本说明书中的内容。对于本领域技术人员来说,对所公开的实施例的各种局部修改是显而易见的,并且在不脱离本说明书的精神和范围的情况下,可以将这里定义的一般原理应用于其他实施例和应用。因此,本说明书不限于所示的实施例,而是与权利要求一致的最宽范围。The following description provides specific application scenarios and requirements of this specification, with the purpose of enabling those skilled in the art to manufacture and use the contents of this specification. Various local modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and embodiments without departing from the spirit and scope of the description. application. Accordingly, the description is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the claims.
这里使用的术语仅用于描述特定示例实施例的目的,而不是限制性的。比如,除非上下文另有明确说明,这里所使用的,单数形式“一”,“一个”和“该”也可以包括复数形式。当在本说明书中使用时,术语“包括”、“包含”和/或“含有”意思是指所关联的整数,步骤、操作、元素和/或组件存在,但不排除一个或多个其他特征、整数、步骤、操作、元素、组件和/或组的存在或在该系统/方法中可以添加其他特征、整数、步骤、操作、元素、组件和/或组。The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. For example, as used herein, the singular forms "a", "an" and "the" may also include the plural forms unless the context clearly dictates otherwise. When used in this specification, the terms "comprising", "comprising" and/or "containing" mean that the associated integer, step, operation, element and/or component is present, but does not exclude one or more other features , integers, steps, operations, elements, components and/or groups exist or other features, integers, steps, operations, elements, components and/or groups may be added in the system/method.
考虑到以下描述,本说明书的这些特征和其他特征、以及结构的相关元件的操作和功能、以及部件的组合和制造的经济性可以得到明显提高。参考附图,所有这些形成本说明书的一部分。然而,应该清楚地理解,附图仅用于说明和描述的目的,并不旨在限制本说明书的范围。还应理解,附图未按比例绘制。These and other features of this specification, as well as the operation and function of the relevant elements of the structure, and the economy of assembly and manufacture of the components, may be significantly enhanced in view of the following description. Reference is made to the accompanying drawings, all of which form a part of this specification. It should be clearly understood, however, that the drawings are for purposes of illustration and description only and are not intended to limit the scope of the specification. It should also be understood that the figures are not drawn to scale.
本说明书中使用的流程图示出了根据本说明书中的一些实施例的系统实现的操作。应该清楚地理解,流程图的操作可以不按顺序实现。相反,操作可以以反转顺序或同时实现。此外,可以向流程图添加一个或多个其他操作。可以从流程图中移除一个或多个操作。The flowcharts used in this specification illustrate the operation of system implementations according to some embodiments in this specification. It should be clearly understood that the operations of the flowcharts may be performed out of order. Instead, operations may be performed in reverse order or concurrently. Additionally, one or more other operations can be added to the flowchart. One or more operations can be removed from the flowchart.
在对本说明书具体实施例说明之前,先对本说明书的应用场景进行如下介绍:Before describing the specific embodiments of this specification, the application scenarios of this specification are introduced as follows:
用户使用终端在电子商务平台(简称电商平台)上对某笔订单支付时,需要使用支付平台来完成支付。所述电商平台可以是能够实现支付行为的平台,比如购物平台、外卖平台、打车平台,等等。所述支付平台可以是与电商平台之间通信连接的第三方支付平台,如支付宝,也可以是电商平台自己的电商支付平台。图1示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐方法的应用场景示意图。如图1所示,用户可以通过终端向支付平台发送支付请求。支付平台可以结合该用户所属的人群(如人群1/人群2)的支付转化率为该用户确定推荐信息,即,指示终端在支付页面上显示权益和权益获取方式。其中,所述支付页面可以包括支付前置页、支付首页、退出挽回页、失败挽回页、支付后推荐页、支付结果页中的至少一种。所述权益可以包括权益类型和权益值。所述权益类型可以包括现金、积分、饲料、红包、券、能量、道具中的至少一种。所述权益获取方式可以包括支付立减、支付返、银行充公中的至少一种。比如,支付平台可以指示终端在支付前置页展示本单支付可得现金0.8元、在支付首页展示本单支付可返1元缴费红包和5个积分、在退出挽回页展示本单支付可立减1.88元、在支付失败页展示绑卡并继续支付可得10元红包和/或在支付结果页展示可领取10个积分和5个森林能量。为了便于描述,本说明书将主要描述用户在确认支付之前的信息推荐方法。When a user uses a terminal to pay for an order on an e-commerce platform (referred to as an e-commerce platform), he needs to use a payment platform to complete the payment. The e-commerce platform may be a platform that can realize payment behavior, such as a shopping platform, a food delivery platform, a taxi platform, and the like. The payment platform may be a third-party payment platform communicated with the e-commerce platform, such as Alipay, or it may be the e-commerce payment platform of the e-commerce platform itself. Fig. 1 shows a schematic diagram of an application scenario of an information recommendation method based on payment behavior according to some embodiments of this specification. As shown in Figure 1, the user can send a payment request to the payment platform through the terminal. The payment platform may determine recommendation information for the user based on the payment conversion rate of the group to which the user belongs (such as group 1/group 2), that is, instruct the terminal to display rights and rights acquisition methods on the payment page. Wherein, the payment page may include at least one of a payment front page, a payment home page, an exit recovery page, a failure recovery page, a post-payment recommendation page, and a payment result page. The equity may include an equity type and an equity value. The benefit type may include at least one of cash, points, feed, red envelopes, coupons, energy, and props. The method of acquiring rights and interests may include at least one of payment instant discount, payment rebate, and bank confiscation. For example, the payment platform can instruct the terminal to display the payment of this order on the front page of the payment to get 0.8 yuan in cash, to display the payment of this order on the payment home page to return a red envelope of 1 yuan payment and 5 points, and to display the payment of this order can be immediately received on the exit recovery page. Minus 1.88 yuan, display card binding on the payment failure page and continue to pay to get a 10 yuan red envelope, and/or display on the payment result page to receive 10 points and 5 forest energy. For ease of description, this specification will mainly describe the information recommendation method for the user before confirming the payment.
为了方便描述,本说明书对以下描述将会出现的术语进行解释:For the convenience of description, this manual explains the terms that will appear in the following descriptions:
用户画像:通过对用户的海量信息进行分析而得到高度精炼的特征标签,用多个特征标签可视化地表征该用户而得到的结果。User portrait: A highly refined feature label is obtained by analyzing the massive information of the user, and the result obtained by visually representing the user with multiple feature labels.
权益:泛指用户在支付平台里可获得的包括现金、积分、话费红包、能量、饲料、道具等各种激励或者工具。Rights and interests: generally refers to various incentives or tools that users can obtain in the payment platform, including cash, points, red envelopes for phone calls, energy, feed, props, etc.
召回:从全量信息集合中按照一定要求挑选出符合要求的推荐信息。Recall: Select the recommended information that meets the requirements according to certain requirements from the full amount of information collection.
图2示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐系统001的示意图。如图2所示,系统001可以包括目标用户100和支付平台200。支付平台200包括目标终端210、服务器230以及网络250。Fig. 2 shows a schematic diagram of an
目标用户100可以使用目标终端210对某笔交易订单实施支付行为。The
目标终端210可以检测到目标用户100发起的支付请求。在一些实施例中,所述信息推荐方法可以在服务器230上执行。此时,目标终端210可以将该支付请求通过网络250发送给服务器230。服务器230可以接收该支付请求,并确定与目标用户100的用户画像匹配的推荐信息集合,从而从该推荐信息集合中确定目标推荐信息并通过网络250发送给目标终端210。从而,目标终端210上可以显示所述目标推荐信息,供目标用户100查看。在一些实施例中,所述信息推荐方法可以在目标终端210上执行。此时,当目标终端210检测到目标用户100发起的支付请求时,可以自己确定与目标用户100的用户画像匹配的推荐信息集合,并从该推荐信息集合中确定目标推荐信息且显示,以供目标用户100查看。在一些实施例中,所述信息推荐方法可以部分在目标终端210上执行,部分在服务器230上执行。比如,目标终端210可以将检测到的支付请求通过网络250发送给服务器230。服务器230可以接收该支付请求,并确定与目标用户100的用户画像匹配的推荐信息集合,并将该推荐信息集合通过网络250发送给目标终端210。目标终端210从该推荐信息集合中确定目标推荐信息并显示,以供目标用户100查看。The
当所述信息推荐的方法在目标终端210上执行时,目标终端210可以存储有执行本说明书描述的信息推荐方法的数据或指令,并可以执行或用于执行所述数据或指令。在一些实施例中,目标终端210可以包括具有数据信息处理功能的硬件设备和驱动该硬件设备工作所需必要的程序。如图2所示,目标终端210可以与服务器230进行通信连接。在一些实施例中,服务器230可以与多个目标终端210进行通信连接。在一些实施例中,目标终端210可以通过网络250与服务器230交互,以接收或发送消息等。在一些实施例中,目标终端210可以包括移动设备、平板电脑、笔记本电脑、机动车辆的内置设备或类似内容、支付宝的蜻蜓设备、售货机、售货柜,或其任意组合。在一些实施例中,所述移动设备可包括智能家居设备、智能移动设备、虚拟现实设备、增强现实设备或类似设备,或其任意组合。在一些实施例中,所述智能家居装置可包括智能电视、台式电脑等,或任意组合。在一些实施例中,所述智能移动设备可包括智能手机、个人数字辅助、游戏设备、导航设备等,或其任意组合。在一些实施例中,所述虚拟现实设备或增强现实设备可能包括虚拟现实头盔、虚拟现实眼镜、虚拟现实补丁、增强现实头盔、增强现实眼镜、增强现实补丁或类似内容,或其中的任何组合。例如,所述虚拟现实设备或所述增强现实设备可能包括谷歌眼镜、头戴式显示器、VR等。在一些实施例中,所述机动车中的内置装置可包括车载计算机、车载电视等。在一些实施例中,目标终端210可以包括图像采集设备,用于采集生物图像,比如目标用户100的人脸图像、指纹图像、虹膜图像、视网膜图像、手型图像等。当然,目标终端210还可以采集目标用户100的声音特征、行为特征、步态特征等。所述图像采集设备可以是二维图像采集设备(比如RGB摄像头),也可以是二维图像采集设备(比如RGB摄像头)和深度图像采集设备(比如3D结构光摄像头、激光探测器,等等)。在一些实施例中,目标终端210可以是具有定位技术的设备,用于定位目标终端210的位置。When the information recommendation method is executed on the
在一些实施例中,目标终端210可以安装有一个或多个应用程序(APP)。所述APP能够为目标用户110提供通过网络250同外界交互的能力以及界面。所述APP包括但不限于:网页浏览器类APP程序、搜索类APP程序、聊天类APP程序、购物类APP程序、视频类APP程序、理财类APP程序、即时通信工具、邮箱终端、社交平台软件等等。在一些实施例中,目标终端210上可以安装有目标APP,所述目标APP可以是电商平台对应的APP。目标用户100可以在所述目标APP中发起一笔交易订单,并触发支付请求。支付平台200可以响应于该支付请求时,执行信息推荐方法。所述目标APP比如为购物APP、外卖APP、打车APP,等等。In some embodiments, the
服务器230可以是提供各种服务的服务器,例如对目标终端210上显示的页面提供支持的后台服务器。当所述信息推荐的方法在服务器230上执行时,服务器230可以存储有执行本说明书描述的信息推荐方法的数据或指令,并可以执行或用于执行所述数据或指令。在一些实施例中,服务器230可以包括具有数据信息处理功能的硬件设备和驱动该硬件设备工作所需必要的程序。服务器230可以与多个目标终端210通信连接,并接收目标终端210发送的数据。The
网络250用以在目标终端210和服务器230之间提供通信连接的介质。网络250可以促进信息或数据的交换。如图2所示,目标终端210和服务器230可以同网络250连接,并且通过网络250互相传输信息或数据。在一些实施例中,网络250可以是任何类型的有线或无线网络,也可以是其组合。比如,网络250可以包括电缆网络,有线网络、光纤网络、电信通信网络、内联网、互联网、局域网(LAN)、广域网(WAN)、无线局域网(WLAN)、大都市市区网(MAN)、广域网(WAN)、公用电话交换网(PSTN)、蓝牙网络、ZigBee网络、近场通信(NFC)网络或类似网络。在一些实施例中,网络250可以包括一个或多个网络接入点。例如,网络250可以包括有线或无线网络接入点,如基站或互联网交换点,通过该接入点,目标终端210和服务器230的一个或多个组件可以连接到网络250以交换数据或信息。The
应该理解,图2中的目标终端210、服务器230和网络250的数目仅仅是示意性的。根据实现需要,可以具有任意数目的目标终端210、服务器230和网络250。It should be understood that the numbers of
图3示出了根据本说明书的一些实施例提供的一种计算设备600的硬件结构图。计算设备600可以执行本说明书描述的基于支付行为的信息推荐方法。所述信息推荐方法在本说明书中的其他部分介绍。当所述信息推荐方法在目标终端210上执行时,计算设备600可以是目标终端210。当所述信息推荐方法在服务器230上执行时,计算设备600可以是服务器230。当所述信息推荐方法可以部分在目标终端210上执行,部分在服务器230上执行时,计算设备600可以是目标终端210和服务器230。FIG. 3 shows a hardware structure diagram of a
如图3所示,计算设备600可以包括至少一个存储介质630和至少一个处理器620。在一些实施例中,计算设备600还可以包括通信端口650和内部通信总线610。同时,计算设备600还可以包括I/O组件660。As shown in FIG. 3 ,
内部通信总线610可以连接不同的系统组件,包括存储介质630、处理器620和通信端口650。
I/O组件660支持计算设备600和其他组件之间的输入/输出。I/
通信端口650用于计算设备600同外界的数据通信,比如,通信端口650可以用于计算设备600同网络250之间的数据通信。通信端口650可以是有线通信端口也可以是无线通信端口。The
存储介质630可以包括数据存储装置。所述数据存储装置可以是非暂时性存储介质,也可以是暂时性存储介质。比如,所述数据存储装置可以包括磁盘632、只读存储介质(ROM)634或随机存取存储介质(RAM)636中的一种或多种。存储介质630可以存储有至少一组指令集用于实现基于支付行为的信息推荐。所述指令是计算机程序代码,所述计算机程序代码可以包括执行本说明书提供的基于支付行为的信息推荐方法的程序、例程、对象、组件、数据结构、过程、模块等等。The
至少一个处理器620可以同至少一个存储介质630以及通信端口650通过内部通信总线610通信连接。至少一个处理器620用以执行上述至少一个指令集。当计算设备600运行时,至少一个处理器620可以读取所述至少一个指令集,并且根据所述至少一个指令集的指示,执行本说明书提供的基于支付行为的信息推荐方法。处理器620可以执行基于支付行为的信息推荐方法包含的所有步骤。处理器620可以是一个或多个处理器的形式,在一些实施例中,处理器620可以包括一个或多个硬件处理器,例如微控制器,微处理器,精简指令集计算机(RISC),专用集成电路(ASIC),特定于应用的指令集处理器(ASIP),中心处理单元(CPU),图形处理单元(GPU),物理处理单元(PPU),微控制器单元,数字信号处理器(DSP),现场可编程门阵列(FPGA),高级RISC机器(ARM),可编程逻辑器件(PLD),能够执行一个或多个功能的任何电路或处理器等,或其任何组合。仅仅为了说明问题,在本说明书中计算设备600中仅描述了一个处理器620。然而,应当注意,本说明书中计算设备600还可以包括多个处理器,因此,本说明书中披露的操作和/或方法步骤可以如本说明书所述的由一个处理器执行,也可以由多个处理器联合执行。例如,如果在本说明书中计算设备600的处理器620执行步骤A和步骤B,则应该理解,步骤A和步骤B也可以由两个不同处理器620联合或分开执行(例如,第一处理器执行步骤A,第二处理器执行步骤B,或者第一和第二处理器共同执行步骤A和B)。At least one
图4示出了根据本说明书的一些实施例提供的一种基于支付行为的信息推荐方法P100的流程图。如前所述,目标终端210和/或服务器230可以执行本说明书所述的基于支付行为的信息推荐方法P100。以下以服务器230执行为例描述所述方法P100。如图4所示,所述方法P100可以包括:Fig. 4 shows a flow chart of a payment behavior-based information recommendation method P100 provided according to some embodiments of this specification. As mentioned above, the
S120:接收目标用户100通过目标终端210发起的支付请求。S120: Receive a payment request initiated by the
目标用户100可以通过目标终端210针对电商平台的某笔交易触发支付请求,目标终端210可以在检测到该支付请求时,将其发送给支付平台200的服务器230,从而服务器230接收到该支付请求。The
所述支付请求用于请求对当前交易订单进行支付。在一些实施例中,所述支付请求可以是目标用户100在订单详情页上通过订单提交选项所触发。在一些实施例中,所述支付请求可以是目标用户100在支付前置页上通过确认交易选项所触发。比如,订单详情页上显示购买商品的详细信息和“提交订单”选项,目标用户100点击“提交订单”选项的同时可以触发该支付请求。所述支付前置页可以在目标终端210检测到“提交订单”选项的触发指令后显示。所述支付前置页上可以包括多个候选支付平台的名称和确认选项,当目标用户100从多个候选支付平台中选择一个作为支付平台200时,在点击“确认交易”选项的同时可以发起该支付请求。所述候选支付平台可以包括第三方支付平台,如支付宝,也可以包括电商平台自己的电商支付平台。图5示出了根据本说明书的一些实施例提供的一种支付前置页的示意图。如图5所示,支付前置页显示了三种候选支付平台XX支付、支付宝支付以及A电商平台对应的A电商支付,以及显示了这三种候选支付平台分别对应的确认选项。目标用户100点击了支付宝支付的确认选项,当点击“确认支付50元”的确认交易选项时,便能够触发支付请求。The payment request is used to request payment for the current transaction order. In some embodiments, the payment request may be triggered by the
S140:确定与目标用户100的用户画像匹配的推荐信息集合。S140: Determine a set of recommended information matching the user portrait of the
在一些实施例中,服务器230可以先确定目标用户100的原始信息集合,并从原始信息集合中进行召回,从而得到所述推荐信息集合。在一些实施例中,服务器230也可以将所述原始信息集合作为目标用户100的推荐信息集合,而不执行召回动作,本说明书实施例对此不作限定。In some embodiments, the
如果服务器230从原始信息集合中召回得到推荐信息集合,所述方法S140可以包括确定原始信息集合和召回两个过程。如果服务器230直接将原始信息集合作为推荐信息集合而不执行召回动作,所述方法S140可以包括确定原始信息集合的过程。以下分别介绍确定原始信息集合和召回的过程。If the
(1)确定原始信息集合(1) Determine the original information set
服务器230可以获取目标用户100的用户画像,确定与目标用户100的用户画像匹配的原始信息集合,原始信息集合中可以包括多条推荐信息。其中,目标用户100的用户画像包括目标用户100的多个特征标签,比如,兴趣爱好、使用支付平台200支付的活跃程度、消费水平、购买类型、教育程度、消费信用水平,等等。The
所述原始信息集合可以包括至少一个配置平台下的推荐信息。每个配置平台可以配置与自身相关的多种推荐信息,包括配置推荐信息中的权益类型和权益值。需要说明的是,所述推荐信息还可以包括权益类型和权益值所要显示的位置,即显示在哪个支付页面,如显示在支付前置页、支付首页或者推出挽回页等。所述推荐信息还可以包括权益类型和权益值的获取方式,比如支付立减、支付返等。The original information set may include recommended information under at least one configuration platform. Each configuration platform can configure a variety of recommended information related to itself, including configuring the type and value of equity in the recommended information. It should be noted that the recommendation information may also include the type of equity and the location to display the equity value, that is, which payment page to display on, such as the payment front page, the payment home page, or the redemption page. The recommendation information may also include the type of equity and the method of acquiring the equity value, such as immediate discount on payment, rebate on payment, and the like.
当目标用户100完成支付行为(如目标用户100的账户中的虚拟货币到达电商平台的账户中)并获得该配置平台下的权益后,则可能进入该配置平台中核销该权益,从而实现该配置平台引流的目标。When the
在一些实施例中,所述配置平台可以是相对于目标用户100和支付平台200而言的第三方服务平台。在这种情况下,该第三方服务平台可以包括服务器230,第三方服务平台的服务器230可以与支付平台200的服务器230之间通信连接,以实现该第三方服务平台与支付平台200之间的通信连接。该第三方服务平台可以将配置的推荐信息发送给支付平台200,进而由支付平台200确定是否向目标终端210发送该推荐信息。所述支付平台200比如为支付宝,所述第三方服务平台比如为花呗、蚂蚁森林、蚂蚁庄园、芭芭农场、充值中心、饿了么中的一个或多个。In some embodiments, the configuration platform may be a third-party service platform relative to the
在一些实施例中,所述配置平台可以是支付平台200自身,比如支付平台200为了与其他支付平台200之间竞争,可以配置推荐信息。需要说明的是,所述推荐信息集合基于原始信息集合得来,因此所述推荐信息集合同样可以包括至少一个配置平台下的推荐信息。In some embodiments, the configuration platform may be the
所述推荐信息可以是关于权益的推荐信息,即向目标用户100推荐权益,以通过权益激励目标用户100的支付转化。具体地,所述推荐信息可以表征完成所述支付请求对应的支付行为可获得的权益类型及其对应的权益值。其中,所述权益类型包括现金、积分、话费红包、能量、饲料以及道具中的至少一种类型。每种权益类型可以存在多种取值,比如,现金或话费红包的取值可以为0.1元、0.5元、1元、1.5元、2元等;积分的基础权益值可以为2个、10个、50个等;能量的基础权益值可以为1g(克)、5g、10g等;饲料的基础权益值可以为30g(克)、50g、100g、180g等;道具的基础权益值可以为1个能量保护罩、1个时光加速器、1个能量双击卡等。The recommendation information may be recommendation information about rights and interests, that is, recommend rights and interests to the
由于每种权益类型存在多种取值,为了能在每个配置平台的预算下给每个配置平台最大的支付转化率,服务器230需要进行权益定价。即,当配置平台为其对应的用户集合中的每个用户分配推荐信息时,每个用户的推荐信息中的权益值应该设置多少,才能使得在该配置平台的预算约束(成本约束)下,该配置平台对应的用户集合的支付转化率最大。所述原始信息集合和所述推荐信息集合中权益定价的过程可以相同。Since each equity type has multiple values, in order to give each configuration platform the maximum payment conversion rate under the budget of each configuration platform, the
具体地,每个配置平台可以为与其匹配的每个用户配置一个或多个权益类型,并为每个权益类型定价。比如,用户A经常使用花呗,是花呗所匹配的一个用户,花呗为用户A配置了现金和积分两种权益类型,服务器230可以为用户A分别确定这两种权益类型各自对应的权益值。Specifically, each configuration platform can configure one or more benefit types for each matched user, and set a price for each benefit type. For example, user A often uses Huabei and is a user matched by Huabei. Huabei configures two types of rights and interests for user A: cash and points. The
所述权益定价的方法可以是支付平台200的服务器230执行,也可以是配置平台的服务器230执行。以下以支付平台200的服务器230执行为例描述权益定价。图6示出了根据本说明书的一些实施例提供的一种权益定价方法的流程图。如图6所述,所述权益定价的方法可以包括:The equity pricing method may be executed by the
S210:对于每个配置平台对应的用户集合,服务器230可以确定与该用户集合中的每个用户匹配的至少一个权益类型,及所述至少一个权益类型中每个权益类型对应的多个候选权益值。S210: For the user set corresponding to each configuration platform, the
所述用户集合可以是每个配置平台对应的权益类型集合相匹配的用户集合。所述权益类型集合可以是该配置平台在推荐信息中所配置的权益类型的集合。比如,花呗对应的权益类型集合包括现金、积分、能量。蚂蚁森林对应的权益类型集合包括能量、道具、饲料。在一些实施例中,所述用户集合可以包括开通对应配置平台的用户的集合。在一些实施例中,所述用户集合可以是在当前时间之前的第二预设时段内使用过/进入过对应配置平台的用户的集合,该第二预设时段比如为1年、半年、1个季度、1个月,等等。在一些实施例中,所述用户集合可以是用户画像满足对应配置平台的配置条件的用户的集合,该配置条件比如为使用该配置平台的频率高于预设频率和/或在该配置平台上的消费总额高于预设金额等等。The set of users may be a set of users matching the set of benefit types corresponding to each configuration platform. The set of benefit types may be a set of benefit types configured by the configuration platform in the recommendation information. For example, the set of equity types corresponding to Huabei includes cash, points, and energy. The set of rights and interests corresponding to Ant Forest includes energy, props, and feed. In some embodiments, the user set may include a set of users who subscribe to the corresponding configuration platform. In some embodiments, the set of users may be a set of users who have used/entered the corresponding configuration platform within a second preset period before the current time, such as 1 year, half a year, 1 year quarter, month, etc. In some embodiments, the set of users may be a set of users whose user portraits meet the configuration conditions of the corresponding configuration platform, such as the frequency of using the configuration platform is higher than a preset frequency and/or on the configuration platform The total amount of consumption is higher than the preset amount and so on.
在一些实施例中,服务器230可以获取所述至少一个权益类型中每个权益类型对应的多个权益值区间,所述权益值区间包括多个基础权益值,所述多个权益值区间之间按照所述基础权益值的大小顺序排列。比如,服务器230将花呗下的积分配置给用户A时,获取积分对应的多个权益值区间amt(a1,a2...an,b1,b2...bn,c1,c2,c3...cn)。其中,(a1,a2...an)是一个权益值区间,(b1,b2...bn)是一个权益值区间,(c1,c2,c3...cn)是一个权益值区间,amt代表权益值区间内的基础权益值。该多个权益值区间比如按照从小到大的顺序排序。服务器230可以根据Uplift模型计算出所述用户集合中每个用户对每个基础权益值的增益,也即是计算所述每个用户对每个基础权益值的敏感度,根据敏感度确定所述多个权益值区间。比如,每个权益值区间内包括敏感度大于预设敏感度阈值的基础权益值。In some embodiments, the
进而,服务器230可以从所述多个权益值区间的每个权益值区间内确定一个候选权益值,以得到所述多个候选权益值。在一些实施例中,服务器230可以根据多个权益值区间的随机定价样本表现,在各个权益值区间内选择边际效益最高的基础权益值作为候选权益值,所述边际效益是每个配置平台每增加或减少一单位基础权益值所增加或减少的收益。此时,在每个权益值区间内,所选的候选权益值对应的边际效益大于除候选权益值之外的任一非候选权益值对应的边际效益。比如,服务器230从上述示例的三个权益值区间内选择出三个候选权益值amt`(a2,b4,c3)。在一些实施例中,服务器230还可以不基于权益值区间,而是通过其他方式确定所述多个候选权益值,比如,直接从数据库中获取所述多个候选权益值。在一些实施例中,服务器230还可以根据其他指标从每个权益值区间内选取候选权益值,所述其他指标比如为成本、利润等。Further, the
所述权益定价的方法还可以包括:The method for pricing equity may also include:
S230:以总权益值约束为约束条件,以总支付转化率最大为约束目标,对每个用户的每个权益类型,从多个候选权益值中确定一个为其对应的权益值。S230: Taking the total equity value constraint as the constraint condition and taking the maximum total payment conversion rate as the constraint target, for each equity type of each user, determine one of multiple candidate equity values as its corresponding equity value.
服务器230可以先计算当前权益类型的多个候选权益值中的每个候选权益值对应的候选支付转化率,比如计算候选权益值amt`(a2,b4,c3)各自对应的候选支付转化率p(pa2,pb4,pc3)。在一些实施例中,服务器230可以先确定候选支付转化率的初始概率值,比如s(sa2,sb4,sc3),并获取所述初始概率值与所述候选支付转化率的目标概率值之间的映射关系,比如f(s',p'),进而基于该映射关系对所述初始概率值进行预估校准,得到所述目标概率值,并将所述目标概率值确定为所述候选支付转化率。其中,所述映射关系f(s',p')可以是服务器230预先对随机定价下样本训练得到的训练概率值和样本表现的真实概率值之间的关系进行训练得到。在一些实施例中,服务器230也可以训练候选权益值与候选支付转化率之间的映射关系,并根据该映射关系和候选权益值确定候选支付转化率。The
进而,服务器230可以基于每个候选权益值对应的候选支付转化率分别计算一个总支付转化率。具体地,服务器230可以先为配置平台对应的用户集合中的每个用户的每个权益类型选定一个候选权益值,并获取每个用户对应的所选定的候选权益值的候选支付转化率,这样,所述用户集合中的每个用户都对应一个候选支付转化率,服务器230可以将这些候选支付转化率相加,得到一个总支付转化率。也即是,服务器230可以将所述用户集合对应的支付转化率的和作为所述总支付转化率。Furthermore, the
在一些实施例中,所述总权益值约束可以包括第一约束。所述第一约束可以是:对于每个配置平台而言,与当前配置平台对应的权益类型集合相匹配的用户集合的权益值的和小于第一总预算,所述用户集合包括目标用户100。假设花呗对应的用户集合中的用户数量为2万个,花呗的第一总预算B为1万元,则所述第一约束可以为:花呗为该2万个用户分配的权益值的和小于1万元。In some embodiments, the total equity value constraint may include a first constraint. The first constraint may be: for each configuration platform, the sum of rights and interests values of user sets that match the rights and interests type set corresponding to the current configuration platform is less than the first total budget, and the user set includes the
在一些实施例中,所述总权益值约束可以包括第二约束。所述第二约束可以是:对每个配置平台的权益类型集合中的每个权益类型而言,与当前权益类型相匹配的子用户集合的权益值的和小于第二总预算Bi。所述Bi可以是每个配置平台下第i个权益类型对应的第二总预算。In some embodiments, the total equity value constraint may include a second constraint. The second constraint may be: for each benefit type in the benefit type set of each configuration platform, the sum of the benefit values of the sub-user sets matching the current benefit type is less than the second total budget Bi. The Bi may be the second total budget corresponding to the i-th equity type under each configuration platform.
如前所述,描述了每个配置平台匹配的用户集合,所述用户集合可以包括所述子用户集合。比如,每个配置平台下多个权益类型对应的多个子用户集合的加和可以是所述用户集合。在一些实施例中,所述子用户集合可以是在所述第二预设时段内核销过对应权益类型的用户的集合,比如,话费红包匹配的子用户集合可以是近1个月内核销过该话费红包的用户集合。在一些实施例中,所述子用户集合可以是用户画像满足该权益类型的匹配条件的用户的集合,该匹配条件可以是在该权益类型的相关活动上的花费时长大于预设花费时长、执行所述相关活动的频率高于预设频率和/或执行所述相关活动的消费总额高于预设金额,等等。比如,某用户经常打电话,通话时长、通话次数和/或话费消耗金额比较大,则该用户可以属于话费红包所匹配的子用户集合。再比如,某用户经常玩蚂蚁森林,每天定时收取森林能量,则该用户可以属于能量所匹配的子用户集合。As mentioned above, the user set matched by each configuration platform is described, and the user set may include the sub-user set. For example, the sum of multiple sub-user sets corresponding to multiple benefit types under each configuration platform may be the user set. In some embodiments, the set of sub-users may be a set of users who have canceled the corresponding benefit type within the second preset period of time. The collection of users who have sold the call credit red envelope. In some embodiments, the sub-user set may be a set of users whose user profile meets the matching condition of the benefit type, and the matching condition may be that the time spent on related activities of the benefit type is longer than the preset time spent, execution The frequency of the related activity is higher than a preset frequency and/or the total amount of consumption for performing the related activity is higher than a preset amount, and so on. For example, if a user makes frequent calls, and the call duration, number of calls, and/or call charge consumption are relatively large, the user may belong to the sub-user set matched by the call charge red envelope. For another example, if a user often plays Ant Forest and collects forest energy regularly every day, the user can belong to the sub-user set whose energy matches.
假设花呗下的权益类型——积分对应的子用户集合中的用户数量为5000个,花呗为积分配置的第二总预算Bi为3000元,则所述第二约束可以为:花呗为该5000个用户分配积分的权益值的和小于3000元。Assuming that the type of benefits under Huabei—the number of users in the sub-user set corresponding to points is 5,000, and the second total budget Bi configured by Huabei for points is 3,000 yuan, the second constraint can be: Huabei is The sum of the rights and interests of the 5,000 user-distributed points is less than 3,000 yuan.
需要说明的是,所述总权益值约束可以包括第一约束,可以包括第二约束,也可以既包括第一约束也包括第二约束,本说明书实施例对此不作限定。It should be noted that the total equity value constraint may include the first constraint, may include the second constraint, or may include both the first constraint and the second constraint, which is not limited in this embodiment of the specification.
由于每个候选权益值都可以计算出一个总支付转化率,为了得到最大的总支付转化率,因此本说明书实施例可以总支付转化率最大为约束目标,同时以总权益值约束为约束条件,对每个用户的每个权益类型,从多个候选权益值中确定一个为其对应的权益值。在为每个权益类型从多个候选权益值中确定出所述权益值后,所述总支付转化率包括每个配置平台对应的用户集合的权益值对应的支付转化率的和。Since a total payment conversion rate can be calculated for each candidate equity value, in order to obtain the maximum total payment conversion rate, the embodiment of this specification can take the maximum total payment conversion rate as the constraint goal, and at the same time take the total equity value constraint as the constraint condition, For each equity type of each user, one of the multiple candidate equity values is determined as its corresponding equity value. After the equity value is determined from multiple candidate equity values for each equity type, the total payment conversion rate includes the sum of payment conversion rates corresponding to the equity values of the user set corresponding to each configuration platform.
经过所述权益定价后,原始数据集/推荐数据集中每个配置平台下的推荐信息中的权益值就确定了,且该权益值是以所述总权益值约束为约束条件、以所述总支付转化率最大为约束目标得到的,因此,本说明书能够从权益定价方面给出每个配置平台在最低成本控制下的最大支付转化率。After the equity pricing, the equity value in the recommendation information under each configuration platform in the original data set/recommended data set is determined, and the equity value is constrained by the total equity The maximum payment conversion rate is obtained from the constraint objective. Therefore, this manual can give the maximum payment conversion rate of each configuration platform under the minimum cost control from the perspective of equity pricing.
(2)召回(2) recall
服务器230可以按照一种或多种召回方式从目标用户100的原始信息集合中进行召回,以得到推荐数据集。The
在一些实施例中,服务器230可以召回与召回平台的第一召回条件匹配的第一子推荐信息。所述至少一个配置平台包括所述召回平台,所述召回平台可以是所述至少一个配置平台中部分或全部的配置平台。所述召回平台可以预先按照其第一召回条件训练召回模型,通过召回模型召回所述第一子推荐信息。所述第一召回条件可以是与所述召回平台的活动相关的条件。比如,花呗为目标用户100配置了现金和积分两种权益类型,花呗的第一召回条件可以是召回积分相关的推荐信息,如召回的推荐信息为q1(c1,c3,c4...)。再比如,充值中心为目标用户100配置了话费红包和积分两种权益类型,充值中心的第一召回条件可以是召回话费红包相关的推荐信息,如召回的推荐信息为q2(c2,c5,c7...)。所述第一子推荐信息比如包括q1(c1,c3,c4...)和q2(c2,c5,c7...)。In some embodiments, the
在一些实施例中,服务器230可以召回目标用户100的第一支付转化率大于预设转化率阈值的第二子推荐信息。所述预设转化率阈值比如为50%、60%、75%、80%、95%,第二子推荐信息比如为q3(c1,c2,c5,...)。目标用户100对某推荐信息的第一支付转化率大,表明该推荐信息对目标用户100起正向作用,有助于目标用户100的支付转化。在一些实施例中,服务器230可以训练Uplift模型(增益模型),通过Uplift模型召回第二子推荐信息。通过召回对目标用户100起正向作用的第二子推荐信息,能够降低目标用户100的营销反感度,从而提高支付转化率。In some embodiments, the
在一些实施例中,服务器230可以召回包含目标用户100的偏好权益类型的第三子推荐信息,所述偏好权益类型是基于目标用户100的第一历史行为数据确定的偏好值大于预设偏好阈值的权益类型。所述第一历史行为数据可以是第一预设时段内的支付行为数据,也可以是第一预设次数的支付行为数据。比如,目标用户100在近一天支付或者在近十笔支付中,服务器230给目标用户100的目标终端210推送了充话费5毛的红包,但目标用户100并没有支付而是退出支付,而当推送立减红包时却支付成功,可以表明目标用户100的偏好权益类型是现金,因此可以为目标用户100召回包含现金的第三子推荐信息。所述第三子推荐信息比如为q5(c6,c7,...)。In some embodiments, the
在一些实施例中,服务器230可以召回与第二召回条件匹配的第四子推荐信息,所述第二召回条件基于专家经验数据确定。所述第二召回条件可以与目标用户100的用户画像相关。比如,目标用户100属于低活人群(支付活跃度低的人群)、但经常充话费,基于专家经验数据可以确定所述第二召回条件是召回包含话费红包的推荐信息。所述第三子推荐信息比如为q4(c2,c4,...)。In some embodiments, the
其中,第一子推荐信息、第二子推荐信息、第三子推荐信息以及第四子推荐信息中的任意两种可以包括相同的推荐信息。所述推荐信息集合可以包括第一子推荐信息、第二子推荐信息、第三子推荐信息以及第四子推荐信息中的至少一种。Wherein, any two of the first sub-recommendation information, the second sub-recommendation information, the third sub-recommendation information and the fourth sub-recommendation information may include the same recommendation information. The recommendation information set may include at least one of first sub-recommendation information, second sub-recommendation information, third sub-recommendation information, and fourth sub-recommendation information.
需要说明的是,服务器230可以对经过召回得到的推荐信息集合进行排序。在一些实施例中,所述推荐信息集合按照目标用户100对推荐信息集合中推荐信息的支付转化率的大小顺序进行排序。比如,服务器230对q1(c1,c3,c4...)、q2(c2,c5,c7...)、q3(c1,c2,c5,...)、q4(c2,c4,...)、q5(c6,c7,...)按照所述第一支付转化率从大到小的顺序排列成c'(c1,c2,c3,c4,c5,c6,c7...)。具体地,服务器230可以使用支付转化率模型对推荐信息集合进行排序。It should be noted that the
在一些实施例中,服务器230可以在接收所述支付请求之前,预先为支付用户数据集中的每位用户确定与各自用户画像对应的推荐信息集合,该支付用户数据集中包括目标用户100。进而,服务器230在接收到所述支付请求时,可以直接获取目标用户100的推荐信息集合。其中,所述支付用户数据集可以是支付平台200的注册用户的集合,也可以是在当前时间之前的第一预设时段内登录过支付平台200的用户的集合,该第一预设时段比如为1年、半年、1个季度、1个月,等等。在一些实施例中,服务器230也可以在接收到所述支付请求之后,确定目标用户100对应的推荐信息集合。In some embodiments, before receiving the payment request, the
S160:从所述推荐信息集合中确定目标推荐信息并发送给所述目标终端210。S160: Determine target recommendation information from the recommendation information set and send it to the
为了能够推荐更准确的推荐信息,服务器230可以基于目标用户100的属性数据确定目标用户100对所述推荐信息集合中推荐信息的第二支付转化率,并按照第二支付转化率的大小顺序,对所述推荐信息集合中的推荐信息进行排序,得到目标信息集合。进而,在所述目标信息集合中确定所述目标推荐信息。In order to be able to recommend more accurate recommendation information, the
其中,所述属性数据包括如下至少一种:目标用户100的第二历史行为数据、所述目标终端210所处的地理位置、当前的天气状况、以及所述至少一个配置平台的当前预算。Wherein, the attribute data includes at least one of the following: the second historical behavior data of the
所述第二历史行为数据可以是第二预设时段内的支付行为数据,也可以是第二预设次数的支付行为数据。第二预设时段内和第二预设次数的值可以设置较小,比如,获取目标用户100近1天的第二历史行为数据或者获取近五笔的支付记录,从而可以快速地确定出目标推荐信息,提高信息推荐的效率。对于所述地理位置,比如,目标用户100在发起支付请求时位于乡村,那么服务器230可以不给目标用户100推荐包含单车券的推荐信息。所述当前的天气状况可以是支付平台200在接收到支付请求时的天气状况,比如,当前的天气是雨天,目标用户100点外卖的概率提升,所以服务器230可以向目标用户100推荐包含外卖券的推荐信息。每个配置平台的预算可能会发生变化,所述至少一个配置平台的当前预算可以是支付平台200在接收到支付请求时的预算。比如,所述当前预算比较紧张,服务器230可以向目标用户100推荐权益值较低的推荐信息。The second historical behavior data may be payment behavior data within a second preset period, or payment behavior data for a second preset number of times. The value of the second preset time period and the second preset number of times can be set smaller, for example, to obtain the second historical behavior data of the
需要说明的是,所述目标推荐信息的数量可以是一条或多条。也即是,本说明书实施例可以推荐单推荐信息和组合推荐信息,丰富了信息推荐的推荐形式。It should be noted that the number of target recommendation information may be one or more. That is, the embodiment of this specification can recommend single recommendation information and combined recommendation information, which enriches the recommendation forms of information recommendation.
如前所述,每条推荐信息还可以包括权益类型和权益值所要显示的位置,即显示在哪个支付页面,如显示在支付前置页、支付首页或者推出挽回页等。所述推荐信息还可以包括权益类型和权益值的获取方式,比如支付立减、支付返等。因此,服务器230可以指示目标终端210在所述目标推荐信息对应的支付页面上显示所述目标推荐信息。所述支付页面可以包括支付前置页、支付首页、退出挽回页、失败挽回页、支付结果页、支付后推荐页中的至少一种。如图5所示,在支付前置页上显示目标推荐信息,支付平台200为支付宝,目标推荐信息211为“可立减1元”。图7示出了根据本说明书的一些实施例提供的在支付首页上显示目标推荐信息的示意图,其他支付页面不再展示。如图7所示,目标推荐信息211为“本单可返1元话费红包和5g森林能量”。As mentioned above, each piece of recommendation information may also include the type of equity and the location where the equity value is to be displayed, that is, which payment page it is displayed on, such as the payment front page, the payment home page, or the redemption page. The recommendation information may also include the type of equity and the method of acquiring the equity value, such as immediate discount on payment, rebate on payment, and the like. Therefore, the
在一些实施例中,服务器230可以在接收到所述支付请求后,可以为每种支付页面都确定一个目标推荐信息,每个支付页面的目标推荐信息都可以是以总权益值约束为约束条件、以总支付转化率最大为约束目标确定的。目标终端210显示哪个支付页面,该支付页面上就可以显示对应的目标推荐信息。这样,通过目标用户100的一个支付请求就可以为多种支付页面分别配置目标推荐信息,无需单独一个一个地为每个支付页面配置,提高信息推荐的效率,且信息推荐比较智能化。In some embodiments, after receiving the payment request, the
当目标用户100在通过目标终端210完成所述支付请求对应的支付行为时,就可以获取所述目标推荐信息中的权益类型和权益值,并且可以核销所述目标推荐信息中的权益类型和权益值。When the
在一些实施例中,图8示出了根据本说明书的一些实施例提供的一种信息推荐的示意框图。如图8所示,目标用户100可以通过目标终端210发起支付决策。目标终端210可以向支付平台200咨询推荐信息。支付平台200可以向权益平台咨询推荐信息的获取方式和权益(包括权益类型和权益值)。所述权益平台可以是支付平台200的服务器230,也可以是配置平台的服务器230。权益平台可以先进行权益定价。所述权益定价的过程可以包括定价策略、预估校准、在线规划。所述定价策略用于确定候选权益值,所述预估校准用于确定候选权益值的候选支付转化率,所述在线规划用于在以总权益值约束为约束条件、以总支付转化率最大为约束目标计算总支付转化率,从而得到原始信息集合。权益定价之后可以从原始信息集合中召回得到推荐信息集合,包括平台活动召回、增益模型(Uplift)召回、支付动线历史召回和专家经验召回。所述平台活动召回用于召回第一子推荐信息,所述增益模型召回用于召回第二子推荐信息,所述支付动线历史召回用于召回第三子推荐信息,所述专家经验召回用于召回四子推荐信息。之后可以按照目标用户100的第一支付转化率对推荐信息集合进行精排。再根据目标用户100的属性数据对精排后的推荐信息集合进行重排。进而,从重排后的推荐信息集合中确定目标推荐信息返回到支付平台200,从而支付平台200可以将目标推荐信息发送到目标终端210。In some embodiments, FIG. 8 shows a schematic block diagram of information recommendation provided according to some embodiments of this specification. As shown in FIG. 8 , the
综上所述,本说明书提供的信息推荐方法和系统,支付平台能够为推荐信息集合中的推荐信息给出最低成本控制下转化率最大的定价,因此基于该推荐信息集合为目标用户确定目标推荐信息,能够提升目标用户的支付转化率,降低陪跑成本。To sum up, with the information recommendation method and system provided in this specification, the payment platform can provide the recommendation information in the recommendation information set with the lowest cost control and the highest conversion rate pricing. Therefore, based on the recommendation information set, the target recommendation for the target user is determined. Information can improve the payment conversion rate of target users and reduce the cost of accompanying running.
本说明书另一方面提供一种非暂时性存储介质,存储有至少一组用来进行基于支付行为的信息推荐方法的可执行指令。当所述可执行指令被处理器执行时,所述可执行指令指导所述处理器实施本说明书所述的基于支付行为的信息推荐方法P100的步骤。在一些可能的实施方式中,本说明书的各个方面还可以实现为一种程序产品的形式,其包括程序代码。当所述程序产品在基于支付行为的信息推荐系统001上运行时,所述程序代码用于使基于支付行为的信息推荐系统001执行本说明书描述的基于支付行为的信息推荐方法P100的步骤。用于实现上述方法的程序产品可以采用便携式紧凑盘只读存储器(CD-ROM)包括程序代码,并可以在基于支付行为的信息推荐系统001上运行。然而,本说明书的程序产品不限于此,在本说明书中,可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统使用或者与其结合使用。所述程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以为但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件,或者任意以上的组合。可读存储介质的更具体的例子包括:具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。所述计算机可读存储介质可以包括在基带中或者作为载波一部分传播的数据信号,其中承载了可读程序代码。这种传播的数据信号可以采用多种形式,包括但不限于电磁信号、光信号或上述的任意合适的组合。可读存储介质还可以是可读存储介质以外的任何可读介质,该可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。可读存储介质上包含的程序代码可以用任何适当的介质传输,包括但不限于无线、有线、光缆、RF等等,或者上述的任意合适的组合。可以以一种或多种程序设计语言的任意组合来编写用于执行本说明书操作的程序代码,所述程序设计语言包括面向对象的程序设计语言—诸如Java、C++等,还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在基于支付行为的信息推荐系统001上执行、部分地在基于支付行为的信息推荐系统001上执行、作为一个独立的软件包执行、部分在基于支付行为的信息推荐系统001上部分在远程计算设备上执行、或者完全在远程计算设备上执行。Another aspect of the present specification provides a non-transitory storage medium storing at least one set of executable instructions for performing an information recommendation method based on payment behavior. When the executable instruction is executed by the processor, the executable instruction instructs the processor to implement the steps of the payment behavior-based information recommendation method P100 described in this specification. In some possible implementations, various aspects of this specification can also be implemented in the form of a program product, which includes program codes. When the program product runs on the payment behavior-based
上述对本说明书特定实施例进行了描述。其他实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者是可能有利的。The foregoing describes specific embodiments of this specification. Other implementations are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or advantageous in certain embodiments.
综上所述,在阅读本详细公开内容之后,本领域技术人员可以明白,前述详细公开内容可以仅以示例的方式呈现,并且可以不是限制性的。尽管这里没有明确说明,本领域技术人员可以理解本说明书需求囊括对实施例的各种合理改变,改进和修改。这些改变,改进和修改旨在由本说明书提出,并且在本说明书的示例性实施例的精神和范围内。To sum up, after reading this detailed disclosure, those skilled in the art can understand that the foregoing detailed disclosure may be presented by way of example only, and may not be restrictive. Although not explicitly stated here, those skilled in the art can understand that this description needs to cover various reasonable changes, improvements and modifications to the embodiments. Such alterations, improvements and modifications are intended to be suggested by this description, and are within the spirit and scope of the exemplary embodiments of this description.
此外,本说明书中的某些术语已被用于描述本说明书的实施例。例如,“一个实施例”,“实施例”和/或“一些实施例”意味着结合该实施例描述的特定特征,结构或特性可以包括在本说明书的至少一个实施例中。因此,可以强调并且应当理解,在本说明书的各个部分中对“实施例”或“一个实施例”或“替代实施例”的两个或更多个引用不一定都指代相同的实施例。此外,特定特征,结构或特性可以在本说明书的一个或多个实施例中适当地组合。In addition, certain terms in this specification have been used to describe the embodiments of this specification. For example, "one embodiment," "an embodiment" and/or "some embodiments" means that a particular feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. Thus, it is emphasized and should be understood that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various parts of this specification are not necessarily all referring to the same embodiment. Also, specific features, structures or characteristics may be properly combined in one or more embodiments of the present specification.
应当理解,在本说明书的实施例的前述描述中,为了帮助理解一个特征,出于简化本说明书的目的,本说明书将各种特征组合在单个实施例、附图或其描述中。然而,这并不是说这些特征的组合是必须的,本领域技术人员在阅读本说明书的时候完全有可能将其中一部分设备标注出来作为单独的实施例来理解。也就是说,本说明书中的实施例也可以理解为多个次级实施例的整合。而每个次级实施例的内容在于少于单个前述公开实施例的所有特征的时候也是成立的。It should be understood that in the foregoing description of the embodiments of the present specification, in order to facilitate the understanding of one feature, for the purpose of simplifying the present specification, the present specification combines various features in a single embodiment, drawing or description thereof. However, this does not mean that the combination of these features is necessary, and it is entirely possible for those skilled in the art to label some of the devices as separate embodiments when reading this specification. That is to say, the embodiments in this specification can also be understood as the integration of multiple sub-embodiments. It is also true that each sub-embodiment lies in less than all features of a single preceding disclosed embodiment.
本文引用的每个专利,专利申请,专利申请的出版物和其他材料,例如文章、书籍、说明书、出版物、文件、物品等,除了与其相关的任何历史起诉文件、可能与本文件不一致或相冲突的任何相同的、或者任何可能对权利要求的最宽范围具有限制性影响的任何相同的历史起诉文件,均可以通过引用结合于此,并用于现在或以后与本文件相关联的所有目的。此外,如果在与任何所包含的材料相关联的术语的描述、定义和/或使用与本文档相关的术语、描述、定义和/或之间存在任何不一致或冲突时,使用本文件中的术语为准。Each patent, patent application, publication of a patent application, and other materials, such as articles, books, specifications, publications, documents, articles, etc., cited herein, except for any historical prosecution documents to which it relates, which may be inconsistent with or Any identities that conflict, or any identities that may have a restrictive effect on the broadest scope of the claims, are hereby incorporated by reference for all purposes now or hereafter associated with this document. Furthermore, in the event of any inconsistency or conflict between the description, definition, and/or use of a term associated with any contained material, the term used in this document shall prevail.
最后,应理解,本文公开的申请的实施方案是对本说明书的实施方案的原理的说明。其他修改后的实施例也在本说明书的范围内。因此,本说明书披露的实施例仅仅作为示例而非限制。本领域技术人员可以根据本说明书中的实施例采取替代配置来实现本说明书中的申请。因此,本说明书的实施例不限于申请中被精确地描述过的实施例。Finally, it is to be understood that the embodiments of the application disclosed herein are illustrative of the principles of the embodiments of the specification. Other modified embodiments are also within the scope of this specification. Therefore, the embodiments disclosed in this specification are only illustrative and not restrictive. Those skilled in the art may implement the applications in this specification by adopting alternative configurations according to the embodiments in this specification. Therefore, the embodiments of the present specification are not limited to the embodiments precisely described in the application.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310391485.0ACN116433287A (en) | 2023-04-13 | 2023-04-13 | A method and system for information recommendation based on payment behavior |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310391485.0ACN116433287A (en) | 2023-04-13 | 2023-04-13 | A method and system for information recommendation based on payment behavior |
| Publication Number | Publication Date |
|---|---|
| CN116433287Atrue CN116433287A (en) | 2023-07-14 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310391485.0APendingCN116433287A (en) | 2023-04-13 | 2023-04-13 | A method and system for information recommendation based on payment behavior |
| Country | Link |
|---|---|
| CN (1) | CN116433287A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118569852A (en)* | 2024-07-29 | 2024-08-30 | 支付宝(杭州)信息技术有限公司 | Equity processing method and device based on IoT device |
| CN119248136A (en)* | 2024-09-18 | 2025-01-03 | 北京字跳网络技术有限公司 | Method, device, equipment and storage medium for electronic payment |
| CN119831651A (en)* | 2024-12-10 | 2025-04-15 | 天翼电子商务有限公司 | Payment score determining method and device and electronic equipment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118569852A (en)* | 2024-07-29 | 2024-08-30 | 支付宝(杭州)信息技术有限公司 | Equity processing method and device based on IoT device |
| CN119248136A (en)* | 2024-09-18 | 2025-01-03 | 北京字跳网络技术有限公司 | Method, device, equipment and storage medium for electronic payment |
| CN119831651A (en)* | 2024-12-10 | 2025-04-15 | 天翼电子商务有限公司 | Payment score determining method and device and electronic equipment |
| CN119831651B (en)* | 2024-12-10 | 2025-09-30 | 天翼支付科技有限公司 | Payment score determination method, device and electronic equipment |
| Publication | Publication Date | Title |
|---|---|---|
| US11830041B2 (en) | System and methods for generating dynamic market pricing for use in real-time auctions | |
| US11727452B1 (en) | Invoice financing and repayment | |
| CN116433287A (en) | A method and system for information recommendation based on payment behavior | |
| US10963915B2 (en) | Machine-learning based systems and methods for optimizing search engine results | |
| US9892458B1 (en) | Invoice financing and repayment | |
| US11599904B2 (en) | Method, apparatus, and computer program product for predicting consumer behavior | |
| CN106779641A (en) | Information processing method and information processor | |
| WO2019183156A1 (en) | Machine-learning based systems and methods for optimizing search engine results | |
| US20230092475A1 (en) | Information processing method, apparatus and device, and storage medium | |
| KR102704710B1 (en) | Method and apparatus for providing advertising cost-based matching service for marketing | |
| US20160042416A1 (en) | Method and system for collecting commodity consumption data and a method and system for generating an offer | |
| KR20210053162A (en) | System for providing pre-paying of online seller and method thereof | |
| CN112819547B (en) | Information recommendation method and device | |
| KR101601880B1 (en) | Credit Card Marketing Method and System using Personal Expenditure Statement | |
| CN113011912A (en) | Media information processing method, device, electronic equipment and storage medium | |
| CN115358805A (en) | Meta universe-based shop operation and maintenance management method, electronic equipment and storage medium | |
| US12211063B2 (en) | Linking merchant transactions with AR ads | |
| KR102531363B1 (en) | Card discount prediction device and method | |
| CN118013116A (en) | Search result display method and device and electronic equipment | |
| WO2019212843A1 (en) | Machine-learning based systems and methods for optimizing search engine results | |
| TWM578433U (en) | Intellengent customer data processing system and conversational setting system | |
| CN113222566B (en) | A payment management system, method, device and storage medium | |
| CN109858952A (en) | Data processing method and device under service scenarios | |
| CN113450127B (en) | Information display method, information display device, computer equipment and storage medium | |
| CN114254206A (en) | Method and device for generating information recommendation strategy, storage medium, and computing device |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |