CN113379511A

CN113379511A - Method and apparatus for outputting information

Info

Publication number: CN113379511A
Application number: CN202110749450.0A
Authority: CN
Inventors: 张秀军
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Wodong Tianjun Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Wodong Tianjun Information Technology Co Ltd
Priority date: 2021-07-02
Filing date: 2021-07-02
Publication date: 2021-09-10

Abstract

The embodiment of the disclosure discloses a method and a device for outputting information. The specific implementation mode of the method comprises the following steps: responding to the detected that the user accesses the home page, and acquiring historical behavior data of the user; marking a user label for the user according to the historical behavior data; acquiring a recommendation information set matched with a user tag; determining a repurchase period of each brand related to the recommendation information set according to the historical behavior data; and selecting and outputting recommendation information with the difference between the repurchase period and the current time being less than a preset time threshold from the recommendation information set. The implementation method realizes rich and targeted information recommendation and improves the conversion rate.

Description

Method and apparatus for outputting information

Technical Field

The embodiment of the disclosure relates to the technical field of computers, in particular to a method and a device for outputting information.

Background

The daily platform operation can give the user a coupon or a rights touch by experience for the in-station user to pull new or repurchase. Regardless of the processes of renewing, repurchasing and loss recalling, the conventional method is to screen out coupons with different strengths for different crowds when applying for the coupons, and the transformation effect is hopefully improved to the maximum extent.

However, in the related technology, preferential activities depend on manual maintenance, manual orientation, high operation difficulty and serious information touch delay, and a user is not taken measures to retain the preferential activities when the user is about to run away.

Disclosure of Invention

Embodiments of the present disclosure propose methods and apparatuses for outputting information.

In a first aspect, an embodiment of the present disclosure provides a method for outputting information, including: in response to detecting that a user accesses a home page, acquiring historical behavior data of the user; marking a user tag for the user according to the historical behavior data; acquiring a recommendation information set matched with the user tag; determining a repurchase period of each brand related to the recommendation information set according to the historical behavior data; and selecting and outputting recommendation information with the difference between the repeated purchase period and the current time being less than a preset time threshold from the recommendation information set.

In some embodiments, the selecting and outputting recommendation information from the recommendation information set, where a difference between a repurchase period and a current time is less than a predetermined time threshold, includes: determining the brand with the difference between the repeated purchase period and the current time being less than a preset time threshold value as a promotion brand; scoring each promotional brand according to the historical behavior data; and outputting the recommendation information of each promotion brand from high to low according to the scores, wherein the recommendation information of the same promotion brand is sorted from high to low according to the preferential strength.

In some embodiments, said scoring each promotional brand according to said historical behavior data comprises: for each promotional brand, counting from the historical behavior data a number of at least one of the following behaviors relating to the promotional brand within a first predetermined time: adding a shopping cart, searching and clicking to enter a merchant detailed page, and clicking to browse commodities; for each promotional brand, a weighted sum of the number of various actions involving the promotional brand is calculated as a score for the promotional brand.

In some embodiments, weight of joining shopping cart > weight of searching and clicking into merchant detailed page > weight of clicking through goods, and weight of same behavior is decremented within the same day.

In some embodiments, the method further comprises: acquiring a pull information set of the brand not purchased by the user; and alternately outputting the pull information set and the recommendation information set.

In some embodiments, the determining a repurchase period for each brand to which the recommendation information set relates based on the historical behavior data comprises: determining a set of ordering time intervals of each brand related to the recommendation information set according to the historical behavior data; for each brand, sequencing the next single time interval of the brand and then taking the median as the repurchase period of the brand of the user; and if the repurchase period of the brand is more than or equal to the repurchase period of the platform full-quantity users, adjusting the repurchase period of the brand of the users according to the repurchase period of the platform full-quantity users.

In some embodiments, the method further comprises: in response to receiving a target brand input by a user, calculating a user brand passenger unit price of the target brand according to the historical behavior data; acquiring a unit price range of a target brand of an area where the user is located; increasing the unit price of the user brand of the target brand according to the unit price range of the target brand; acquiring a candidate recommendation information set of a target brand matched with the user tag; and selecting and outputting candidate recommendation information matched with the heightened user brand passenger order from the candidate recommendation information set.

In some embodiments, the calculating a user brand customer price for the target brand based on the historical behavior data includes: counting a per-unit purchase amount set of the target brand in a second preset time according to the historical behavior data; and taking the median in the per-purchase amount set as the customer unit price of the user brand of the target brand.

In some embodiments, the obtaining the unit price range of the target brand of the area where the user is located includes: acquiring user brand passenger prices of target brands of all users in the area where the users are located, and putting the user brand passenger prices into an area brand passenger price set; if the target brand does not meet within second preset time, putting the median of the basic prices of all commodities under the target brand into a regional brand passenger order set; and determining the unit price range of the target brand in the area according to the minimum value and the maximum value in the regional brand passenger unit price set.

In some embodiments, the increasing the unit price of the user brand of the target brand according to the unit price range of the target brand includes: sequencing the regional brand passenger order sets from small to large, and dividing the unit price range into a preset number of intervals according to quantiles; calculating the amplification according to the user brand passenger price interval of the target brand of the user; and increasing the customer unit price of the user brand of the target brand according to the increase.

In some embodiments, the selecting and outputting candidate recommendation information matching the heightened user brand customer price from the candidate recommendation information set includes: selecting candidate recommendation information with a threshold higher than the increased customer unit price of the user brand from the candidate recommendation information set; determining the priority according to the interval where the threshold of the selected candidate recommendation information is located, wherein the closer the threshold distance is to the customer unit price of the user brand after being increased, the higher the priority of the candidate recommendation information is; and outputting the selected candidate recommendation information in the order of the priority from high to low, wherein the candidate recommendation information with the same priority is sorted in the order of the preferential strength from high to low.

In some embodiments, the method further comprises: and if the target brand input by the user is not matched with the brand words in the brand library, selecting the closest brand word from the brand library as the target brand.

In a second aspect, an embodiment of the present disclosure provides an apparatus for outputting information, including: an acquisition unit configured to acquire historical behavior data of a user in response to detecting that the user accesses a home page; a marking unit configured to mark a user label for the user according to the historical behavior data; a matching unit configured to acquire a recommendation information set matched with the user tag; a determining unit configured to determine a repurchase period of each brand to which the recommendation information set relates according to the historical behavior data; and the recommending unit is configured to select and output the recommending information with the difference between the repurchasing period and the current time being less than a preset time threshold from the recommending information set.

In a third aspect, an embodiment of the present disclosure provides an electronic device for outputting information, including: one or more processors; storage means having one or more computer programs stored thereon, which when executed by the one or more processors, cause the one or more processors to carry out the method according to the first aspect.

In a fourth aspect, embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method according to the first aspect.

The method and the device for outputting information provided by the embodiment of the disclosure determine the user tags and the repurchase cycles of various brands by analyzing the historical behavior data of the users. And outputting recommendation information of a certain brand when the brand is about to reach a repurchase period for attracting users to repurchase. The target user can be timely and accurately determined to recommend information, and the user loss is avoided.

Drawings

Other features, objects and advantages of the disclosure will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:

FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present disclosure may be applied;

FIG. 2 is a flow diagram for one embodiment of a method for outputting information, according to the present disclosure;

FIG. 3 is a schematic illustration of a repurchase cycle calculation process of a method for outputting information according to the present disclosure;

FIG. 4 is a flow diagram of yet another embodiment of a method for outputting information in accordance with the present disclosure;

5a-5d are schematic diagrams of a user brand customer order calculation process for a method of outputting information according to the present disclosure;

FIG. 6 is a schematic block diagram illustrating one embodiment of an apparatus for outputting information according to the present disclosure;

FIG. 7 is a schematic block diagram of a computer system suitable for use with an electronic device implementing embodiments of the present disclosure.

Detailed Description

The present disclosure is described in further detail below with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.

It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of the disclosed method for outputting information or apparatus for outputting information may be applied.

As shown in fig. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The

terminal devices

101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.

The

terminal apparatuses

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture Experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like. When the

terminal apparatuses

101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.

The server 105 may be a server providing various services, such as a background server providing support for shopping on the

terminal devices

101, 102, 103. The background server may analyze and otherwise process the received data such as the order request, and feed back a processing result (for example, recommendation information such as a coupon distributed to the user) to the terminal device.

The server may be hardware or software. When the server is hardware, it may be implemented as a distributed server cluster formed by multiple servers, or may be implemented as a single server. When the server is software, it may be implemented as multiple pieces of software or software modules (e.g., multiple pieces of software or software modules used to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein. The server may also be a server of a distributed system, or a server incorporating a blockchain. The server can also be a cloud server, or an intelligent cloud computing server or an intelligent cloud host with artificial intelligence technology.

It should be noted that the method for outputting information provided by the embodiment of the present disclosure is generally performed by the server 105, and accordingly, the apparatus for outputting information is generally disposed in the server 105.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The following are explanations of some of the terms referred to in this application:

and (4) recalling the pool: a pool of goods to be exposed to a user.

To the manual price: the user places an order and uses a discount coupon red envelope and the like to offset the price of the order.

Ticket label: threshold, denomination, e.g., 100-5 coupons, threshold is 100 and denomination is 5 dollars.

GMV: for each item, all users submit orders and pay the total amount to complete the purchase order.

ROI is total GMV/coupon offer total ordered with coupon.

And (3) re-purchasing: the number of times that the user purchases the commodity is more than or equal to 2, and the method is called as repurchase.

And (4) carrying out refreshment: and newly downloading the APP and authenticating the registered user.

And (3) loss: the user has not placed an order within a certain time X days (the definition of each app X is different).

And (3) transformation: the user's ordering and transaction is regarded as conversion.

With continued reference to FIG. 2, a flow 200 of one embodiment of a method for outputting information in accordance with the present disclosure is shown. The method for outputting information comprises the following steps:

step 201, responding to the detected user accessing the home page, acquiring historical behavior data of the user.

In the present embodiment, an execution subject (e.g., a server shown in fig. 1) of the method for outputting information may receive an access request from a terminal with which a user performs online shopping through a wired connection manner or a wireless connection manner. The access request may be to the home page of the platform or to a home page of a specific brand store. The access request may include user identity information, such as an account, a mobile phone number, etc. Historical behavior data of the account can be obtained according to the user identity information. Historical behavior data may include, but is not limited to, joining a shopping cart, searching and clicking into a merchant details page, clicking through merchandise. The historical behavior data also includes information such as time of behavior, brand, commodity, hand price, and receiving address.

And 202, marking a user tag for the user according to the historical behavior data.

In the embodiment, user layering is performed according to the user behavior characteristics, so that target users suitable for various service scenes are mined, and refined user operation basic data are built. The tools for user layering render or call user labels for the user.

The target crowd is the user who buys the brand again, and the resource is the coupon, so the user label needing to be printed on the crowd can also be called the user coupon label.

Description of the drawings: the tag ID of the user tag is not modifiable for the system definition and is only used as an identification for the group marking.

1. And (4) calculating T +1, namely calculating the tag data of the previous day by the zero point of today.

According to the following steps: under the channel of brand (ID: 1), all the coupon brands in the coupon pool are transmitted by a Message Queue (MQ) to the newly added and removed coupons in the coupon pool and the brands bound under the coupons, and the big data end receives the message queue and then recalculates the user tag data in real time.

Because the coupon has a validity period, which is generally one month, in order to ensure that the resources triggered to the user are valid, the coupon pool is updated, when the coupon pool is updated, the coupon in the coupon pool is changed, and a part of coupons are removed or added, the user needs to recalculate the portrait data of the coupon in the current coupon pool (matching condition).

A coupon pool: i.e., a summary of coupons applied by business personnel.

MQ: if the coupon in the coupon pool is newly added or deleted, MQ messages are sent to inform the big data end that the data are changed, and the data need to be updated to ensure the accuracy of the data.

2. Label definitions

Judging whether the brands in the ticket pool and the circle layer to which the user belongs are hit according to the orders in the user preset time (for example, the last 90 days), wherein the order state is as follows: the payment on delivery is completed, or the payment on line is completed.

1) If a user (pull-up) of a certain brand has not purchased for 90 days, the pull-up 0 to 1 (ID: 11)

2) for a user (repurchase) with a certain brand order number 1 in 90 days, the repurchase of the brand is 1 to N (ID: 13)

3) a user (repurchase) with a certain brand order number of 1 within 90 days is the repurchase 1 to 2 (ID: 12)

4) regardless of the order situation, any group of people, this brand of unlimited group (ID: 10)

3. user label format [ belonged circle layer, value, threshold, denomination ]

The label format is standardized and is defined as a four-level label, so that the label can be applied when a coupon to be made is accurately touched subsequently.

The ring layer belongs to: 1 to N for repurchase (ID: 13), unlimited population (ID: 10), the tag ID is system-defined and not modifiable

The value: specific brand of repurchase

Examples of such applications are

If the user id1 hits LS-Laxin 0 to 1 ticket 100-10, UV-1 to 2 ticket 1000 to 100 and PG-1 to N ticket 200-5, the applicable user label of the scheme is [ 13, PG, 200, 5]

Step 203, acquiring a recommendation information set matched with the user label.

In the present embodiment, the recommendation information may be various types of coupons, such as full discount coupons, gift certificates, and the like. The recommendation information may also be some advertising information. The recommendation information is related to the brand. The user label comprises information of a belonging circle layer (new, 1-to-N and the like), a brand, a threshold, a denomination and the like. The recommendation information has a ticket tag in the same format as the user tag.

And when the business personnel applies for the recommendation information, the label screening is carried out, namely the background function supports marking the recommendation information, and only the recommendation information needs to be applied and selected. A ticket label format [ ticket location (corresponding to the belonging circle of the user label), value (brand), threshold, denomination ], and the label format specification is defined as a four-level label. And the ID of the ticket label is consistent with the ID of the user ticket label, and the ticket pop window can be triggered by matching.

And if the circle layer and the brand of the user label are respectively matched with the ticket positioning and value of the ticket label, the recommendation information corresponding to the ticket label is the recommendation information matched with the user label.

And step 204, determining the repurchase period of each brand related to the recommendation information set according to the historical behavior data.

In this embodiment, the follow-up purchase is a purchase of a brand of goods in the user's historical order. The application needs to analyze when and why a push (e.g., pop-up) of recommendation information (e.g., coupons) is triggered. The recent bill-drawing user does not need to issue a coupon pop-up window to stimulate the user consumption when actively purchasing, so that the target group is positioned as the user who has a historical order and does not purchase for a period of time, and the period of time is defined as the re-purchasing period of the user. That is, there is a risk of losing if the user does not place an order within the re-purchase period, and a coupon is needed to reach the purpose that the user stimulates the user to place an order.

The home page is a flow distribution scene, the brand repurchase user touch is carried out under the scene, and in order to ensure the accuracy and the strong correlation of users, the definition of a crowd is as follows: and (4) enabling the users who do not make orders to enter the brand survival promoting pool (1 to N for repurchase) in the brand repurchase period. And the user can immediately remove the order after placing the order, namely the user placing the order gets out of the brand activation pool.

The user brand repurchase period calculation logic is as follows:

in the order of the user within a predetermined time (for example, the last half year), the order is distinguished according to the back-end brand, and the order placing time interval is calculated as a repurchase cycle base value. And judging whether the user needs to enter a survival promoting pool or not according to the time interval between the current time and the last time when the user places the order of the certain brand of goods. If the user counts more than one unit in one day.

Several data index update periods are as follows: the Bpin repurchase period is updated on 1 day per month, the total number of users are updated on 1 day per month, and whether the users promote the update of the survival pool T +1 or not is judged.

1) Pulling all order data of the user in the last half year (firstly, payment is finished or goods arrive and pay is not considered to be returned);

2) calculating the back-end brand to which the commodity belongs in the order;

3) obtaining ordering time point t of each back-end brand₁、t₂……t_tCalculate t₁～t₂、t₂～t₃、t₃～t₄……t_t-1～t_tTime interval T of placing orders every two adjacent times_t；

4) Will T_tTaking the median T from small to large_{Repurchase of}A re-purchasing period of a certain brand for the user;

5) calculating the time interval T between the current time and the last purchasing time of a brand_iNow-T4 (T4 shown in FIG. 3) and T_{Repurchase of}A relationship;

if T_i>＝T_{Repurchase of}Entering a survival promoting pool after the step-1, marking the type of the ticket with a value (brand name 1/ID: T)_{Repurchase 1}Brand name 2/ID T_{Repurchase 2}Name of brand 3/ID: T_{Buyback 3}】

In some optional implementations of this embodiment, determining a repurchase period of each brand involved in the recommendation information set according to the historical behavior data includes: determining a set of ordering time intervals of each brand related to the recommendation information set according to the historical behavior data; for each brand, sequencing the next single time interval of the brand and then taking the median as the repurchase period of the brand of the user; and if the repurchase period of the brand is more than or equal to the repurchase period of the platform full-quantity users, adjusting the repurchase period of the brand of the users according to the repurchase period of the platform full-quantity users.

If the user brand buys cycle T again_{Repurchase of}>Total user buyback period T_All-purposeWhen Ti is equal to T_All-purposeEntering a brand activation pool at a time point, marking the type of the ticket with a value (brand name 1/ID: T)_All-purposeBrand name 2/ID T_All-purposeName of brand 3/ID: T_All-purpose】

T_All-purposeThe repurchase period for the platform full number of users is an empirical value obtained according to the probability distribution of the repurchase periods of the brand of all users.

Once the user enters the survival promotion pool, the time node for issuing the coupons is determined and cannot be influenced by the repeated purchase cycle updated every month until the user exits the survival promotion pool, and the total repeated purchase cycle of the user and the individual is updated when the user enters the next round of calculation.

The user repurchase cycle can be divided into platform, type and brand repurchase cycles, each type of commodity repurchase cycle is different, the brand repurchase cycle of the user needs to be calculated when the brand user is activated, the B2B platform user repurchase frequency is high, and personal behaviors of each person have randomness, so that the repurchase cycle of each person needs to be adjusted by the platform repurchase cycle, data is subjected to smoothing processing, and therefore the calculation logic is also specific to the B2B.

And step 205, selecting and outputting recommendation information with the difference between the repurchase period and the current time being less than the threshold value of the preset time length from the recommendation information set.

In this embodiment, when the current time is close to a re-purchasing period of a certain brand in the recommendation information set (e.g., 1 day away from the re-purchasing period), the upcoming due recommendation information may be output. If a plurality of pieces of recommendation information need to be output, the pieces of recommendation information can be sorted in the order of the discount degrees from large to small, for example, a coupon with a discount of 7 is arranged in front of a coupon with a discount of 8. The output mode is not limited to the pop-up window displaying of the coupons for the user to get, and the coupons can be directly put into the user account and prompted to be used by the user.

A person can have a plurality of brands of commodities in a re-purchasing period at the same time, but coupons of the brands cannot be issued to avoid disturbing users, and the demands of the users for commodities under the brands are different, so that the priority of distributing the brand coupons needs to be defined. In the part, the fact that the B2B platform user frequently visits the same commodity for many times is considered, so that the strength of the user appeal is different from that of other platforms, and strategies fused with the user re-purchase cycle are output.

Alternatively, it may be desirable to control the frequency of outputting the recommendation information, such as popping up coupon pops of up to two brands a day, and popping up coupon pops only once for the same user for one searched brand during the day.

In some optional implementation manners of this embodiment, selecting and outputting recommendation information from the recommendation information set, where a difference between a repurchase period and a current time is less than a predetermined duration threshold, includes: determining the brand with the difference between the repeated purchase period and the current time being less than a preset time threshold value as a promotion brand; scoring each promotional brand according to the historical behavior data; and outputting the recommendation information of each promotion brand from high to low according to the scores, wherein the recommendation information of the same promotion brand is sorted from high to low according to the preferential strength.

If the brand users who reach the repurchase period do not place an order, all corresponding brand tickets are recalled, the sequencing logic of each brand needs to be calculated, and if multiple coupons exist in each brand, the sequencing of the coupons in each brand needs to be calculated. Each promotional brand may be scored according to historical behavioral data and then sorted according to the score.

In some optional implementations of this embodiment, scoring each promotional brand according to the historical behavior data includes: for each promotional brand, counting from the historical behavior data a number of at least one of the following behaviors relating to the promotional brand within a first predetermined time: adding a shopping cart, searching and clicking to enter a merchant detailed page, and clicking to browse commodities; for each promotional brand, a weighted sum of the number of various actions involving the promotional brand is calculated as a score for the promotional brand.

The specific process is as follows:

1) since the home page coupon is strongly related to the user, the recalled brand ranking is judged according to the strength of the user behavior

The calculation logic:

i. the user behavior is divided into strong and weak, and the labor cost is different and the weight is different. Different weights are assigned according to different behaviors: weight of entering shopping cart > weight of searching and clicking into merchant detailed page > weight of clicking through goods, for example, weight ratio between them is 5: 3: 1, adding a shopping cart, 5, searching and clicking to enter a merchant detailed page, 3, clicking to browse the commodity, 1.

And ii, because the B2B user has a habit of repeated access, the user can access the merchant detailed page for price comparison for multiple times, and the B-end user can add multiple cars to the same sku, namely, one user can perform multiple operations on the same behavior of the same commodity in one day, the operations are recorded as n times (n >1), and the weight of the same behavior in the same day is decreased progressively. For example, the score weight is increased by 1+1/2+1/4 … +1/2n-1, and the score interval is [1,2 ].

For example, the first time the adding operation for the commodity A is 5 points, and the second time the adding operation is 5 points (1+ 1/2). If the vehicle is added for 2 times, the entering merchant detailed page is searched and clicked for 1 time, and the browsed commodity is clicked for 1 time, the score is given: 5*(1+1/2)+3+1

Aggregate all behavioral data under each brand for 30 days

Examples are: the user has behavioral operation on the commodity A under the brand 1 for two days, the behavior 1 is divided into 1.3, the behavior 3 is divided into 1.8, and then the behavior of the commodity A is divided into 1.3+1.8 which is 3.1 points;

under brand 1, the B commodity has behavior operation for two days, the behavior 1 is divided into 1.3, the behavior 3 is divided into 1.8, then the behavior of the a commodity is divided into 1.3+1.8 ═ 3.1, then the user is divided into 6.2 to the behavior of brand 1, and calculate the behavior scores of all brands.

And T +1, calculating the accumulated behavior score of the user on the brand in the last 30 days, recalling the brand tickets, and sequencing the recalled brand tickets according to the score, wherein the behavior score is not provided with a threshold value.

v. rank the brand behavior from high to low, then the absolute order of recalling brand tickets.

2) If the user behavior scores are the same or the user does not act, calculating the strength of the Top1 ticket under each brand, sorting the brands according to the strength, and if the strengths are the same, selecting the threshold to be low, and if the strengths are the same, randomly sorting.

Examples are: recalling brand note as a-Top1 note: 100-10 (strength 0.1), ticket 105-10 of Top2, ticket 100-5 of brand B-Top1 (strength 0.05), ticket 200-10 of Top2, then ranking of brands is obtained according to strength Top1, i.e. brand A > brand B

3) Within-brand coupon ordering. The logic for ranking each brand ticket is the same as the search "brand ticket. repurchase 1 to N ticket ranking" shown in flow 400, except that: the unlimited crowd tickets are also arranged according to rules.

Examples are: top1, Top2 and Top3 of brand A are obtained through the method; top1, Top2, Top3 of brand B; top1 ranking for brand C.

Top1, Top2, Top3 of Tokyo pool brand L; top1, Top2, Top3 of brand M; top1 ranking for brand N. The underpinning coupon pool brand is a brand that the user has not purchased.

Normal brand pool ordering: A-Top1> B-Top1> C-Top1> A-Top2> B-Top2> A-Top3> B-Top3

Ordering brand pools at the bottom of the support: L-Top1> M-Top1> N-Top1> L-Top2> M-Top2> L-Top3> M-Top3

Comprehensive sequencing: A-Top1> B-Top1> C-Top1> A-Top2> B-Top2> A-Top3> B-Top3> L-Top1> M-Top1> N-Top1> L-Top2> M-Top2> L-Top3> M-Top3

1) The same user can hit a plurality of brand tickets at the same time, and the tickets simultaneously containing 0-to-1-pulling-new and 1-to-N-repurchasing-1-to-N-pulling-new are existed, then the tickets are sorted according to the absolute sequence, and 1: 1, inserting; namely: pulling new 0 to 1 ticket 1, repurchasing 1 to N tickets 1, pulling new 0 to 1 ticket 1 … …

2) Updating the ticket pool, adding new tickets and performing incremental calculation;

3) the ticket pool updates rejected tickets and invalid tickets (the valid period is passed, and the like) to be filtered;

4) the exposed coupons need to be filtered, and after all the coupons are distributed for one turn, the user does not pick up the coupons and then recalls the coupons to participate in distribution.

Since the service applies for a plurality of coupons for the same brand, such as 100-5, 100-10, 200-10, etc., and which of the different threshold-denomination coupons is preferentially distributed to the user, the priority also needs to be defined, since B2B is the purchasing platform of the merchant, the user has the feature of centralized purchasing, that is, a large number of commodities are purchased at the same time. Therefore, the insertion and sorting of the coupons are required in consideration of the customer unit price and the variety of the coupons.

The users meeting the conditions are obtained through user insights, the coupon pop-up windows can be triggered after the users log in, one keyword pops up the same user only once in one day, and the coupon pop-up windows of two brands at most are popped up in one day. And each scene independently controls the popup touch logic without influencing the popup logic of the whole station.

The front-end buried point can recycle data, continuously adjust the algorithm model according to data expression, evaluate marketing effect and serve as a basis for adjusting strategies, and finally maximize the effect.

According to the method provided by the embodiment of the disclosure, the coupons conforming to the daily behaviors of the user are recalled, and when the user is touched, the coupons most possibly converted by the user are preferentially exposed, so that the conversion rate is improved while the unit price of the customer is ensured, and the system recommendation replaces manual operation.

With further reference to fig. 4, a flow 400 of yet another embodiment of a method for outputting information is shown. The process 400 of the method for outputting information includes the steps of:

step 401, in response to receiving the target brand input by the user, calculating the customer unit price of the user brand of the target brand according to the historical behavior data.

In the present embodiment, an electronic device (e.g., a server shown in fig. 1) on which the method for outputting information is operated may receive a target brand input by a user from a terminal with which the user performs online shopping through a wired connection manner or a wireless connection manner. The customer brand customer price for the target brand is calculated based on historical behavior data already obtained by the process 200. And counting a per-unit purchase amount set of the target brand in a second preset time according to the historical behavior data. The median in the per-purchase amount set may be taken as the customer brand price for the user of the target brand. The average in the per-purchase amount set may also be used as the user brand guest price for the target brand.

The user brand guest unit price calculation process is as follows:

1) calculating the purchase amount per unit of each brand in an order of a user in a preset time (such as the last half year), and accumulating and summing up the purchase amount as a single amount if the number of the orders is more than one day;

2) taking the median of the hand Price as the brand passenger Price of the User;

updating frequency: 1 day per month (can be updated by incremental data, reducing the amount of calculation).

Optionally, if the target brand input by the user does not match the brand word in the brand library, selecting the closest brand word from the brand library as the target brand. If the brand words input by the user are not accurate (wrong characters and the like), the brand words need to be called into a synonym word list, so that the brand to which the user input words belong can be accurately identified, for example, if the user inputs '3 squirrels', the user should be matched with 'three squirrels'; for example, if the brand word entered by the system is "happy (Lay's)" and the user search keyword is often "happy" or "Lay's", these three words are also synonymous. The brand ticket pop-up window can be accurately popped up regardless of any brand word input by the user.

Step 402, acquiring a unit price range of a target brand of an area where a user is located.

In this embodiment, user brand customer prices of target brands of all users in an area where the user is located are obtained and put into an area brand customer price set; if the target brand does not meet within second preset time, putting the median of the basic prices of all commodities under the target brand into a regional brand passenger order set; and determining the unit price range of the target brand in the area according to the minimum value and the maximum value in the regional brand passenger unit price set.

The preamble description is as follows: all of the commercial products include: the commodity with transaction in the last half year (on shelf + off shelf), the commodity without transaction (on shelf at the present time)

1) In the last half year there was a deal order, user + brand dimension.

And according to regional distinction, taking the median of each brand to the hand Price of all the user order tables as the brand passenger order Price of the user. If a user purchases a plurality of times a day, the brand customer price is accumulated and summed as the brand customer price of the user (sku does not deduplicate).

For example, brand a is purchased by User1 (1 st hand price), User2 (3 st hand price), User3 (5 st hand price), User4 (3 st hand price), and User5 (1.2 st hand price) in beijing, respectively, brand guest prices of brand a in beijing are (1, 3, 5, 3, 1.2), and brand a in beijing are within the range of [1,5 ].

2) Nearly half year without transaction brand, sku dimension

Each sku is distinguished according to regions, and the median of the basic prices of all skus under each brand is taken as the Price of the brand passenger

The above two cases calculate the Price of brand guest in all brand regions.

Updating frequency: 1 day per month (can be updated according to incremental data, reduce calculated amount)

And 403, increasing the unit price of the user brand of the target brand according to the unit price range of the target brand.

In this embodiment, in order to secure the ROI for using the coupon, it is desirable that the price for the user to place a purchase order using the coupon is higher than the daily price for placing a purchase order, and therefore, the adjusted price is required in the user dimension.

The specific process is shown in fig. 5a-5 d:

s4031, the regional brand guest unit sets are sorted from small to large, and the unit range is divided into a preset number of intervals according to quantiles.

Sorting the Price from small to large, dividing 10 intervals [0.1, 0.2, 0.3,0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1 ] according to quantiles]I.e. the quantile interval q ═ q_m，q_n]And Q ═ Q for each quantile value_m，Q_n]。

The noun explains:

Price-User: user's certain brand passenger order

Q value corresponding to five-point number

q_m: a left or right section of quantile corresponding to a certain brand passenger order of the user; q. q.s_m<0.5 left interval q_m>0.5 Right interval

q_n: q value corresponding to quintile number is 0.5

Examples are: suppose that the brand guest unit Price sold in Beijing region by a certain brand A is [60,600,600%]If a User has a Price-User of 160, the Price is [60,600]10 quantiles are divided, and 10 intervals are corresponding to the quantiles. Each quantile corresponds to a value of Q ═ Q_m，Q_n]. As shown in fig. 5 a. It can be known that

q_n＝0.5。

S4032, the user brand customer price of the target brand is increased according to the section where the user brand customer price of the target brand of the user is located.

And judging the region where the Price-User falls in the Price to obtain the Q value and the Q value. Find the Q-interval [0.3,0.4] Q-interval [160,200] corresponding to brand A for this user as shown in FIG. 5 b.

The brand guest unit price of the user can be increased, but the brand guest unit price does not exceed the interval. For example, the brand guest price of the user after the increase is 180.

The increased User brand customer Price-User-new can also be calculated according to the following formula:

the User brand customer Price-User-new after the increase calculated in the interval of fig. 5 b:

Price-User-new＝160+|160-260|*|0.3-0.5|＝160+18＝178

i.e., pushing a 178 dollar threshold coupon to the user can improve conversion.

And step 404, acquiring a candidate recommendation information set of the target brand matched with the user label.

In this embodiment, the matching process is the same as step 203, and candidate recommendation information of the same belonging circle layer and brand is matched. The candidate recommendation information is named only for distinguishing from the recommendation information in the process 200, and the two types of information have substantially the same attribute and have four-level labels. The recommendation information has a threshold, i.e. the minimum condition to be met for using the coupon, e.g. a threshold of 100-5 is 100, and the user needs to buy 100 yuan to use the coupon.

And 405, selecting and outputting candidate recommendation information matched with the heightened user brand customer order from the candidate recommendation information set.

In this embodiment, the candidate recommendation information having a threshold equal to or higher than the increased brand passenger price of the user is selected from the candidate recommendation information set. And setting the priority of the candidate recommendation information with the threshold equal to the increased brand passenger unit price of the user as the highest priority. And determining the priority according to the interval where the threshold of the selected candidate recommendation information is located, wherein the priority of the candidate recommendation information is higher when the threshold is closer to the adjusted user brand customer unit price. And outputting the selected candidate recommendation information in the order of the priority from high to low, wherein the candidate recommendation information with the same priority is sorted in the order of the preferential strength from high to low.

And acquiring the coupons in the coupon pool through the MQ, and recalculating the lower data when the coupons are newly added in the coupon pool. And filtering invalid tickets in real time.

Absolute order: the ticket pool has a threshold (Price-User-new) (sequence 1) and the ticket pool has a brand ticket (sequence 2) with a threshold (Price-User-new)

1. Brand tickets with threshold of Price-User-new [ brand tickets available for brand keywords searched by users ] in ticket pool

Coupon [ threshold-denomination ], force λ ═ denomination/threshold

Brand coupons containing the same threshold are ranked from high to low according to strength λ.

2. The brand ticket with threshold not equal to Price-User-new in ticket pool

1) And judging the quantile range of the brand guest unit Price in the region where the User adjusts the brand guest unit Price of the brand.

In the above example, Price-User-new is 178, Q [0.3,0.4], Q [160,200], and the coupon thresholds are 10, 100, 110, 200, 300, 460, respectively, as indicated by the dots in fig. 5c and 5 d.

2) Then the coupon priority 1 recall within Q of the Price-User-new

The multiple tickets are ordered from high to low according to the strength lambda.

3) The coupons in the sub-position range adjacent to the Q left and right of the Price-User-new are recalled according to the priority 2

4) Other coupon priority 3 recalls

I.e., sorted in absolute order of priority by the recall pool of coupons, sorted from high to low by the degree λ.

And (3) inputting a keyword (brand name) by a user, clicking a search button, immediately triggering a popup window, and specifically popping a coupon of which brand according to the coupon sequencing logic. Considering the user experience problem, one keyword is popped once for the same user in one day, and the coupon pop window of two keywords is popped at most in one day.

And each scene independently controls the popup touch logic without influencing the popup logic of the whole station.

As can be seen from fig. 4, compared with the embodiment corresponding to fig. 2, the flow 400 of the method for outputting information in the present embodiment represents a process for increasing the coupon usage threshold. In order to ensure the ROI for using the coupon, it is desirable that the price for the user to place an order using the coupon is higher than the daily price for placing an order, so that there is a need for adjusting the price in the user dimension. The recalled coupons are ranked, and the interests of the users and the interests of the brand merchants are considered in a fusion mode in consideration of ranking logics of two dimensions, namely user interest, affordability (unit price of customers) and brand merchant premium (unit price of customers is pulled up).

With further reference to fig. 6, as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an apparatus for outputting information, which corresponds to the method embodiment shown in fig. 2, and which is particularly applicable in various electronic devices.

As shown in fig. 6, the apparatus 600 for outputting information of the present embodiment includes: the system comprises an acquisition unit 601, a marking unit 602, a matching unit 603, a determination unit 604 and a recommendation unit 605. The obtaining unit 601 is configured to obtain historical behavior data of a user in response to detecting that the user accesses a home page; a marking unit 602 configured to mark a user label for the user according to the historical behavior data; a matching unit 603 configured to obtain a recommendation information set matching the user tag; a determining unit 604 configured to determine a repurchase period for each brand to which the recommendation information set relates, based on the historical behavior data; and a recommending unit 605 configured to select and output recommendation information from the recommendation information set, wherein the difference between the repurchase period and the current time is less than a predetermined time threshold.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: determining the brand with the difference between the repeated purchase period and the current time being less than a preset time threshold value as a promotion brand; scoring each promotional brand according to historical behavior data; and outputting the recommendation information of each promotion brand from high to low according to the scores, wherein the recommendation information of the same promotion brand is sorted from high to low according to the preferential strength.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: for each promotional brand, counting from the historical behavior data a number of at least one of the following behaviors relating to the promotional brand within a first predetermined time: adding a shopping cart, searching and clicking to enter a merchant detailed page, and clicking to browse commodities; for each promotional brand, a weighted sum of the number of various actions involving the promotional brand is calculated as a score for the promotional brand.

In some optional implementations of this embodiment, the weight of joining a shopping cart > search and click the weight of entering a merchant detailed page > click the weight of browsing the goods, and the weight of the same action within the same day is decremented.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: acquiring a pull information set of a brand not purchased by a user; the pull information set and the recommendation information set are alternately output.

In some optional implementations of this embodiment, the determining unit 604 is further configured to: determining a list-leaving time interval set of each brand related to the recommendation information set according to the historical behavior data; for each brand, sorting the next single time interval of the brand, and taking the median as the repurchase period of the brand of the user; and if the repurchase period of the brand is more than or equal to the repurchase period of the platform full-quantity users, adjusting the repurchase period of the brand of the users according to the repurchase period of the platform full-quantity users.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: in response to receiving the target brand input by the user, calculating the customer unit price of the user brand of the target brand according to the historical behavior data; acquiring a unit price range of a target brand of an area where a user is located; the unit price of the user brand of the target brand is increased according to the unit price range of the target brand; acquiring a candidate recommendation information set of a target brand matched with a user tag; and selecting candidate recommendation information matched with the heightened user brand passenger order from the candidate recommendation information set and outputting the candidate recommendation information.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: counting a per-unit purchase amount set of the target brand in a second preset time according to the historical behavior data; and taking the median in the per-purchase amount set as the customer unit price of the user brand of the target brand.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: acquiring user brand passenger prices of target brands of all users in an area where the users are located, and putting the user brand passenger prices into an area brand passenger price set; if the target brand does not meet within the second preset time, putting the median of the basic prices of all the commodities under the target brand into the regional brand customer price set; and determining the unit price range of the target brand in the area according to the minimum value and the maximum value in the regional brand passenger unit price set.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: sequencing regional brand passenger unit sets in a descending order, and dividing a unit range into a preset number of intervals according to quantiles; and increasing the user brand passenger price of the target brand according to the section of the user brand passenger price of the target brand of the user.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: selecting candidate recommendation information with a threshold higher than the increased customer unit price of the user brand from the candidate recommendation information set; determining the priority according to the interval where the threshold of the selected candidate recommendation information is located, wherein the closer the threshold distance is to the customer unit price of the user brand after being increased, the higher the priority of the candidate recommendation information is; and outputting the selected candidate recommendation information in the order of the priority from high to low, wherein the candidate recommendation information with the same priority is sorted in the order of the preferential strength from high to low.

In some optional implementations of this embodiment, the recommending unit 605 is further configured to: and if the target brand input by the user is not matched with the brand words in the brand library, selecting the closest brand word from the brand library as the target brand.

According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

An electronic device for outputting information, comprising: one or more processors; a storage device having one or more computer programs stored thereon that, when executed by the one or more processors, cause the one or more processors to implement the methods of

flows

200 and 400.

A computer-readable medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the methods of

flows

200 and 400.

FIG. 7 illustrates a schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the device 700 comprises a computing unit 701, which may perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM)702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Various components in the device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, or the like; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

Computing unit 701 may be a variety of general purpose and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 701 executes the respective methods and processes described above, such as a method for outputting information. For example, in some embodiments, the method for outputting information may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of a computer program may be loaded onto and/or installed onto device 700 via ROM 702 and/or communications unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the method for outputting information described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured by any other suitable means (e.g., by means of firmware) to perform the method for outputting information.

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a server of a distributed system or a server incorporating a blockchain. The server can also be a cloud server, or an intelligent cloud computing server or an intelligent cloud host with artificial intelligence technology. The server may be a server of a distributed system or a server incorporating a blockchain. The server can also be a cloud server, or an intelligent cloud computing server or an intelligent cloud host with artificial intelligence technology.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for outputting information, comprising:

in response to detecting that a user accesses a home page, acquiring historical behavior data of the user;

marking a user tag for the user according to the historical behavior data;

acquiring a recommendation information set matched with the user tag;

determining a repurchase period of each brand related to the recommendation information set according to the historical behavior data;

and selecting and outputting recommendation information with the difference between the repeated purchase period and the current time being less than a preset time threshold from the recommendation information set.

2. The method of claim 1, wherein the selecting and outputting recommendation information from the set of recommendation information that a difference between a repurchase period and a current time is less than a predetermined length threshold comprises:

determining the brand with the difference between the repeated purchase period and the current time being less than a preset time threshold value as a promotion brand;

scoring each promotional brand according to the historical behavior data;

and outputting the recommendation information of each promotion brand from high to low according to the scores, wherein the recommendation information of the same promotion brand is sorted from high to low according to the preferential strength.

3. The method of claim 2, wherein the scoring each promotional brand according to the historical behavior data comprises:

for each promotional brand, counting from the historical behavior data a number of at least one of the following behaviors relating to the promotional brand within a first predetermined time: adding a shopping cart, searching and clicking to enter a merchant detailed page, and clicking to browse commodities;

for each promotional brand, a weighted sum of the number of various actions involving the promotional brand is calculated as a score for the promotional brand.

4. The method of claim 3, wherein the weight of joining a shopping cart > search and click into merchant detailed page > click through weight of browsing merchandise, and the weight of the same action within the same day is decremented.

5. The method of claim 1, wherein the method further comprises:

acquiring a pull information set of the brand not purchased by the user;

and alternately outputting the pull information set and the recommendation information set.

6. The method of claim 1, wherein the determining a repurchase period for each brand to which the set of recommendation information relates from the historical behavior data comprises:

determining a set of ordering time intervals of each brand related to the recommendation information set according to the historical behavior data;

for each brand, sequencing the next single time interval of the brand and then taking the median as the repurchase period of the brand of the user; and if the repurchase period of the brand is more than or equal to the repurchase period of the platform full-quantity users, adjusting the repurchase period of the brand of the users according to the repurchase period of the platform full-quantity users.

7. The method of claim 1, wherein the method further comprises:

in response to receiving a target brand input by a user, calculating a user brand passenger unit price of the target brand according to the historical behavior data;

acquiring a unit price range of a target brand of an area where the user is located;

increasing the unit price of the user brand of the target brand according to the unit price range of the target brand;

acquiring a candidate recommendation information set of a target brand matched with the user tag;

and selecting and outputting candidate recommendation information matched with the heightened user brand passenger order from the candidate recommendation information set.

8. The method of claim 7, wherein the calculating a user brand customer price per unit for a target brand from the historical behavior data comprises:

counting a per-unit purchase amount set of the target brand in a second preset time according to the historical behavior data;

and taking the median in the per-purchase amount set as the customer unit price of the user brand of the target brand.

9. The method of claim 7, wherein the obtaining of the unit price range of the target brand of the area where the user is located comprises:

acquiring user brand passenger prices of target brands of all users in the area where the users are located, and putting the user brand passenger prices into an area brand passenger price set;

if the target brand does not meet within second preset time, putting the median of the basic prices of all commodities under the target brand into a regional brand passenger order set;

and determining the unit price range of the target brand in the area according to the minimum value and the maximum value in the regional brand passenger unit price set.

10. The method of claim 9, wherein the adjusting the unit price of the user brand of the target brand up according to the unit price range of the target brand comprises:

sequencing the regional brand passenger order sets from small to large, and dividing the unit price range into a preset number of intervals according to quantiles;

and increasing the user brand passenger unit price of the target brand according to the interval of the user brand passenger unit price of the target brand of the user.

11. The method of claim 9, wherein the selecting and outputting candidate recommendation information from the set of candidate recommendation information that matches the heightened user brand customer price comprises:

selecting candidate recommendation information with a threshold higher than the increased customer unit price of the user brand from the candidate recommendation information set;

determining the priority according to the interval where the threshold of the selected candidate recommendation information is located, wherein the closer the threshold distance is to the customer unit price of the user brand after being increased, the higher the priority of the candidate recommendation information is;

and outputting the selected candidate recommendation information in the order of the priority from high to low, wherein the candidate recommendation information with the same priority is sorted in the order of the preferential strength from high to low.

12. The method of claims 7-11, wherein the method further comprises:

and if the target brand input by the user is not matched with the brand words in the brand library, selecting the closest brand word from the brand library as the target brand.

13. An apparatus for outputting information, comprising:

an acquisition unit configured to acquire historical behavior data of a user in response to detecting that the user accesses a home page;

a marking unit configured to mark a user label for the user according to the historical behavior data;

a matching unit configured to acquire a recommendation information set matched with the user tag;

a determining unit configured to determine a repurchase period of each brand to which the recommendation information set relates according to the historical behavior data;

and the recommending unit is configured to select and output the recommending information with the difference between the repurchasing period and the current time being less than a preset time threshold from the recommending information set.

14. An electronic device for outputting information, comprising:

one or more processors;

a storage device having one or more computer programs stored thereon,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-12.

15. A computer-readable medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, carries out the method according to any one of claims 1-12.