WO2017092602A1 - 一种信息投放用户的筛选方法和服务器 - Google Patents

一种信息投放用户的筛选方法和服务器 Download PDF

Info

Publication number
WO2017092602A1
WO2017092602A1 PCT/CN2016/107078 CN2016107078W WO2017092602A1 WO 2017092602 A1 WO2017092602 A1 WO 2017092602A1 CN 2016107078 W CN2016107078 W CN 2016107078W WO 2017092602 A1 WO2017092602 A1 WO 2017092602A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
information
behavior data
delivery
server
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2016/107078
Other languages
English (en)
French (fr)
Inventor
胡于响
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of WO2017092602A1 publication Critical patent/WO2017092602A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0253—During e-commerce, i.e. online transactions

Definitions

  • the present application relates to the field of electronic commerce, and in particular, to a screening method and server for an information delivery user.
  • a solution is proposed in the prior art for the above problem, and the solution includes a parameter configuration module, a receiving module, and a commodity page file generating module.
  • the parameter configuration is used to generate an activity page file for displaying a coupon activity to a specific merchant based on the configured activity range control parameter and activity information display control parameter.
  • the activity scope control parameter is configured to configure a commodity category participating in the coupon activity
  • the activity information display control parameter is used to configure a merchant participating in the coupon activity.
  • the activity information display control parameters corresponding to each merchant may be stored in the form of a database table. By configuring the activity information display control parameter, the merchant can be set to participate in the coupon activity and the non-participating coupon activity, and the business page file is not generated for the merchant who cannot participate in the coupon activity.
  • the receiving module is configured to receive an activity request submitted by a merchant in the specific merchant.
  • the product page file generation module is used to generate a product page for the item that meets the review rule among the merchants submitting the activity request
  • the document, the product page file includes product information and coupon activity information of participating in the coupon event.
  • the product page file is generated for the merchant in the white list, and the white list is a pre-set list of merchants that satisfy the review rule.
  • the daily behavior data of the consumer is not analyzed, and the user who is interested in the store coupon cannot be accurately divided, and the parameters are manually set, which is not only smart enough, but also has a large workload, and at the same time, is whitelisted.
  • the way to determine that the user who meets the requirements will make the user change is not flexible enough.
  • the purpose of the application is to provide a screening method and a server for information delivery users.
  • By analyzing user behavior data to determine potential users who need to deliver information not only can the potential users be accurately positioned, but also the hit rate of information delivery is improved. .
  • a screening method for an information delivery user comprising:
  • the server obtains behavior data of the user
  • the server determines a comprehensive score of the user according to a preset behavior data score
  • the server determines a delivery order of information to each user according to the user comprehensive score and a comprehensive score of other users;
  • the server delivers the information to the user according to the delivery type of the information and the preset delivery policy according to the delivery sequence.
  • the behavior data includes any combination of one or more of the following:
  • the number of views to the target object the collection of the target object, the retrieval of the target object, industry preferences, and historical purchase information.
  • the server determines whether the user behavior data meets a preset behavior criterion, specifically:
  • the server determines whether the number of times the user browses the target object exceeds a browsing threshold in the first preset time
  • the server determines whether the user has collected the target object in the second preset time
  • the server determines whether the number of times the user retrieves the target object exceeds a retrieval threshold in a third preset time
  • the server When the behavior data is an industry preference, the server counts the industry preference of the user in a fourth preset time, and determines whether the industry preference of the user matches the industry of the target object;
  • the server collects the shopping information of the user in the fifth preset time, and determines whether the shopping information of the user is associated with the target object.
  • the server determines a comprehensive score of the user according to a preset behavior data score, specifically:
  • the server determines a type of behavior data of the user
  • the server scores the type of behavior data of the user according to a preset behavior data score
  • the server determines a comprehensive score of the user according to a score of the type of behavior data of the user.
  • the server delivers the information to the user according to the delivery type of the information and the preset delivery policy according to the delivery sequence, specifically:
  • the server determines, according to the behavior data of the user and the industry pen price of the user, the type of delivery of the information by using the preset delivery policy;
  • the server delivers the information to the user according to the delivery order according to the type of delivery of the information.
  • a server comprising:
  • An acquisition module configured to acquire behavior data of the user
  • a determining module configured to determine whether the behavior data of the user meets a preset behavior criterion
  • a first determining module configured to determine a comprehensive score of the user according to a preset behavior data score if the behavior data of the user meets a preset behavior criterion
  • a second determining module configured to determine, according to the user comprehensive score and a comprehensive score of other users, a delivery order of information to each user;
  • the delivery module is configured to deliver the information to the user according to the delivery order according to the type of delivery of the information and the preset delivery policy.
  • the behavior data includes any combination of one or more of the following:
  • the number of views to the target object the collection of the target object, the retrieval of the target object, industry preferences, and historical purchase information.
  • the determining module is specifically configured to:
  • the behavior data is the number of times of browsing the target object, determining whether the number of times the user browses the target object exceeds a browsing threshold in the first preset time;
  • the behavior data is a collection situation of the target object, determining whether the user has collected the target object in the second preset time;
  • the behavior data is the number of retrievals to the target object, determining whether the number of times the user retrieves the target object exceeds a retrieval threshold in a third preset time;
  • the industry preference of the user is counted in a fourth preset time, and it is determined whether the industry preference of the user matches the industry of the target object;
  • the shopping information of the user is counted in a fifth preset time, and it is determined whether the shopping information of the user is associated with the target object.
  • the first determining module is specifically configured to:
  • a comprehensive score of the user is determined based on a score of the type of behavior data of the user.
  • the delivery module is specifically configured to:
  • the present application determines whether the behavior data of the user satisfies a preset behavior criterion by analyzing the behavior data of the user, and if yes, confirms the delivery order of the information to the user according to the determined comprehensive score of the user, and then according to the information.
  • the delivery type and the preset delivery policy deliver the information according to the delivery order.
  • the application can determine the potential users who need the delivery information, and not only can the potential users be accurately ranked by sorting the potential users and delivering the information.
  • the hit rate of the information delivery is also improved, and the user who has a great interest in the delivery information can be preferentially obtained, and the application directly analyzes the user based on the big data, thereby reducing the workload of the operator, and When the user's behavior changes, he can flexibly make corresponding judgments on the user's changes.
  • FIG. 1 is a flowchart of a screening method for an information delivery user in an embodiment of the present application
  • FIG. 2 is a schematic structural diagram of a server in an embodiment of the present application.
  • the prior art does not analyze the daily behavior data of the consumer, and cannot accurately distinguish the users who are interested in the store coupon and manually set the parameters, which is not only smart enough, but also has a large workload, and at the same time, The way the whitelist is determined is that the user who meets the requirements will make the user change when the user changes.
  • the embodiment of the present application provides a screening method for the information delivery user, which is determined by analyzing the behavior data of the user to satisfy the delivery standard of the information, after the user meets the delivery standard of the information. Determining the degree of the user's demand for the information, and determining the delivery order of the information to the user, and determining the delivery information corresponding to the user according to the type of the information and the preset delivery policy.
  • the delivery of the information in the order of delivery enables the precise positioning of the delivery information and the accurate delivery of the delivery information, which improves the hit rate of the information delivery, and ensures that users who are interested in the delivery information can be preferentially obtained.
  • FIG. 1 it is a schematic flowchart of a screening method for an information delivery user according to an embodiment of the present application, where the method includes the following steps:
  • step 101 the server acquires behavior data of the user.
  • the behavior data includes any combination of one or more of the following:
  • the number of views to the target object the collection of the target object, the retrieval of the target object, industry preferences, and historical purchase information.
  • the target object here may refer to a store, or a business object sold in a store, and the business object itself may be a physical commodity or a service, for example, car wash, maintenance, Massage, cleaning, chef visit, housekeeping, tutoring, entertainment, eating and drinking, travel, hotels, car rental, etc.
  • Any content that can be included in the user's needs can be used as an analysis object in the embodiment of the present application to identify potential target users. Such a change does not affect the scope of protection of the present application.
  • the meaning of the target object also has the above-mentioned limitations, and the description will not be repeated hereinafter.
  • the server may be a server of an e-commerce platform, and the user may leave an operation trace on the e-commerce platform during daily online shopping. These operation traces are behavior data of the user, and the potential business object of the user may be analyzed according to the operation trace. Demand or favorite business objects in the store.
  • the business object is a physical product
  • the product indicates that the user prefers the product of the store, but for some reasons (such as: price or no discount, etc.) and does not purchase
  • the information of the number of times the user browses the product of the store can determine which user the user is.
  • the goods in the store are interested, and you can further analyze what kind of products are in the store.
  • the user has collected a store's merchandise, or retrieved a certain merchandise a certain number of times, or the user's industry preferences (such as a user often buys some electronic products indicating that the user has a preference for electronic products), and users
  • the shopping information is clearly associated with certain items, and it can also indicate that users are interested in these items.
  • the business object is a specific service content, such as housekeeping service, hotel business, etc.
  • a specific service content such as housekeeping service, hotel business, etc.
  • the user's behavior data may also include other operational traces left by the user on the server (such as the aforementioned e-commerce platform) during the daily consumption process, and any behavior information indicating the potential business needs of the user belongs to the protection of the present application. range.
  • Step 102 The server determines whether the behavior data of the user meets a preset behavior criterion.
  • step 103 is performed, and if not, it ends.
  • the server determines whether the user behavior data meets a preset behavior criterion, specifically:
  • the server determines whether the number of times the user browses the target object exceeds a browsing threshold in the first preset time
  • the server determines whether the user has collected the target object in the second preset time
  • the server determines whether the number of times the user retrieves the target object exceeds a retrieval threshold in a third preset time
  • the server When the behavior data is an industry preference, the server counts the industry preference of the user in a fourth preset time, and determines whether the industry preference of the user matches the industry of the target object;
  • the server collects the shopping information of the user in the fifth preset time, and determines whether the shopping information of the user is associated with the target object.
  • Case 1 behavior data is the number of views on the target object.
  • the target user is determined based on: when a user browses a store's merchandise more than once. When it is counted, it indicates that the user has a relatively strong purchase demand for the goods of the store, but may not be purchased due to reasons such as price or no discount, and therefore, it may be determined that the user is a potential purchase user of the commodity, or The user is a potential purchase user of the store.
  • the number of times the user browses the store or the product in the store within the first preset time period can be obtained, and whether the number of views is determined Exceeded the preset browsing threshold. If it is exceeded, the corresponding user satisfies the preset behavior standard of the store, indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store when the information is served.
  • the preset behavior standard of the store indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store when the information is served.
  • the first preset time and the browsing threshold may be determined according to actual needs.
  • Case 2 Behavior data is a collection of target objects.
  • the target user is determined according to: when a user collects the goods of a certain store, it indicates that the user has a desire to purchase the product of the store or prefers the product, but may be due to price or There is no purchase for reasons such as discounts, or when a user has collected a store, it indicates that the user has a desire to purchase the product of the store or prefers the store, but may not purchase it due to reasons such as price or no discount.
  • it can be determined that the user is a potential purchase user for the item of the store, or the user is a potential purchase user of the store.
  • each user in the second preset time range has collected the store or the store's product. If the collection is over, the corresponding user satisfies the preset behavior standard of the store, indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store as the information is served.
  • the preset behavior standard of the store indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store as the information is served.
  • One of the users On the contrary, if there is no collection, the corresponding user does not meet the preset behavior standard of the store, indicating that the user does not have a strong demand for the purchase of the store or the store, and does not need to be placed when the store delivers the information.
  • the second set time can be determined according to actual needs.
  • Case 3 Behavior data is the number of searches for the target object.
  • the target user is determined based on: when a user searches for a certain product for a certain number of times, indicating that the user has a relatively strong purchase demand for the goods of the store, but may be due to price or There is no purchase for reasons such as discounts, etc., therefore, it can be determined that the user is a potential purchase user of the item, or the user is a potential purchase user of the shop.
  • the number of times of retrieval of the product by the user for the store or the store within the third preset time period may be obtained, and whether the number of searches is determined Exceeded the preset search threshold. If it is exceeded, the corresponding user satisfies the preset behavior standard of the store, indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store when the information is served.
  • the preset behavior standard of the store indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store when the information is served.
  • the corresponding user does not meet the preset behavior standard of the store, indicating that the user does not have a strong demand for the purchase of the store or the store, and does not need to serve the information when the store delivers the information. user.
  • the third preset time and the retrieval threshold may be determined according to actual needs.
  • Some users have industry preferences. For example, if a user's industry preference is an electronic product, then the user will definitely pay attention to or purchase a lot of electronic products, or some users have obvious preferences for certain products over a period of time, such as When a user is doing home improvement, the user has a clear industry preference for home improvement products. If a user's industry prefers a store's merchandise, then that user becomes a potential buyer for those stores.
  • the user's industry preference may be fixed or may have an industry preference for a period of time, it is necessary to count the industry preference of the user within the fourth preset time range, and determine whether or not the product of a certain store is based on the statistical industry preference. Consistent. If the match is met, the predetermined behavior standard of the store is satisfied, indicating that the user has a strong demand for the purchase of the store or the store, so the user can be regarded as one of the users that the store needs to serve when the information is served. .
  • the fourth set time can be determined according to actual needs.
  • the user's shopping information can reflect which kind of product the user needs. For example, if a user has purchased a laptop, the user may need a laptop computer and a computer accessory related to the laptop, if the goods and laptop of a certain store are used. Whether it is related to computer accessories related to laptops, then the user is a potential buyer of the store.
  • the shopping information of each user in the fifth preset time can be counted, and whether the product of a certain store is associated with the shopping information of each user is determined. If there is an association, the corresponding user satisfies the preset behavior standard of the store, indicating that the user has a strong demand for the purchase of the store or the store, so the user may be required to serve the store as the information is served. One of the users. On the contrary, if there is no association, the corresponding user does not meet the behavior standard preset by the store, indicating that the user does not purchase the goods for the store or the store. Too strong, there is no need to serve the user when posting information at the store.
  • the fifth preset time may be determined according to actual needs.
  • Step 103 The server determines a comprehensive score of the user according to a preset behavior data score.
  • the server determines a comprehensive score of the user according to a preset behavior data score, specifically:
  • the server determines a type of behavior data of the user
  • the server scores the type of behavior data of the user according to a preset behavior data score
  • the server determines a comprehensive score of the user according to a score of the type of behavior data of the user.
  • a score is set in advance for the behavior data, for example, the score of a target object is W1, the score of the target object is W2, the score of the target object is W3, and the industry preference is consistent with the target object.
  • the score is W4, and the score of the shopping information associated with the target object is W5.
  • the user may have a plurality of behavior data that meet the behavior standard preset by the store. For example, the number of times the user browses the goods of one store in the first preset time exceeds a threshold, and the user is in the second preset time. If the product of the store is also collected, then the user's comprehensive score is W1+W2. Further, if the preset behavior data type also includes shopping information, and the user's industry preference is also consistent with the store's product. Then, the user's comprehensive score is W1+W2+W4.
  • Step 104 The server determines a delivery order of information to each user according to the user comprehensive score and the comprehensive score of other users.
  • the comprehensive score of the user can reflect the size of interest of the user and the store, and the higher the score indicates that the user has more interest in the store delivery information, then the user uses the The probability of placing information is also greater. Therefore, it is necessary to preferentially deliver information to users with high comprehensive scores to avoid the number of set delivery information, so that users with high comprehensive scores cannot obtain delivery information.
  • Step 105 The server delivers the information to the user according to the delivery sequence according to the type of delivery of the information and the preset delivery policy.
  • the server delivers the information to the user according to the delivery type of the information and the preset delivery policy according to the delivery sequence, specifically:
  • the server determines, according to the behavior data of the user and the industry pen price of the user, the type of delivery of the information by using the preset delivery policy;
  • the server delivers the information to the user according to the delivery order according to the type of delivery of the information.
  • the user's industry pen unit price is the price that the user purchases in an industry, and the average price per order is used to examine the purchasing ability of the user in the industry.
  • the user When the behavior data of the user satisfies the behavior standard set by the store, then the user is a potential user of the store, and then determines an industry to which the store belongs according to the commodity of the store, and then determines that the user is The industry's pen unit price for the industry to determine the purchasing power of the user.
  • the server compares the purchasing power of the user with the price corresponding to the user behavior data, and determines the type of information delivery according to the preset delivery policy. Specifically, the information is a coupon, and the coupon to be served has 300. Yuan minus 30 and over 200 yuan minus 20 two.
  • the behavior data of the user is an item of the store
  • the price of the item of the store is 200 yuan
  • the price of the industry pen of the user is 300
  • the preset delivery policy is The stimulating delivery strategy determines that the type of information to be delivered is a coupon of 300 yuan or less, which can stimulate users to purchase more products in the store. If the default delivery strategy is a robust delivery strategy, it is determined
  • the type of information to be delivered is a coupon of 200 yuan or less, which allows the user to directly consume, so that the probability of the user using the coupon increases.
  • the price of the item of the store in the collection is 200 yuan
  • the price of the industry pen of the user is 100
  • the coupon of 200 minus 20 is paid to the user.
  • the delivery is performed in accordance with the determined user delivery order.
  • the present application determines whether the behavior data of the user satisfies a preset behavior criterion by analyzing the behavior data of the user, and if yes, confirms the delivery order of the information to the user according to the determined comprehensive score of the user, and then according to the information.
  • the delivery type and the preset delivery policy deliver the information according to the delivery order.
  • the application can determine the potential users who need the delivery information, and not only can the potential users be accurately ranked by sorting the potential users and delivering the information.
  • the hit rate of the information delivery is also improved, and the user who has a great interest in the delivery information can be preferentially obtained, and the application directly analyzes the user based on the big data, thereby reducing the workload of the operator, and When the user's behavior changes, he can flexibly make corresponding judgments on the user's changes.
  • the server obtains the behavior data of the user, and specifically includes: the number of times of browsing a product of the store, whether the product is collected, the number of times the product is retrieved, industry preference, and shopping information, according to the The behavior data determines whether the user is a potential purchase user of the store.
  • the server determines whether the number of times the user browses the item of the store within 10 days exceeds 5 times, and if it exceeds, the user is listed as the shop or the shop. Potential buyers of goods.
  • the server determines whether the user is The item that has been stored in the store within 5 days, if it has been collected, the user is listed as a potential purchaser of the item at the store or the store.
  • the server determines whether the number of times the user searches for the item of the store within 10 days exceeds 5 times, and if it exceeds, the user is listed as the shop or the shop. Potential buyers of goods.
  • the server may determine the industry preference of the user according to the daily purchasing behavior of each user within one month (corresponding to the fourth preset time mentioned above), based on the result of the industry preference, the server determines The user's industry preference is the same as the product of a certain store. For example, if it is determined that the user's industry preference is to purchase a book, and the current store's product is a book, then the user's industry preference matches the store's product. The user is then listed as a potential purchaser for the store or the store.
  • the server may collect shopping information of each user within one month. If a user purchases a laptop within one month (equivalent to the aforementioned fifth preset time), then the server Through association analysis algorithms, such as: Apriori algorithm (a frequent item set algorithm for mining association rules), it is determined whether the goods of a certain store are associated with the shopping information of the user, and if there is an association, the user is listed as the Shop or potential purchaser of the store.
  • Apriori algorithm a frequent item set algorithm for mining association rules
  • the server collects behavior data of the user that meets a preset behavior criterion.
  • the server scores the two behavior data. And determining a comprehensive score of the user, for example: the number of times the user browses the store in the behavior data of the user and Collecting the products of the store to meet the preset behavior standard, the server scores according to the preset behavior data scores for the two behavior data of browsing the store and collecting the products of the store, and obtaining The combined score of the user.
  • the server sorts the composite scores of the plurality of users, and delivers the coupons prepared by the store according to the sorting.
  • the server Before the coupon is placed, the server further determines the type of the coupon to be placed. Specifically, the server determines the industry of the commodity of the store, and then determines the unit price of the user in the industry. If the number of times the user browses the store meets the set behavior standard, and the product of the user browsing the store is about 200 yuan, the product sold by the store belongs to an electronic product, and then the user is determined.
  • the industry pen unit price on the electronic product if the user's industry pen price on the electronic product is 300 yuan, the server determines the delivery type of the coupon according to a preset delivery policy, according to the user's delivery order
  • the corresponding coupons are delivered, for example, the coupons to be served are over 300 yuan minus 30 and over 200 yuan minus 20. If the default delivery strategy is a robust delivery strategy, the type of information to be served is determined to be 200. If the default delivery strategy is a stimulating delivery strategy, the type of information to be delivered is determined to be a coupon of 300 yuan or less.
  • the server selects a coupon of 200 yuan minus 20 for delivery.
  • the delivery is performed according to the delivery order of the user at the time of delivery.
  • the present application determines whether the behavior data of the user satisfies a preset behavior criterion by analyzing the behavior data of the user, and if yes, confirms the delivery order of the information to the user according to the determined comprehensive score of the user, and then according to the information.
  • the delivery type and the preset delivery policy deliver the information according to the delivery order.
  • the application can determine the potential users who need the delivery information, and not only can the potential users be accurately ranked by sorting the potential users and delivering the information.
  • the hit rate of the information delivery is also improved, and the user who has a great interest in the delivery information can be preferentially obtained, and the application directly analyzes the user based on the big data, thereby reducing the workload of the operator, and When the user's behavior changes, he can flexibly make corresponding judgments on the user's changes.
  • the present application also proposes a server, as described in FIG. 2, the server includes:
  • the obtaining module 21 is configured to acquire behavior data of the user
  • the determining module 22 is configured to determine whether the behavior data of the user meets a preset behavior criterion
  • the first determining module 23 is configured to determine a comprehensive score of the user according to the preset behavior data score if the behavior data of the user meets a preset behavior criterion;
  • the second determining module 24 determines, according to the user comprehensive score and the comprehensive score of other users, the order of delivery of the information to each user;
  • the delivery module 25 is configured to deliver the information to the user according to the delivery order according to the type of delivery of the information and the preset delivery policy.
  • the behavior data includes any combination of one or more of the following:
  • the number of views to the target object the collection of the target object, the retrieval of the target object, industry preferences, and historical purchase information.
  • the determining module is specifically configured to:
  • the behavior data is the number of times of browsing the target object, determining whether the number of times the user browses the target object exceeds a browsing threshold in the first preset time;
  • the behavior data is a collection situation of the target object, determining whether the user has collected the target object in the second preset time;
  • the behavior data is the number of retrievals to the target object, determining whether the number of times the user retrieves the target object exceeds a retrieval threshold in a third preset time;
  • the industry preference of the user is counted in a fourth preset time, and it is determined whether the industry preference of the user matches the industry of the target object;
  • the shopping information of the user is counted in a fifth preset time, and it is determined whether the shopping information of the user is associated with the target object.
  • the first determining module is specifically configured to:
  • a comprehensive score of the user is determined based on a score of the type of behavior data of the user.
  • the delivery module is specifically configured to:
  • the present application determines whether the behavior data of the user satisfies a preset behavior criterion by analyzing the behavior data of the user, and if yes, confirms the delivery order of the information to the user according to the determined comprehensive score of the user, After the information is delivered according to the delivery type of the information and the preset delivery policy, the application can determine the potential users who need the delivery information, and not only potential users but also potential users can be sorted and delivered. The precise positioning also improves the hit rate of information delivery, and also ensures that users who are interested in the delivery information can be preferentially obtained, and the application directly analyzes the user based on big data, thereby reducing the work of the operator. The burden, and when the user's behavior changes, can flexibly make corresponding judgments on the user's changes.
  • the present application can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is a better implementation. the way.
  • the technical solution of the present application which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for making a
  • the terminal device (which may be a cell phone, a personal computer, a server, or a network device, etc.) performs the methods described in various embodiments of the present application.
  • modules in the apparatus in the embodiments may be distributed in the apparatus of the embodiment according to the description of the embodiments, or the corresponding changes may be located in one or more apparatuses different from the embodiment.
  • the modules of the foregoing embodiments may be integrated into one or may be deployed separately; may be combined into one module, or may be further split into multiple sub-modules.
  • the serial numbers of the embodiments of the present application are merely for the description, and do not represent the advantages and disadvantages of the embodiments.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种信息投放用户的筛选方法和服务器，所述方法包括：服务器获取用户的行为数据（101）；所述服务器判断所述用户的行为数据是否满足预设的行为标准（102）；如果满足，所述服务器根据预设的行为数据得分确定所述用户的综合得分（103）；所述服务器根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序（104）；所述服务器根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息（105）。所述方法可对潜在用户实现精确定位，从而实现信息投放的命中率。

Description

一种信息投放用户的筛选方法和服务器
本申请要求2015年12月04日递交的申请号为201510884335.9、发明名称为“一种信息投放用户的筛选方法和设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及电子商务领域，特别是涉及一种信息投放用户的筛选方法和服务器。
背景技术
近年来随着电子商务的快速发展，越来越多的商家选择在网上开店。为了吸引买家，提高销量，很多卖家都会通过发放优惠券的方式来刺激买家在自己的店铺购买商品。卖家发放优惠券一般采用两种方法，一是对已经在本店铺购买过商品的老客户主动发放。由于卖家已经掌握了老客户的网购账号及手机号码，因而可以直接将优惠券发放到老客户的账户中并短信通知；二是直接将优惠券放在网上店铺显眼位置，由进入店铺的买家主动领取，当买家需要购买店铺商品时直接抵扣使用。
这两种由卖家自身发起的优惠券发放方法均有明显缺陷。对于第一种方法，很多店铺的商品往往购买了一次之后短时间内不会再购买第二次，例如大件商品，或者电子类产品，因而对老客户发放之后优惠券的使用率极低；而第二种方法没有对进入店铺的用户进行细分，所有买家均可以领取，领取后的利用率亦很低。
针对上述问题现有技术提出了一种解决方案，所述方案中包括，参数配置模块、接收模块和商品页文件生成模块。
参数配置用于根据配置的活动范围控制参数和活动信息展示控制参数生成用于向特定商户展示优惠券活动的活动页文件。其中，所述活动范围控制参数用于配置参与优惠券活动的商品类目，所述活动信息展示控制参数用于配置参与优惠券活动的商户。可以以数据库表的形式存储各个商户所对应的活动信息展示控制参数。通过配置该活动信息展示控制参数，可以将商户设置为可参与优惠券活动和不可参与优惠券活动，对于不可参与优惠券活动的商户，不生成所述活动页文件。
接收模块用于接收所述特定商户中的商户提交的活动请求。
商品页文件生成模块用于为提交活动请求的商户中满足审核规则的商品生成商品页 文件，所述商品页文件包括参加优惠券活动的商品信息及优惠券活动信息。为白名单中的商户生成所述商品页文件，所述白名单为预先设定的满足所述审核规则的商户名单。
在实现现有技术的过程，申请人发现现有技术至少存在以下问题：
现有技术中未对消费者的日常行为数据进行分析，不能精确划分出对店铺优惠券感兴趣的用户，且是手动设置参数，不仅不够智能，并且工作量较大，同时，以白名单的方式确定满足规定的用户会使用户发生变化时操作不够灵活。
发明内容
本申请的目的在于提供一种信息投放用户的筛选方法和服务器，通过分析用户的行为数据来确定出需要投放信息的潜在用户，不仅可以实现潜在用户的精准定位，还提高了信息投放的命中率。
本申请的技术方案如下：
一种信息投放用户的筛选方法，所述方法包括：
服务器获取用户的行为数据；
所述服务器判断所述用户的行为数据是否满足预设的行为标准；
如果满足，所述服务器根据预设的行为数据得分确定所述用户的综合得分；
所述服务器根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序；
所述服务器根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
所述行为数据包括以下的一种或多种的任意组合：
对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
所述服务器判断所述用户行为数据是否满足预设的行为标准，具体为：
当所述行为数据为对目标对象的浏览次数时，所述服务器判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
当所述行为数据为对目标对象的收藏情况时，所述服务器判断第二预设时间内所述用户是否收藏过所述目标对象；
当所述行为数据为对目标对象的检索次数时，所述服务器判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
当所述行为数据为行业偏好时，所述服务器统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
当所述行为数据为购物信息时，所述服务器统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
所述服务器根据预设的行为数据得分确定所述用户的综合得分，具体为：
所述服务器确定所述用户的行为数据的种类；
所述服务器根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
所述服务器根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
所述服务器根据所述信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息，具体为：
所述服务器根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
所述服务器根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
一种服务器，所述服务器包括：
获取模块，用于获取用户的行为数据；
判断模块，用于判断所述用户的行为数据是否满足预设的行为标准；
第一确定模块，如果所述用户的行为数据满足预设的行为标准，用于根据预设的行为数据得分确定所述用户的综合得分；
第二确定模块，用于根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序；
投放模块，用于根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
所述行为数据包括以下的一种或多种的任意组合：
对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
所述判断模块，具体用于：
当所述行为数据为对目标对象的浏览次数时，判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
当所述行为数据为对目标对象的收藏情况时，判断第二预设时间内所述用户是否收藏过所述目标对象；
当所述行为数据为对目标对象的检索次数时，判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
当所述行为数据为行业偏好时，统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
当所述行为数据为购物信息时，统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
所述第一确定模块，具体用于：
确定所述用户的行为数据的种类；
根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
所述投放模块，具体用于：
根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
本申请通过分析用户的行为数据来判断所述用户的行为数据是否满足预设的行为标准，如果满足，再根据确定的所述用户的综合得分确认信息向所述用户的投放顺序，然后根据信息的投放种类和预设的投放策略按照所述投放顺序投放所述信息，本申请可以确定出需要所述投放信息的潜在用户，通过对潜在用户进行排序并投放信息不仅可以实现潜在用户的精准定位，还提高了信息投放的命中率，同时还保证了对所述投放信息兴趣较大的用户可以优先获得，而且本申请还是直接基于大数据对用户进行分析，减轻了操作人员的工作负担，并且在用户的行为发生变化时可以灵活的针对用户的变化做出相应的判断。
附图说明
为了更清楚地说明本申请或现有技术中的技术方案，下面将对本申请或现有技术描述中所需要使用的附图作简单的介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，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。
图1为本申请实施例中的一种信息投放用户的筛选方法流程图；
图2为本申请实施例中的一种服务器的结构示意图。
具体实施方式
下面将结合本申请中的附图，对本申请中的技术方案进行清楚、完整的描述，显然，所描述的实施例是本申请的一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员获得的其他实施例，都属于本申请保护的范围。
如背景技术所述现有技术未对消费者的日常行为数据进行分析，不能精确划分出对店铺优惠券感兴趣的用户且是手动设置参数，不仅不够智能，并且工作量较大，同时，以白名单的方式确定满足规定的用户会使用户发生变化时操作不够灵活。
基于此，本申请实施例提出了一种信息投放用户的筛选方法，通过分析用户的行为数据来判断所述用户是满足所述信息的投放标准，在所述用户满足所述信息的投放标准后确定所述用户对所述信息的需求程度，并以此确定所述信息向所述用户的投放顺序，再根据信息的投放种类和预设的投放策略确定出所述用户对应的投放信息按照所述投放顺序投放所述信息，实现了投放信息对应用户的精确定位和投放信息的精确投放，提高了信息投放的命中率，还保证了对所述投放信息兴趣较大的用户可以优先获得。
如图1所述，为本申请实施例提出的一种信息投放用户的筛选方法的流程示意图，所述方法包括以下步骤：
步骤101，服务器获取用户的行为数据。
其中，所述行为数据包括以下的一种或多种的任意组合：
对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
需要说明的是，这里的目标对象，可以指的是一家店铺，也可以是一家店铺中所在出售的业务对象，而业务对象本身，可以是实体商品，也可以是服务，例如：洗车、养护、按摩、清洁、厨师上门、家政、家教、娱乐、吃喝、旅行、酒店、租车等，凡是可以属于用户需求范围的内容都可以作为本申请实施例中的分析对象，用以找出潜在的目标用户，这样的变化并不会影响本申请的保护范围，在本申请的后续说明中，目标对象的含义同样存在上述的限定，后文中不再重复说明。
所述服务器可以为电商平台的服务器，用户在日常网购时会在电商平台上留下操作痕迹，这些操作痕迹即为用户的行为数据，可以根据所述操作痕迹分析出用户潜在的业务对象需求或喜欢哪家店铺中在售的业务对象。
例如：以业务对象为实体商品的情况为例，如果一个用户多次浏览某店铺中在售的 商品，则表明用户对该店铺的商品比较喜欢，但是由于某些原因(如：价格或没有优惠等)而没有购买，通过用户的浏览该店铺的商品的次数这一信息可以确定出用户对哪家店铺的商品感兴趣，并且还可以进一步分析出是对该家店铺的何种商品感兴趣。
同样的，如果用户收藏了某店铺的商品、或者检索某个商品的次数达到了一定数量，或者用户的行业偏好(如某用户经常购买一些电子产品表明该用户对电子产品有偏好)，以及用户的购物信息与某些商品存在明显关联，则同样可以表明用户对这些商品感兴趣。
通过获取用户的行为数据可以分析出用户喜欢哪些商品，进而判断出所述用户是否对某店铺出售的商品相对应，从而得出所述用户是否为所述店铺的潜在用户。
当然，如果业务对象为具体的服务内容，例如家政服务，酒店业务等等，也存在上述的情况，可以得出用户是否为提供这些服务的店铺的潜在用户。
当然，用户的行为数据还可以包括用户在日常消费过程中在服务器(例如前述的电商平台)上留下的其他操作痕迹，凡是能够表明用户潜在的业务需求的行为信息均属于本申请的保护范围。
步骤102，所述服务器判断所述用户的行为数据是否满足预设的行为标准。
如果满足，则执行步骤103，如果不满足，则结束。
所述服务器判断所述用户行为数据是否满足预设的行为标准，具体为：
当所述行为数据为对目标对象的浏览次数时，所述服务器判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
当所述行为数据为对目标对象的收藏情况时，所述服务器判断第二预设时间内所述用户是否收藏过所述目标对象；
当所述行为数据为对目标对象的检索次数时，所述服务器判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
当所述行为数据为行业偏好时，所述服务器统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
当所述行为数据为购物信息时，所述服务器统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
具体的，针对不同的具体应用场景，同样以业务对象为实体商品的情况为例，对以上各情况进行说明如下：
情况一、行为数据为对目标对象的浏览次数。
在此种情况下，目标用户的确定依据为：当某用户浏览某一店铺的商品超过一定次 数时，表明所述用户对该店铺的商品有比较强烈的购买需求，但是可能由于价格或没有折扣等原因而没有购买，因此，可以确定所述用户为该商品的潜在购买用户，或者所述用户为所述店铺的潜在购买用户。
基于上述的依据，在一个店铺或该店铺所销售的商品有促销活动要投放信息时，可以获取第一预设时间内各用户对该店铺或该店铺内商品的浏览次数，并判断浏览次数是否超过预先设定的浏览阈值。如果超过，则相应的用户满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求比较强烈，因此，可以将该用户作为该店铺在投放信息时需要投放的用户之一。相反，如果没有超过，则相应的用户不满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求不太强烈，在该店铺投放信息时不需要投放给该用户。其中，所述第一预设时间和浏览阈值可以根据实际需要进行确定。
情况二、行为数据为对目标对象的收藏情况。
在此种情况下，目标用户的确定依据为：当某用户收藏过某店铺的商品时，表明所述用户对该店铺的这件商品有购买欲望或比较喜欢所述商品，但是可能由于价格或没有折扣等原因而没有购买，或者，当某用户收藏过某店铺时，表明所述用户对该店铺的商品有购买欲望或比较喜欢这个店铺，但是可能由于价格或没有折扣等原因而没有购买，因此，可以确定所述用户为该店铺的这件商品的潜在购买用户，或所述用户为所述店铺的潜在的购买用户。
基于上述的依据，在一个店铺或该店铺所销售的商品有促销活动要投放信息时，判断第二预设时间范围内各用户是否收藏过该店铺或者该店铺的商品。如果收藏过，则相应的用户满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求比较强烈，因此，可以将该用户作为该店铺在投放信息时需要投放的用户之一。相反，如果没有收藏过，则相应的用户不满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求不太强烈，在该店铺投放信息时不需要投放给该用户。其中，所述第二设时间可以根据实际需要进行确定。
情况三、行为数据为对目标对象的检索次数。
在此种情况下，目标用户的确定依据为：当某用户对某一店铺的商品检索次数超过一定次数时，表明所述用户对该店铺的商品有比较强烈的购买需求，但是可能由于价格或没有折扣等原因而没有购买，因此，可以确定所述用户为该商品的潜在购买用户，或者所述用户为所述店铺的潜在购买用户。
基于上述的依据，在一个店铺或该店铺所销售的商品有促销活动要投放信息时，可以获取第三预设时间内各用户对该店铺或该店铺内商品的检索次数，并判断检索次数是否超过预先设定的检索阈值。如果超过，则相应的用户满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求比较强烈，因此，可以将该用户作为该店铺在投放信息时需要投放的用户之一。相反，如果没有超过，则相应的用户不满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求不太强烈，在该店铺投放信息时不需要投放给该用户。其中，所述第三预设时间和检索阈值可以根据实际需要进行确定。
情况四、行为数据为行业偏好。
某些用户有行业偏好，例如：某用户的行业偏好是电子产品，那么，该用户平时肯定会关注或购买很多电子产品，或者某些用户在一段时间内对某些产品具有明显的偏好，例如：某用户在进行家装时，所述用户会对家装产品具有明显的行业偏好。如果某用户的行业偏好于一些店铺的商品相对应时，那么，该用户就成为了这些店铺的潜在买家。
由于用户的行业偏好可能是固定的，也可能是一段时间内具有某行业偏好，因此，需要统计第四预设时间范围内用户的行业偏好，根据统计到的行业偏好判断是否与某店铺的商品相吻合。如果吻合，则满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求比较强烈，因此，可以将该用户作为该店铺在投放信息时需要投放的用户之一。相反，如果不吻合，则相应的用户不满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求不太强烈，在该店铺投放信息时不需要投放给该用户。其中，所述第四设时间可以根据实际需要进行确定。
情况五、行为数据为购物信息。
用户的购物信息能够反映出该用户需要哪方面的商品，例如：某用户购买过笔记本电脑，那么该用户对笔记本电脑和与笔记本电脑有关的电脑配件可能比较需要，如果某店铺的商品与笔记本电脑或笔记本电脑有关的电脑配件有关，那么该用户就是该店铺的潜在买家。
基于以上的依据，可以统计第五预设时间内各用户的购物信息，根据各用户的购物信息判断是否与某店铺的商品有关联。如果有关联，则相应的用户满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求比较强烈，因此，可以将该用户作为该店铺在投放信息时需要投放的用户之一。相反，如果没有关联，则相应的用户不满足该店铺预设的行为标准，表明该用户对该店铺或该店铺在售的商品购买需求不 太强烈，在该店铺投放信息时不需要投放给该用户。其中，所述第五预设时间可以根据实际需要进行确定。
需要进行说明的是，上述的示例都是以业务对象为实体商品的情况为例进行说明的，如果业务对象具体为服务，相应的处理流程也与上述的方案相类似，可以以此类推，在此不再重复说明。
步骤103，所述服务器根据预设的行为数据得分确定所述用户的综合得分。
所述服务器根据预设的行为数据得分确定所述用户的综合得分，具体为：
所述服务器确定所述用户的行为数据的种类；
所述服务器根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
所述服务器根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
具体的，预先为行为数据设定得分，例如：浏览一个目标对象的得分为W1，收藏过该目标对象的得分为W2，检索过该目标对象的得分为W3，行业偏好与该目标对象相吻合的得分为W4，购物信息与该目标对象相关联的得分为W5。
用户可能有多种的行为数据都满足所述店铺预设的行为标准，例如：用户在第一预设时间内浏览一个店铺的商品的次数超过阈值，并且，该用户在第二预设时间内也收藏过该店铺的商品，那么，该用户的综合得分为W1+W2，进一步的，如果预设的行为数据类型中还包括购物信息，并且该用户的行业偏好也与该店铺的商品相吻合，那么,该用户的综合得分为W1+W2+W4。
步骤104，所述服务器根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序。
具体的，所述用户的综合得分能够反映出所述用户与所述店铺投放信息兴趣的大小，得分越高表示所述用户对所述店铺投放信息的兴趣越大，那么所述用户使用所述投放信息的概率也越大，因此需要优先向综合得分高的用户投放信息，以避免设定的投放信息数目的原因使综合得分高的用户无法获得投放信息。
步骤105，所述服务器根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
所述服务器根据所述信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息，具体为：
所述服务器根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
所述服务器根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
用户的行业笔单价为所述用户在某行业购买商品，平均每笔订单的价格，用于考查所述用户在所述行业的购买能力。
当所述用户的行为数据满足所述店铺设定的行为标准时，那么所述用户为所述店铺的潜在用户，再根据所述店铺的商品确定所述店铺所属的行业，然后确定所述用户在所述行业的行业笔单价，以确定所述用户的购买能力。
所述服务器将所述用户的购买能力和所述用户行为数据对应的价格进行对比，根据预设的投放策略确定信息投放种类，具体以信息为优惠券为例，要投放的优惠券有满300元减30和满200元减20两种。
当所述用户的行为数据为收藏过所述店铺的商品时，如果收藏的所述店铺的商品的价格为200元，所述用户的行业笔单价为300时，且如果预设的投放策略为刺激性投放策略，确定要投放的信息的种类为满300元减30的优惠券，这样可以刺激用户购买所述店铺中更多的商品，如果预设的投放策略为稳健型投放策略，确定要投放的信息的种类为满200元减20的优惠券，这样可以使用户直接进行消费，使用户使用所述优惠券的概率增加。当收藏的所述店铺的商品的价格为200元，所述用户的行业笔单价为100时，向用户投放满200减20的优惠券。其中，在投放时按照确定的用户投放顺序进行投放。
本申请通过分析用户的行为数据来判断所述用户的行为数据是否满足预设的行为标准，如果满足，再根据确定的所述用户的综合得分确认信息向所述用户的投放顺序，然后根据信息的投放种类和预设的投放策略按照所述投放顺序投放所述信息，本申请可以确定出需要所述投放信息的潜在用户，通过对潜在用户进行排序并投放信息不仅可以实现潜在用户的精准定位，还提高了信息投放的命中率，同时还保证了对所述投放信息兴趣较大的用户可以优先获得，而且本申请还是直接基于大数据对用户进行分析，减轻了操作人员的工作负担，并且在用户的行为发生变化时可以灵活的针对用户的变化做出相应的判断。
为了进一步阐述本申请的技术思想，现结合具体的应用场景，对本申请的技术方案进行说明，具体如下：
以业务对象为实体商品的情况为例，服务器获取用户的行为数据，具体包括：浏览店铺的一件商品的次数、是否收藏该商品、检索该商品的次数、行业偏好和购物信息，根据所述行为数据判断所述用户是否为所述店铺的潜在购买用户。
对于用户在短时间内频繁浏览一个店铺中的某个商品的情况，如果在服务器中设定的行为标准为10天(相当于前述第一预设时间)内浏览一家店铺中的某个商品超过5次(相当于前述浏览阈值)，则所述服务器判断该用户在10天内浏览该店铺的这件商品的次数是否超过5次，如果超过，则将该用户列为该店铺或该店铺的这件商品的潜在购买者。
对于用户收藏店铺商品的情况，如果在所述服务器中设定的行为标准为最近5天(相当于前述第二预设时间)内收藏过某个店铺的商品，则所述服务器判断该用户是否在5天内收藏过该店铺的商品，如果收藏过，则将该用户列为该店铺或该店铺的这件商品的潜在购买者。
对于用户在短时间内频繁检索一个店铺中的某个商品的情况，如果在服务器中设定的行为标准为10天(相当于前述第三预设时间)内检索一家店铺中的某个商品超过5次(相当于前述检索阈值)，则所述服务器判断该用户在10天内检索该店铺的这件商品的次数是否超过5次，如果超过，则将该用户列为该店铺或该店铺的这件商品的潜在购买者。
在另一种应用场景下，服务器可以根据各用户1个月内(相当于前述的第四预设时间)的日常购买行为确定出所述用户的行业偏好，基于这样的行业偏好结果，服务器判断该用户的行业偏好与某个店铺的商品是否相同，例如：如果确定用户的行业偏好为购买图书，而当前店铺的商品为图书的话，那么，该用户的行业偏好与这个店铺的商品相吻合，则将该用户列为该店铺或该店铺的潜在购买者。
在另一种应用场景下，所述服务器可以统计1个月内各用户的购物信息，如果一个用户在1个月(相当于前述的第五预设时间)内购买过笔记本电脑，那么，服务器通过关联分析算法，如：Apriori算法(一种挖掘关联规则的频繁项集算法)，确定出某个店铺的商品与该用户的购物信息是否存在关联，如果存在关联，则将该用户列为该店铺或该店铺的潜在购买者。
需要进行说明的是，上述的示例都是以业务对象为实体商品的情况为例进行说明的，如果业务对象具体为店铺本身或者服务，相应的处理流程也与上述的方案相类似，可以以此类推，在此不再重复说明。
在具体的处理过程中，服务器统计所述用户满足预设的行为标准的行为数据，当一个用户的行为数据中有2项满足预设的行为标准时，则服务器为这2项行为数据进行打分，并确定出该用户的综合得分，例如：所述用户的行为数据中浏览所述店铺的次数和 收藏所述店铺的商品都满足预设的行为标准，那么所述服务器根据预设的行为数据得分为浏览所述店铺的次数和收藏所述店铺的商品这两项行为数据进行评分，并得出所述用户的综合得分。
所述服务器将多个用户的综合得分进行排序，并按照所述排序投放所述店铺准备的优惠券。
在投放优惠券之前，所述服务器还要确定投放的优惠券的种类，具体的，所述服务器确定所述店铺的商品所述的行业，然后确定出所述用户在所述行业的行业笔单价，如果所述用户浏览所述店铺的次数满足设定的行为标准时，且所述用户浏览所述店铺的商品都在200元左右，所述店铺所售商品属于电子产品，然后确定出所述用户在电子产品上的行业笔单价，如果所述用户在电子产品上的行业笔单价为300元时，所述服务器根据预先设定的投放策略确定优惠券的投放种类，根据所述用户的投放顺序投放对应的优惠券，例如:要投放的优惠券有满300元减30和满200元减20两种,如果预设的投放策略为稳健型投放策略，确定要投放的信息的种类为满200元减20的优惠券,如果预设的投放策略为刺激性投放策略，确定要投放的信息的种类为满300元减30的优惠券。
当所述用户的行业笔单价小于所述用户浏览所述店铺的商品的价格时，所述服务器选择满200元减20的优惠券进行投放。其中，在投放时根据所述用户的投放顺序进行投放。
本申请通过分析用户的行为数据来判断所述用户的行为数据是否满足预设的行为标准，如果满足，再根据确定的所述用户的综合得分确认信息向所述用户的投放顺序，然后根据信息的投放种类和预设的投放策略按照所述投放顺序投放所述信息，本申请可以确定出需要所述投放信息的潜在用户，通过对潜在用户进行排序并投放信息不仅可以实现潜在用户的精准定位，还提高了信息投放的命中率，同时还保证了对所述投放信息兴趣较大的用户可以优先获得，而且本申请还是直接基于大数据对用户进行分析，减轻了操作人员的工作负担，并且在用户的行为发生变化时可以灵活的针对用户的变化做出相应的判断。
基于与上述方法同样的申请构思，本申请还提出了一种服务器，如图2所述，所述服务器包括：
获取模块21，用于获取用户的行为数据；
判断模块22，用于判断所述用户的行为数据是否满足预设的行为标准；
第一确定模块23，如果所述用户的行为数据满足预设的行为标准，用于根据预设的行为数据得分确定所述用户的综合得分；
第二确定模块24，根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序；
投放模块25，用于根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
所述行为数据包括以下的一种或多种的任意组合：
对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
所述判断模块，具体用于：
当所述行为数据为对目标对象的浏览次数时，判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
当所述行为数据为对目标对象的收藏情况时，判断第二预设时间内所述用户是否收藏过所述目标对象；
当所述行为数据为对目标对象的检索次数时，判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
当所述行为数据为行业偏好时，统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
当所述行为数据为购物信息时，统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
所述第一确定模块，具体用于：
确定所述用户的行为数据的种类；
根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
所述投放模块，具体用于：
根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
本申请通过分析用户的行为数据来判断所述用户的行为数据是否满足预设的行为标准，如果满足，再根据确定的所述用户的综合得分确认信息向所述用户的投放顺序，然 后根据信息的投放种类和预设的投放策略按照所述投放顺序投放所述信息，本申请可以确定出需要所述投放信息的潜在用户，通过对潜在用户进行排序并投放信息不仅可以实现潜在用户的精准定位，还提高了信息投放的命中率，同时还保证了对所述投放信息兴趣较大的用户可以优先获得，而且本申请还是直接基于大数据对用户进行分析，减轻了操作人员的工作负担，并且在用户的行为发生变化时可以灵活的针对用户的变化做出相应的判断。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到本申请可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台终端设备(可以是手机，个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述的方法。
以上所述仅是本申请的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本申请原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视本申请的保护范围。
本领域技术人员可以理解实施例中的装置中的模块可以按照实施例描述进行分布于实施例的装置中，也可以进行相应变化位于不同于本实施例的一个或多个装置中。上述实施例的模块可以集成于一体，也可以分离部署；可以合并为一个模块，也可以进一步拆分成多个子模块。上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
以上公开的仅为本申请的几个具体实施例，但是，本申请并非局限于此，任何本领域的技术人员能思之的变化都应落入本申请的保护范围。

Claims (10)

  1. 一种信息投放用户的筛选方法，其特征在于，所述方法包括：
    服务器获取用户的行为数据；
    所述服务器判断所述用户的行为数据是否满足预设的行为标准；
    如果满足，所述服务器根据预设的行为数据得分确定所述用户的综合得分；
    所述服务器根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序；
    所述服务器根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
  2. 如权利要求1所述方法，其特征在于，所述行为数据包括以下的一种或多种的任意组合：
    对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
  3. 如权利要求2所述方法，其特征在于，所述服务器判断所述用户行为数据是否满足预设的行为标准，具体为：
    当所述行为数据为对目标对象的浏览次数时，所述服务器判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
    当所述行为数据为对目标对象的收藏情况时，所述服务器判断第二预设时间内所述用户是否收藏过所述目标对象；
    当所述行为数据为对目标对象的检索次数时，所述服务器判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
    当所述行为数据为行业偏好时，所述服务器统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
    当所述行为数据为购物信息时，所述服务器统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
  4. 如权利要求1所述方法，其特征在于，所述服务器根据预设的行为数据得分确定所述用户的综合得分，具体为：
    所述服务器确定所述用户的行为数据的种类；
    所述服务器根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
    所述服务器根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
  5. 如权利要求1所述方法，其特征在于，所述服务器根据所述信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息，具体为：
    所述服务器根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
    所述服务器根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
  6. 一种服务器，其特征在于，所述服务器包括：
    获取模块，用于获取用户的行为数据；
    判断模块，用于判断所述用户的行为数据是否满足预设的行为标准；
    第一确定模块，如果所述用户的行为数据满足预设的行为标准，用于根据预设的行为数据得分确定所述用户的综合得分；
    第二确定模块，用于根据所述用户综合得分和其他用户的综合得分确定信息向各个用户的投放顺序；
    投放模块，用于根据信息的投放种类和预设的投放策略按照所述投放顺序向所述用户投放所述信息。
  7. 如权利要求6所述服务器，其特征在于，所述行为数据包括以下的一种或多种的任意组合：
    对目标对象的浏览次数、对目标对象的收藏情况、对目标对象的检索情况、行业偏好和历史购买信息。
  8. 如权利要求7所述服务器，其特征在于，所述判断模块，具体用于：
    当所述行为数据为对目标对象的浏览次数时，判断第一预设时间内所述用户浏览所述目标对象的次数是否超过浏览阈值；
    当所述行为数据为对目标对象的收藏情况时，判断第二预设时间内所述用户是否收藏过所述目标对象；
    当所述行为数据为对目标对象的检索次数时，判断第三预设时间内所述用户检索所述目标对象的次数是否超过检索阈值；
    当所述行为数据为行业偏好时，统计第四预设时间内所述用户的行业偏好，判断所述用户的行业偏好是否与所述目标对象的所属行业相匹配；
    当所述行为数据为购物信息时，统计第五预设时间内所述用户的购物信息，判断所述用户的购物信息是否与所述目标对象有关联。
  9. 如权利要求6所述服务器，其特征在于，所述第一确定模块，具体用于：
    确定所述用户的行为数据的种类；
    根据预设的行为数据得分为所述用户的行为数据的种类进行打分；
    根据所述用户的行为数据的种类的打分确定所述用户的综合得分。
  10. 如权利要求6所述服务器，其特征在于，所述投放模块，具体用于：
    根据所述用户的行为数据和所述用户的行业笔单价通过所述预设的投放策略确定所述信息的投放种类；
    根据所述信息的投放种类按照所述投放顺序向所述用户投放所述信息。
PCT/CN2016/107078 2015-12-04 2016-11-24 一种信息投放用户的筛选方法和服务器 Ceased WO2017092602A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510884335.9A CN106846019A (zh) 2015-12-04 2015-12-04 一种信息投放用户的筛选方法和设备
CN201510884335.9 2015-12-04

Publications (1)

Publication Number Publication Date
WO2017092602A1 true WO2017092602A1 (zh) 2017-06-08

Family

ID=58796257

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/107078 Ceased WO2017092602A1 (zh) 2015-12-04 2016-11-24 一种信息投放用户的筛选方法和服务器

Country Status (2)

Country Link
CN (1) CN106846019A (zh)
WO (1) WO2017092602A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110175858A (zh) * 2019-03-27 2019-08-27 阿里巴巴集团控股有限公司 一种调整发放电子资源的方法及系统
CN111198984A (zh) * 2018-11-16 2020-05-26 阿里巴巴集团控股有限公司 用户群体筛选方法、装置及电子设备
CN112488768A (zh) * 2020-12-10 2021-03-12 深圳市欢太科技有限公司 特征提取方法、特征提取装置、存储介质与电子设备
CN114697889A (zh) * 2020-12-31 2022-07-01 广州三星通信技术研究有限公司 处理信息的方法及其装置

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107516236A (zh) * 2017-07-22 2017-12-26 长沙兔子代跑网络科技有限公司 一种根据用户行为数据挖掘代跑客户的方法及装置
CN107526778A (zh) * 2017-07-22 2017-12-29 长沙兔子代跑网络科技有限公司 一种根据用户行为数据挖掘代跑客户的方法及装置
CN107979639A (zh) * 2017-11-21 2018-05-01 重庆金窝窝网络科技有限公司 基于区块链的动作处理方法及装置
CN110197408B (zh) * 2018-02-26 2021-06-29 苏宁易购集团股份有限公司 保价商品的定价预警方法及系统
CN109636530B (zh) * 2018-12-14 2021-02-09 拉扎斯网络科技（上海）有限公司 产品确定方法、装置、电子设备及计算机可读存储介质
CN111429163B (zh) * 2019-01-10 2023-08-15 百度在线网络技术（北京）有限公司 户外广告投放资源的推荐方法、装置及计算机设备
CN113643073A (zh) * 2021-09-01 2021-11-12 深圳市聚商鼎力网络技术有限公司 一种基于大数据的自动化信息投放系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130268317A1 (en) * 2010-12-07 2013-10-10 Digital Foodie Oy Arrangement for facilitating shopping and related method
CN103995839A (zh) * 2014-04-30 2014-08-20 兴天通讯技术（天津）有限公司 基于协同过滤的商品推荐优化方法和系统
CN104331819A (zh) * 2014-11-05 2015-02-04 中国建设银行股份有限公司 信息处理方法及装置
CN104539740A (zh) * 2015-01-26 2015-04-22 北京奇艺世纪科技有限公司 一种信息推送方法及装置
US20150242866A1 (en) * 2012-09-27 2015-08-27 Nec Corporation Determination device, determination method, and computer-readable recording medium
CN104901817A (zh) * 2014-03-07 2015-09-09 腾讯科技（北京）有限公司 目标信息推送方法及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714067B (zh) * 2012-09-29 2018-01-26 腾讯科技（深圳）有限公司 一种信息推送方法和装置
CN103729780A (zh) * 2013-12-23 2014-04-16 北京掌阔技术有限公司 一种广告优化投放方法和系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130268317A1 (en) * 2010-12-07 2013-10-10 Digital Foodie Oy Arrangement for facilitating shopping and related method
US20150242866A1 (en) * 2012-09-27 2015-08-27 Nec Corporation Determination device, determination method, and computer-readable recording medium
CN104901817A (zh) * 2014-03-07 2015-09-09 腾讯科技（北京）有限公司 目标信息推送方法及装置
CN103995839A (zh) * 2014-04-30 2014-08-20 兴天通讯技术（天津）有限公司 基于协同过滤的商品推荐优化方法和系统
CN104331819A (zh) * 2014-11-05 2015-02-04 中国建设银行股份有限公司 信息处理方法及装置
CN104539740A (zh) * 2015-01-26 2015-04-22 北京奇艺世纪科技有限公司 一种信息推送方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111198984A (zh) * 2018-11-16 2020-05-26 阿里巴巴集团控股有限公司 用户群体筛选方法、装置及电子设备
CN110175858A (zh) * 2019-03-27 2019-08-27 阿里巴巴集团控股有限公司 一种调整发放电子资源的方法及系统
CN112488768A (zh) * 2020-12-10 2021-03-12 深圳市欢太科技有限公司 特征提取方法、特征提取装置、存储介质与电子设备
CN114697889A (zh) * 2020-12-31 2022-07-01 广州三星通信技术研究有限公司 处理信息的方法及其装置

Also Published As

Publication number Publication date
CN106846019A (zh) 2017-06-13

Similar Documents

Publication Publication Date Title
WO2017092602A1 (zh) 一种信息投放用户的筛选方法和服务器
CN113205382B (zh) 确定对象的方法、装置、设备、存储介质和程序产品
JP6753578B2 (ja) 電子配信プラットフォーム上で提示対象を選択および推奨するための方法および装置
TW201501059A (zh) 訊息推薦方法及裝置
JP6679451B2 (ja) 選択装置、選択方法および選択プログラム
CN105096151A (zh) 信息推荐方法及装置、服务器
CN106980999A (zh) 一种用户推荐的方法和设备
JP5957164B1 (ja) 情報処理装置、情報処理方法、及び情報処理プログラム
WO2017158798A1 (ja) 情報処理装置、情報配信システム、情報処理方法、および情報処理プログラム
JP2014522056A (ja) 検索結果へのスクリーニング情報の適用
CN107944897A (zh) 数据标签分发系统、方法、服务器和客户端
CN102073737A (zh) 一种自动提供网络销售商实时采购建议的方法和系统
KR20220117425A (ko) 빅데이터를 활용한 시장성 분석 및 사업화방법론 분석 시스템
JP6702628B1 (ja) 提供装置、提供方法および提供プログラム
JP2016512351A (ja) ユーザの買い物履歴、小売業者のｉｄ及び小売業者が販売促進する物品に基づいて、カスタマイズされた検索結果を提供するシステム及び方法
JP7271151B2 (ja) 情報処理装置、情報処理方法および情報処理プログラム
JP6679706B1 (ja) 情報処理装置、情報処理方法、及び情報処理プログラム
CN114549127B (zh) 订单对象替换方法、装置、电子设备和计算机存储介质
JP6917348B2 (ja) 提供装置、提供方法、及び提供プログラム
JP2020107269A (ja) 情報処理装置、情報処理方法および情報処理プログラム
JP2020140242A (ja) 情報処理装置、情報処理方法及び情報処理プログラム
TWI726981B (zh) 資訊投放用戶的篩選方法和設備
CN107239988A (zh) 商品数据呈现系统
KR102014773B1 (ko) 소셜 커머스와 소셜 네트워크를 통한 선물 추천 방법
JP6450040B1 (ja) 情報処理装置、情報処理方法および情報処理プログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16869912

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16869912

Country of ref document: EP

Kind code of ref document: A1