WO2017148272A1 - 一种目标用户的识别方法和装置 - Google Patents

一种目标用户的识别方法和装置 Download PDF

Info

Publication number
WO2017148272A1
WO2017148272A1 PCT/CN2017/073866 CN2017073866W WO2017148272A1 WO 2017148272 A1 WO2017148272 A1 WO 2017148272A1 CN 2017073866 W CN2017073866 W CN 2017073866W WO 2017148272 A1 WO2017148272 A1 WO 2017148272A1
Authority
WO
WIPO (PCT)
Prior art keywords
index
user
behavior
identified
determining
Prior art date
Application number
PCT/CN2017/073866
Other languages
English (en)
French (fr)
Inventor
胡于响
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2017148272A1 publication Critical patent/WO2017148272A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data
    • G06Q30/0203Market surveys; Market polls

Definitions

  • the present application relates to the field of Internet technologies, and in particular, to a method for identifying a target user and a device for identifying a target user.
  • social marketing is an important marketing tool.
  • the traditional social marketing is to establish contact with the buyer through the e-commerce platform or the seller, and then to recommend the new customer by the buyer.
  • This kind of marketing has many shortcomings. One is that it requires more manpower to operate, and it costs more. The second is that the effect is poor, because the platform or the seller is not clear whether the buyer is authoritative and has sufficient influence.
  • the input and output in marketing is relatively low.
  • embodiments of the present application have been made in order to provide a target user identification method and a corresponding target user identification device that overcome the above problems or at least partially solve the above problems.
  • the present application discloses a method for identifying a target user, including:
  • the behavior information includes a user behavior number and a user behavior object, where the behavior information is
  • the step of determining the behavior index of the user to be identified includes:
  • the behavior index is determined using the number of times index and the object index.
  • the step of determining the number of times of the user to be identified for the number of times of the user behavior includes:
  • the user behavior object includes a first attribute behavior object and a second attribute behavior object
  • the step of determining an object index of the to-be-identified user for the user behavior object includes:
  • the object index of the user to be identified is determined by using the first attribute behavior object index and the second attribute behavior object index.
  • the method before the step of determining the first attribute behavior object index of the to-be-identified user for the first attribute behavior object, the method further includes:
  • the time of occurrence of the user behavior for the first attribute behavior object is determined.
  • the determining, by the first attribute behavior object, the first attribute behavior object index of the user to be identified includes:
  • the determining, by the second attribute behavior object, the second attribute behavior object index of the user to be identified includes:
  • the step of determining the behavior index by using the number of times index and the object index comprises:
  • the number of times index, the first attribute behavior object index, and the second attribute behavior object index Weighted summation to obtain the behavioral index.
  • association information includes the number of associated users and associated object information
  • determining, by the association information, the association index of the user to be identified includes:
  • the association index is determined using the user association index and the object association index.
  • the step of determining the user association index of the to-be-identified user for the number of the associated users includes:
  • the user association index is a second user association index.
  • the determining, by the association object information, the object association index of the user to be identified includes:
  • the determining, according to the similarity, the object association index of the user to be identified includes:
  • the step of determining the association index by using the user association index and the object association index comprises:
  • the determining, according to the behavior index and the association index, whether the user to be identified is a target user includes:
  • an identification device for a target user including:
  • a user information obtaining module configured to acquire user information of the user to be identified, where the user information includes behavior information and associated information;
  • a behavior index determining module configured to determine, according to the behavior information, a behavior index of the user to be identified
  • a correlation index determining module configured to determine, according to the association information, a correlation index of the user to be identified
  • the target user identification module is configured to determine, according to the behavior index and the association index, whether the user to be identified is a target user.
  • the behavior information includes a user behavior number and a user behavior object
  • the behavior index determination module includes:
  • a number index determining submodule configured to determine an index of times of the user to be identified for the number of times of the user behavior
  • An object index determining submodule configured to determine an object index of the user to be identified for the user behavior object
  • the behavior index determination sub-module is configured to determine the behavior index by using the number of times index and the object index.
  • the number of times index determining submodule includes:
  • the user behavior number determining sub-module is configured to determine whether the number of user behaviors in the preset time period is greater than a first preset threshold
  • the first number index determining submodule is configured to determine, when the number of user behaviors in the preset time period is greater than the first preset threshold, that the number of times index is the first number of times index;
  • the second number index determining submodule is configured to determine that the number of times index is the second number of times index when the number of user actions in the preset time period is less than the first preset threshold.
  • the user behavior object includes a first attribute behavior object and a second attribute behavior object
  • the object index determination submodule includes:
  • a first attribute behavior object index determining submodule configured to determine, for the first attribute behavior object, a first attribute behavior object index of the to-be-identified user
  • a second attribute behavior object index determining submodule configured to determine, for the second attribute behavior object, a second attribute behavior object index of the user to be identified
  • the user object index determining submodule to be used is configured to determine an object index of the user to be identified by using the first attribute behavior object index and the second attribute behavior object index.
  • the object index determining submodule further includes:
  • a first attribute behavior object determining submodule configured to determine whether the first attribute behavior object is included in the user behavior object
  • the user behavior occurrence time determining submodule is configured to determine a time of occurrence of the user behavior for the first attribute behavior object when the first attribute behavior object is included in the user behavior object.
  • the first attribute behavior object index determining submodule includes:
  • a first attribute behavior object quantity obtaining submodule configured to acquire the number of the first attribute behavior objects whose occurrence time is earlier than a preset time
  • a first attribute behavior object ratio determining submodule configured to determine, according to the quantity of the first attribute behavior object, that the first attribute behavior object whose occurrence time is earlier than the preset time is occupied by the user behavior object proportion.
  • the second attribute behavior object index determining submodule includes:
  • a second attribute behavior object quantity obtaining submodule configured to acquire the quantity of the second attribute behavior object
  • the second attribute behavior object ratio determining submodule is configured to determine, according to the quantity of the second attribute behavior object, a proportion of the second attribute behavior object in the user behavior object.
  • the behavior index determining submodule includes:
  • the behavior index weighted summation submodule is configured to weight the sum of the number of times index, the first attribute behavior object index, and the second attribute behavior object index to obtain the behavior index.
  • association information includes the number of associated users and associated object information
  • association index determining module includes:
  • a user association index determining submodule configured to determine a user association index of the to-be-identified user for the number of the associated users
  • An object association index determining submodule configured to determine an object association index of the user to be identified for the associated object information
  • the association index determining submodule is configured to determine the association index by using the user association index and the object association index.
  • the user association index determining submodule includes:
  • the associated user quantity determining sub-module is configured to determine whether the number of the associated users is greater than a second preset threshold
  • a first user association index determining submodule configured to determine, when the number of the associated users is greater than a second preset threshold, that the user association index is a first user association index
  • the second user association index determining submodule is configured to determine that the user association index is a second user association index when the number of the associated users is less than a second preset threshold.
  • the object association index determining submodule includes:
  • a similarity calculation sub-module configured to calculate a similarity between the user to be identified and other users one by one for the associated object information
  • the to-be-identified user object association index determining sub-module is configured to determine an object association index of the to-be-identified user according to the similarity.
  • the to-be-identified user object association index determining sub-module includes:
  • a user quantity searching unit configured to find out the number of users whose similarity with the to-be-identified user is greater than a third preset threshold
  • the object association index determining unit is configured to determine an object association index of the user to be identified according to the number of users.
  • association index determining submodule includes:
  • a correlation index weighted summation submodule configured to weight the user association index and the object association index to obtain a correlation index of the user to be identified.
  • the target user identification module includes:
  • An index sorting sub-module configured to respectively sort the behavior index and the correlation index to obtain a behavior index ranking ratio and a correlation index ranking ratio
  • a sorting ratio comparison submodule configured to compare the behavior index ranking ratio with a first preset ratio, and/or, the correlation index sorting ratio is compared with a second preset ratio
  • the target user identification submodule is configured to determine, according to the comparison result, whether the user to be identified is a target user.
  • the embodiments of the present application include the following advantages:
  • the embodiment of the present application can obtain the user information of the user to be identified, and then determine the behavior index and the association index of the user to be identified according to the user information, thereby determining whether the user to be identified is the target user, and the identification method can be implemented. Standardization and quantification make the identification of target users more accurate and effective.
  • the embodiment of the present application determines the user's authority by determining the behavior index of the user's purchase of the product (such as the favorable rate) and the time point of purchasing the product; determining the association index through the strong relationship and the weak relationship of the user, and determining the user Whether it has influence, wherein the weak relationship is similar to the commodity, which can specifically characterize the potential influence of the user on others, further clarify the calculation method of the user behavior index and the association index, and help to quickly identify Out of the target user, it helps the merchant to conduct social marketing, operation selection and category recommendation based on the identified target users, so that the marketing and recommendation of the merchant is more targeted.
  • Embodiment 1 is a flow chart showing the steps of Embodiment 1 of a method for identifying a target user according to the present application;
  • FIG. 2 is a flow chart of steps of a second embodiment of a method for identifying a target user according to the present application
  • FIG. 3 is a schematic diagram of determining a user's behavior index of the present application.
  • Figure 4 is a schematic diagram showing the growth of a certain commodity sales volume of the present application.
  • FIG. 5 is a structural block diagram of an embodiment of an identification device for a target user according to the present application.
  • FIG. 1 a flow chart of a first embodiment of a method for identifying a target user of the present application is shown, which may specifically include the following steps:
  • Step 101 Acquire user information of a user to be identified.
  • the user information may include behavior information and associated information.
  • the behavior information may be a certain behavior of the recorded user, such as an introduction information of a user browsing an item, collecting the item in a favorite, or purchasing the item.
  • the association information may be other users directly associated with the user, for example, friend information of the user; or other users indirectly associated with the user, for example, information of other users who have browsed or purchased a certain product. .
  • Step 102 Determine, according to the behavior information, a behavior index of the user to be identified;
  • Step 103 Determine, according to the association information, an association index of the user to be identified;
  • the behavior index and the association index of the user to be identified may be determined for the behavior information and the association information, respectively.
  • Step 104 Determine, according to the behavior index and the association index, whether the user to be identified is a target user.
  • the behavior index and the association index can be directly added, and the result is compared with the preset judgment condition. Comparing to identify whether the user is a target user, and also assigning different weight values to the behavior index and the association index, and then judging the result of the weighted summation to identify the target user, or separately determining the behavior index And the correlation index is compared with the preset judgment condition, and the judgment of whether the user to be identified is the target user is obtained.
  • the technical person skilled in the art can select the manner of identifying the target user according to actual needs, which is not specifically limited in this application.
  • the user information of the user to be identified is obtained, and then the behavior index and the association index of the user to be identified are determined according to the user information, thereby determining whether the user to be identified is a target user, Standardization and quantification of the identification method are implemented, making the identification of the target user more accurate and effective.
  • the method may include the following steps:
  • Step 201 Obtain user information of the user to be identified.
  • the user information may include behavior information and associated information of the user on a certain e-commerce platform.
  • the behavior information may be a certain behavior of the recorded user, such as an introduction information of a user browsing an item, collecting the item in a favorite, or purchasing the item.
  • the association information may be other users directly associated with the user, for example, friend information of the user; or other users indirectly associated with the user, for example, information of other users who have browsed or purchased a certain product. .
  • user information may be obtained through a data warehouse of the e-commerce platform.
  • the data warehouse is generally used for data reading and writing, and can store user transaction information table, commodity transaction information table, product DSR (Detail Seller Rating, seller service rating system) information table and the like.
  • product DSR Delivery Seller Rating, seller service rating system
  • Step 202 Determine, according to the number of times of the user behavior, an index of times of the user to be identified;
  • the number of user behaviors may be user transaction volume data.
  • the step of determining the number of times of the user to be identified for the number of times of the user behavior may specifically include the following sub-steps:
  • Sub-step 2021 determining whether the number of user behaviors in the preset time period is greater than a first preset threshold
  • Sub-step 2022 if yes, determining that the number of times index is the first number of times index
  • Sub-step 2023 if no, determining that the number of times index is the second number of times index.
  • determining the number of times of the user to be identified according to the number of times of the user behavior may be obtained by comparing the number of times of the user behavior in the preset time period with a preset threshold.
  • the preset time period may be 30 days, 90 days, or 180 days, etc., and the length of the preset time period may be determined by a person skilled in the art according to actual needs, which is not specifically limited in this application.
  • the transaction amount data of the user to be identified within 90 days may be extracted, and then compared with a first preset threshold. If the transaction amount is greater than the first preset threshold, the number of times index may be determined as the first number of times index. And if the transaction amount is less than or equal to the first preset threshold, the number of times index may be determined as the second number of times index.
  • the first preset threshold may be specifically set according to the length of the preset time period. For example, if the preset time period is set to 30 days, the first preset threshold may be set to 10 times, that is, the transaction within 30 days. The amount is 10 times. If the preset time period is set to 90 days, the first preset threshold can be set to 45 times, that is, the transaction volume reaches 90 times in 90 days.
  • the first number of times index should be greater than the second number of times index.
  • the second number of times index can be set accordingly to be 0.5.
  • the size of the first-order index and the second-number index may also be determined by a person skilled in the art according to actual needs, which is not specifically limited in the present application.
  • Step 203 Determine, according to the user behavior object, an object index of the user to be identified.
  • the user behavior object may include a first attribute behavior object and a second attribute behavior object.
  • the first attribute behavior object may be a commodity with a high sales volume (explosive product)
  • the second property behavior object may be a commodity with a high favorable rate (excellent product).
  • the step of determining an object index of the user to be identified for the user behavior object may specifically include the following sub-steps:
  • Sub-step 2031 determining, for the first attribute behavior object, a first attribute behavior object index of the user to be identified
  • first determining whether the first attribute behavior object is included in the user behavior object, that is, the user Whether the purchased product includes a product with a high sales amount, and if so, the occurrence time of the user behavior for the first attribute behavior object may be further determined.
  • the user behavior can occur at a specific time when the user purchases such a high-volume item.
  • the sub-step of determining, by the first attribute behavior object, the first attribute behavior object index of the user to be identified may further include:
  • the preset time may be a time when the sales volume growth rate reaches a certain value.
  • FIG. 4 it is a schematic diagram of the growth of the sales volume of a certain product of the present application, wherein the horizontal axis is the time (unit: day) of the product on-line sales, and the vertical axis is the sales volume. It can be seen from the figure that the sales volume in the first 1-3 days is slower. After the third day, the sales volume of the goods increases due to various reasons (such as good quality of goods, spontaneous promotion, promotion, repeat customers of previous buyers). The speed is obviously faster, and the growth rate reaches the maximum on the 6th-7th day. After the 7th day, for some reasons (such as the emergence of market replicas, competitors, etc.), the sales growth rate begins to fall.
  • T the critical point
  • the preset time may be different, and a specific value of the critical point may be determined by a person skilled in the art according to actual needs, which is not specifically limited in this application.
  • the number of items whose purchase time is earlier than the preset time can be obtained, and then the proportion of the item in all the items is determined.
  • Sub-step 2032 determining, for the second attribute behavior object, a second attribute behavior object index of the user to be identified
  • the determining, by the second attribute behavior object, the sub-step of determining the second attribute behavior object index of the user to be identified may further include:
  • the product with a favorable rate of 99% or more can be considered as a high-quality product.
  • the number of items with a favorable rate of 99% or more can be obtained from the user information, and then the proportion of the goods in all the goods is determined.
  • Sub-step 2033 determining the object index of the user to be identified by using the first attribute behavior object index and the second attribute behavior object index.
  • the first attribute behavior object index and the second attribute behavior object index may be directly added to obtain an object index of the user to be identified, and the first attribute behavior object index may also be The second attribute behavior object index is respectively assigned different weight values, and then the result of the weighted summation is used as the object index of the user to be identified.
  • a person skilled in the art can select the manner in which the object index of the user to be identified is determined according to actual needs, which is not specifically limited in the present application.
  • Step 204 using the number of times index and the object index to determine the behavior index
  • the step of determining the behavior index by using the number of times index and the object index may specifically include the following substeps:
  • Sub-step 2041 weighting the number of times index, the first attribute behavior object index, and the second attribute behavior object index to obtain the behavior index.
  • the behavior index of the user to be identified may be calculated by using the following formula:
  • a is the index of the number of times the user is to be identified
  • b is the first attribute behavior object index
  • c is the second attribute behavior object index
  • w 1 , w 2 , w 3 are the weight values of a, b, and c, respectively.
  • w 1 + w 2 + w 3 100
  • the first attribute behavior object index b 0.1
  • the second attribute behavior object index c 0.9
  • w 1 30
  • w 2 50
  • w 3 20
  • Step 205 Determine, according to the number of the associated users, a user association index of the user to be identified.
  • the number of the associated users may be the number of friends of the user on the e-commerce platform, and the friend relationship reflects a strong relationship between the users.
  • the determining, for the number of the associated users, determining the to-be-identified may specifically include the following sub-steps:
  • Sub-step 2051 determining whether the number of the associated users is greater than a second preset threshold
  • Sub-step 2052 if yes, determining that the user association index is a first user association index
  • Sub-step 2053 if no, determining that the user association index is a second user association index.
  • the number of friends of the user to be identified may be compared with a preset threshold to determine a user association index.
  • the preset threshold may be 100, 150, or 180, etc., and the size of the preset threshold may be determined by a person skilled in the art according to actual needs, which is not specifically limited in this application.
  • the second preset threshold may be set to 150, and then the number of friends of the user to be identified is compared with a second preset threshold. If the number of friends is greater than the second preset threshold, the user association index may be determined as The first user association index may be determined to be a second user association index if the number of friends is less than or equal to a second preset threshold.
  • the first number of times index should be greater than the second number of times index.
  • user A has 100 buyer friends
  • user B has 200 buyer friends.
  • the second preset threshold is 150
  • the number can be A user association index is set to 0.7
  • the first user association index is set to 0.4 correspondingly, that is, user A's user association index is 0.4
  • user B's user association index is 0.8
  • the first user association index and the second user association index are The size may also be determined by a person skilled in the art according to actual needs, which is not specifically limited in the present application.
  • Step 206 Determine an object association index of the user to be identified for the associated object information.
  • the associated object information may be information determined according to a user behavior object, for example, a certain commodity is purchased at the same time.
  • the step of determining the object association index of the user to be identified for the associated object information may specifically include the following sub-steps:
  • Sub-step 2061 calculating the similarity between the user to be identified and other users one by one for the associated object information
  • the similarity between users can be calculated according to the products purchased by the user. When the similarity reaches a certain level, it can be considered that there is a weak relationship between the two users.
  • the similarity between the user to be identified and other users may be calculated one by one by using the following formula:
  • each user of the e-commerce platform may be scanned to calculate the similarity between the user to be identified and each of the other users.
  • Sub-step 2062 determining an object association index of the user to be identified according to the similarity.
  • the sub-step of determining the object association index of the user to be identified according to the similarity may further include:
  • a weak relationship is established between the users, and then the number of users whose similarity with the user to be identified is greater than 0.8 is found. Then, according to the number of users, an object association index of the user to be identified is determined.
  • a specific size of the third preset threshold ⁇ can be determined by a person skilled in the art according to the needs of the service, which is not specifically limited in this application.
  • Step 207 Determine the association index by using the user association index and the object association index.
  • the step of determining the association index by using the user association index and the object association index may specifically include the following sub-steps:
  • Sub-step 2071 weighting the user association index and the object association index to obtain a correlation index of the user to be identified.
  • association index of the user to be identified may be calculated by using the following formula:
  • P is the user association index of the user to be identified
  • Q is the object association index
  • w 1 , w 2 are the weight values of P and Q, respectively
  • w 1 +w 2 100, and those skilled in the art can The weight value is adjusted according to the needs of the specific target, which is not specifically limited in this application.
  • Step 208 Determine, according to the behavior index and the association index, whether the user to be identified is a target Household.
  • whether the user to be identified is the target user may be determined according to the behavior index and the association index.
  • the behavior index of the user to be identified may be added to the association index, and the user to be identified may be identified according to the obtained result, and the user may also be identified according to the behavior index or the association index, respectively.
  • the step of determining whether the user to be identified is a target user according to the behavior index and the association index may specifically include the following sub-steps:
  • Sub-step 2081 the behavior index and the correlation index are respectively sorted, and the behavior index ranking ratio and the correlation index ranking ratio are obtained;
  • Sub-step 2082 comparing the behavior index ranking ratio with a first preset ratio, and/or comparing the correlation index ranking ratio with a second preset ratio;
  • Sub-step 2083 determining, according to the comparison result, whether the user to be identified is a target user.
  • the obtained behavior index and the association index of all users may be respectively sorted, and the behavior index ranking ratio and the association index ranking ratio are obtained, for example, the behavior index and the association may be in descending order.
  • the index is sorted. If the total number of users is M, the behavior index of the user to be identified is ranked Nth among all users, and the association index ranks Kth among all users, then the ranking ratio of the behavior index is N/M*100. %, the correlation index sorting ratio is K/M*100%, and then the behavior index ranking ratio is compared with the first preset ratio, and/or the correlation index sorting ratio is compared with the second preset ratio. According to the comparison result, it is determined whether the user to be identified is a target user.
  • the behavior index ranking ratio may be first compared with the first preset ratio. If it is determined that the behavior index ranking ratio is within the second preset ratio range, then the correlation index ranking ratio and the second preset are further determined. a ratio, determining whether the relevance index ranking ratio is within a second preset ratio range, and if yes, determining that the to-be-identified user is a target user, or first determining that the correlation index ranking ratio is within a second preset ratio range Then, the behavior index ranking ratio is compared with the first preset ratio, and according to the comparison result, it is determined whether the user to be identified is the target user.
  • a person skilled in the art may determine the order of comparison according to actual needs, or may simultaneously compare the ranking ratio of the behavior index with the first preset ratio, and the ranking ratio of the associated index with the second preset ratio, which is not specifically described in this application. limited.
  • the embodiment of the present application may also identify the target user by using only the behavior index or the ranking ratio of the association index, which is not specifically limited in this application.
  • the first preset ratio and the second preset ratio may be the same or different, and examples may be used.
  • the first preset ratio and the second preset ratio may both be set to 10%, or the first preset ratio is set to 10%, and the second preset ratio is set to 8%.
  • a specific value of the first preset ratio and the second preset ratio may be determined by a person skilled in the art according to actual needs, which is not specifically limited in this application.
  • the embodiment of the present application determines the authority index by determining the behavior index of the user's purchase of the commodity (such as the favorable rate) and the time point of purchasing the commodity, and determines whether the user has authority by using the strong relationship and the weak relationship of the user to determine whether the user has the Influence, in which the weak relationship is similar to the commodity, it can specifically represent the potential influence of the user on others, further clarify the calculation method of the user behavior index and the association index, which helps to quickly identify the target user and helps The merchants conduct social marketing, operational selection and category recommendation based on the identified target users, which makes the marketing and recommendation of the merchant more targeted.
  • FIG. 5 a structural block diagram of an embodiment of an identification device of a target user of the present application is shown, which may specifically include the following modules:
  • the user information obtaining module 501 is configured to obtain user information of the user to be identified, where the user information includes behavior information and associated information.
  • a behavior index determining module 502 configured to determine, for the behavior information, a behavior index of the user to be identified;
  • the association index determining module 503 is configured to determine, according to the association information, an association index of the user to be identified;
  • the target user identification module 504 is configured to determine, according to the behavior index and the association index, whether the user to be identified is a target user.
  • the behavior information may include a user behavior number and a user behavior object
  • the behavior index determination module 502 may specifically include the following sub-modules:
  • the number index determination sub-module 5021 is configured to determine a frequency index of the user to be identified for the number of times of the user behavior
  • An object index determining sub-module 5022 configured to determine, for the user behavior object, a pair of the user to be identified Elephant index
  • the behavior index determination sub-module 5023 is configured to determine the behavior index by using the number of times index and the object index.
  • the number-of-times index determining sub-module 5021 may specifically include the following sub-modules:
  • the user behavior number determining sub-module 211 is configured to determine whether the number of user behaviors in the preset time period is greater than a first preset threshold
  • the first number index determining sub-module 212 is configured to determine, when the number of user behaviors in the preset time period is greater than the first preset threshold, that the number of times index is the first number of times index;
  • the second number index determining sub-module 213 is configured to determine that the number of times index is a second number of times index when the number of user actions in the preset time period is less than the first preset threshold.
  • the user behavior object may include a first attribute behavior object and a second attribute behavior object
  • the object index determination sub-module 5022 may specifically include the following sub-modules:
  • a first attribute behavior object index determining sub-module 221, configured to determine, for the first attribute behavior object, a first attribute behavior object index of the user to be identified
  • a second attribute behavior object index determining sub-module 222 configured to determine, for the second attribute behavior object, a second attribute behavior object index of the user to be identified
  • the user object index determining sub-module 223 is configured to determine an object index of the user to be identified by using the first attribute behavior object index and the second attribute behavior object index.
  • the object index determining sub-module 5022 may further include the following sub-modules:
  • the first attribute behavior object determining sub-module 224 is configured to determine whether the first attribute behavior object is included in the user behavior object;
  • the user behavior occurrence time determining sub-module 225 is configured to determine a time of occurrence of the user behavior for the first attribute behavior object when the first attribute behavior object is included in the user behavior object.
  • the first attribute behavior object index determining sub-module 221 may specifically include the following sub-modules:
  • the first attribute behavior object quantity obtaining sub-module 221A is configured to acquire the number of the first attribute behavior objects whose occurrence time is earlier than the preset time;
  • a first attribute behavior object ratio determining sub-module 221B configured to determine, according to the quantity of the first attribute behavior object, that the first attribute behavior object whose occurrence time is earlier than the preset time is in the user behavior object The proportion.
  • the second attribute behavior object index determining sub-module 222 may specifically include the following Submodule:
  • a second attribute behavior object quantity obtaining sub-module 222A configured to acquire the quantity of the second attribute behavior object
  • the second attribute behavior object ratio determining sub-module 222B is configured to determine, according to the quantity of the second attribute behavior object, a proportion of the second attribute behavior object in the user behavior object.
  • the behavior index determining sub-module 5023 may specifically include the following sub-modules:
  • the behavior index weighted summation sub-module 231 is configured to weight the sum of the number of times index, the first attribute behavior object index, and the second attribute behavior object index to obtain the behavior index.
  • the association information may include the number of associated users and associated object information
  • the association index determining module 503 may specifically include the following sub-modules:
  • a user association index determining sub-module 5031 configured to determine a user association index of the to-be-identified user for the number of the associated users
  • An object association index determining submodule 5032 configured to determine an object association index of the user to be identified for the associated object information
  • the association index determining sub-module 5033 is configured to determine the association index by using the user association index and the object association index.
  • the user association index determining submodule 5031 may specifically include the following submodules:
  • the associated user quantity determining sub-module 311 is configured to determine whether the number of the associated users is greater than a second preset threshold
  • the first user association index determining sub-module 312 is configured to determine, when the number of the associated users is greater than a second preset threshold, that the user association index is a first user association index;
  • the second user association index determining sub-module 313 is configured to determine that the user association index is a second user association index when the number of the associated users is less than a second preset threshold.
  • the object association index determining submodule 5032 may specifically include the following submodules:
  • the similarity calculation sub-module 321 is configured to calculate the similarity between the user to be identified and other users one by one for the associated object information
  • the to-be-identified user object association index determining sub-module 322 is configured to determine an object association index of the to-be-identified user according to the similarity.
  • the similarity between the user to be identified and other users may be calculated one by one by using the following formula:
  • the to-be-identified user object association index determining sub-module 322 may specifically include the following units:
  • the number of users locating unit 322A is configured to find the number of users whose similarity with the user to be identified is greater than a third preset threshold.
  • the object association index determining unit 322B is configured to determine an object association index of the user to be identified according to the number of users.
  • association index determining submodule 5033 may specifically include the following submodules:
  • the association index weighted summation sub-module 331 is configured to weight the user association index and the object association index to obtain an association index of the user to be identified.
  • the target user identification module 504 may specifically include the following sub-modules:
  • the index ordering sub-module 5041 is configured to separately sort the behavior index and the correlation index to obtain a behavior index ranking ratio and a correlation index ranking ratio;
  • a sorting ratio comparison sub-module 5042 configured to compare the behavior index ranking ratio with a first preset ratio, and/or the correlation index sorting ratio is compared with a second preset ratio;
  • the target user identification sub-module 5043 is configured to determine, according to the comparison result, whether the user to be identified is a target user.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
  • embodiments of the embodiments of the present application can be provided as a method, apparatus, or computer program product. Therefore, the embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, embodiments of the present application may be included in one or more of them.
  • a computer program product embodied on a computer usable storage medium including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the computer device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device.
  • computer readable media does not include non-persistent computer readable media, such as modulated data signals and carrier waves.
  • Embodiments of the present application are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal device to produce a machine such that instructions are executed by a processor of a computer or other programmable data processing terminal device
  • Means are provided for implementing the functions specified in one or more of the flow or in one or more blocks of the flow chart.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the instruction device implements the functions specified in one or more blocks of the flowchart or in a flow or block of the flowchart.

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (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)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请实施例提供了一种目标用户的识别方法和装置,所述方法包括:获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;针对所述行为信息,确定所述待识别用户的行为指数;以及,针对所述关联信息,确定所述待识别用户的关联指数;依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户,明确了用户行为指数和关联指数的计算方式,可以实现识别方法的标准化和定量化,使得对目标用户的识别更准确、更有效。

Description

一种目标用户的识别方法和装置
本申请要求2016年02月29日递交的申请号为201610112500.3、发明名称为“一种目标用户的识别方法和装置”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
本申请涉及互联网技术领域,特别是涉及一种目标用户的识别方法和一种目标用户的识别装置。
背景技术
互联网技术的进步推动了电子商务的发展,通过网络,用户能够快速地选购自己所需的商品,因此,商家对商品的营销也越来越多地从线下向线上转移。
在电商营销中,社会化营销是一种重要的营销手段。传统的社会化营销是通过电商平台或卖家与买家建立联系,然后以买家推荐新客户的方式来进行。这种营销方式有诸多缺点,其一是需要较多的人力运营,耗费较大;其二是效果较差,因为平台或者卖家并不清楚买家是否权威,是否有足够的影响力,从而使得营销中的投入产出比较低。
因此,如果能够及时地发现平台或某类目下具有足够的权威性和影响力的用户,则可以通过个性化、重点化运营该用户的方式,对其进行定向优质服务,从而达到影响其他消费者的目的,使得营销更有针对性,更具效果。
发明内容
鉴于上述问题,提出了本申请实施例以便提供一种克服上述问题或者至少部分地解决上述问题的一种目标用户的识别方法和相应的一种目标用户的识别装置。
为了解决上述问题,本申请公开了一种目标用户的识别方法,包括:
获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;
针对所述行为信息,确定所述待识别用户的行为指数;以及,
针对所述关联信息,确定所述待识别用户的关联指数;
依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
可选地,所述行为信息包括用户行为次数和用户行为对象,所述针对所述行为信息, 确定所述待识别用户的行为指数的步骤包括:
针对所述用户行为次数,确定所述待识别用户的次数指数;
针对所述用户行为对象,确定所述待识别用户的对象指数;
采用所述次数指数和对象指数,确定所述行为指数。
可选地,所述针对所述用户行为次数,确定所述待识别用户的次数指数的步骤包括:
判断在预设时间段内的用户行为次数是否大于第一预设阈值;
若是,则确定所述次数指数为第一次数指数;
若否,则确定所述次数指数为第二次数指数。
可选地,所述用户行为对象包括第一属性行为对象和第二属性行为对象,所述针对所述用户行为对象,确定所述待识别用户的对象指数的步骤包括:
针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
可选地,在所述针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数的步骤前,还包括:
判断所述用户行为对象中是否包括第一属性行为对象;
若是,则确定针对所述第一属性行为对象的用户行为的发生时间。
可选地,所述针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数的步骤包括:
获取所述发生时间早于预设时间的第一属性行为对象的数量;
根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属性行为对象在所述用户行为对象中所占的比例。
可选地,所述针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数的步骤包括:
获取所述第二属性行为对象的数量;
根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
可选地,所述采用所述次数指数和对象指数,确定所述行为指数的步骤包括:
将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数 加权求和,获得所述行为指数。
可选地,所述关联信息包括关联用户的数量和关联对象信息,所述针对所述关联信息,确定所述待识别用户的关联指数的步骤包括:
针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
针对所述关联对象信息,确定所述待识别用户的对象关联指数;
采用所述用户关联指数和对象关联指数,确定所述关联指数。
可选地,所述针对所述关联用户的数量,确定所述待识别用户的用户关联指数的步骤包括:
判断所述关联用户的数量是否大于第二预设阈值;
若是,则确定所述用户关联指数为第一用户关联指数;
若否,则确定所述用户关联指数为第二用户关联指数。
可选地,所述针对所述关联对象信息,确定所述待识别用户的对象关联指数的步骤包括:
针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
根据所述相似度,确定所述待识别用户的对象关联指数。
可选地,所述根据所述相似度,确定所述待识别用户的对象关联指数的步骤包括:
查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
根据所述用户数量,确定所述待识别用户的对象关联指数。
可选地,所述采用所述用户关联指数和对象关联指数,确定所述关联指数的步骤包括:
将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
可选地,所述依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户的步骤包括:
将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
将所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
根据比较结果,确定所述待识别用户是否为目标用户。
为了解决上述问题,本申请公开了一种目标用户的识别装置,包括:
用户信息获取模块,用于获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;
行为指数确定模块,用于针对所述行为信息,确定所述待识别用户的行为指数;以及,
关联指数确定模块,用于针对所述关联信息,确定所述待识别用户的关联指数;
目标用户识别模块,用于依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
可选地,所述行为信息包括用户行为次数和用户行为对象,所述行为指数确定模块包括:
次数指数确定子模块,用于针对所述用户行为次数,确定所述待识别用户的次数指数;
对象指数确定子模块,用于针对所述用户行为对象,确定所述待识别用户的对象指数;
行为指数确定子模块,用于采用所述次数指数和对象指数,确定所述行为指数。
可选地,所述次数指数确定子模块包括:
用户行为次数判断子模块,用于判断在预设时间段内的用户行为次数是否大于第一预设阈值;
第一次数指数确定子模块,用于在预设时间段内的用户行为次数大于第一预设阈值时,确定所述次数指数为第一次数指数;
第二次数指数确定子模块,用于在预设时间段内的用户行为次数小于第一预设阈值时,确定所述次数指数为第二次数指数。
可选地,所述用户行为对象包括第一属性行为对象和第二属性行为对象,所述对象指数确定子模块包括:
第一属性行为对象指数确定子模块,用于针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
第二属性行为对象指数确定子模块,用于针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
待识别用户对象指数确定子模块,用于采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
可选地,所述对象指数确定子模块还包括:
第一属性行为对象判断子模块,用于判断所述用户行为对象中是否包括第一属性行为对象;
用户行为发生时间确定子模块,用于在所述用户行为对象中包括有第一属性行为对象时,确定针对所述第一属性行为对象的用户行为的发生时间。
可选地,所述第一属性行为对象指数确定子模块包括:
第一属性行为对象数量获取子模块,用于获取所述发生时间早于预设时间的第一属性行为对象的数量;
第一属性行为对象比例确定子模块,用于根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属性行为对象在所述用户行为对象中所占的比例。
可选地,所述第二属性行为对象指数确定子模块包括:
第二属性行为对象数量获取子模块,用于获取所述第二属性行为对象的数量;
第二属性行为对象比例确定子模块,用于根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
可选地,所述行为指数确定子模块包括:
行为指数加权求和子模块,用于将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数加权求和,获得所述行为指数。
可选地,所述关联信息包括关联用户的数量和关联对象信息,所述关联指数确定模块包括:
用户关联指数确定子模块,用于针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
对象关联指数确定子模块,用于针对所述关联对象信息,确定所述待识别用户的对象关联指数;
关联指数确定子模块,用于采用所述用户关联指数和对象关联指数,确定所述关联指数。
可选地,所述用户关联指数确定子模块包括:
关联用户数量判断子模块,用于判断所述关联用户的数量是否大于第二预设阈值;
第一用户关联指数确定子模块,用于在所述关联用户的数量大于第二预设阈值时,确定所述用户关联指数为第一用户关联指数;
第二用户关联指数确定子模块,用于在所述关联用户的数量小于第二预设阈值时,确定所述用户关联指数为第二用户关联指数。
可选地,所述对象关联指数确定子模块包括:
相似度计算子模块,用于针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
待识别用户对象关联指数确定子模块,用于根据所述相似度,确定所述待识别用户的对象关联指数。
可选地,所述待识别用户对象关联指数确定子模块包括:
用户数量查找单元,用于查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
对象关联指数确定单元,用于根据所述用户数量,确定所述待识别用户的对象关联指数。
可选地,所述关联指数确定子模块包括:
关联指数加权求和子模块,用于将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
可选地,所述目标用户识别模块包括:
指数排序子模块,用于将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
排序比例比较子模块,用于对所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
目标用户识别子模块,用于根据比较结果,确定所述待识别用户是否为目标用户。
与背景技术相比,本申请实施例包括以下优点:
本申请实施例通过获取待识别用户的用户信息,然后根据所述用户信息确定出所述待识别用户的行为指数和关联指数,从而确定出所述待识别用户是否为目标用户,可以实现识别方法的标准化和定量化,使得对目标用户的识别更准确、更有效。
其次,本申请实施例通过对用户购买商品的相关情况(如好评率)及购买商品的时间点来确定行为指数,判断用户是否权威;通过用户的强关系和弱关系来确定关联指数,判断用户是否具有影响力,其中,弱关系采用商品相似作关联,可以具体表征用户对他人的潜在影响力,进一步明确了用户行为指数和关联指数的计算方式,有助于快速识别 出目标用户的,有助于商家基于识别出的目标用户进行社会化营销,运营选品及类目推荐等,使得商家的营销和推荐更具针对性。
附图说明
图1是本申请的一种目标用户的识别方法实施例一的步骤流程图;
图2是本申请的一种目标用户的识别方法实施例二的步骤流程图;
图3是本申请的一种确定用户的行为指数的原理图;
图4是本申请的某一商品销售量增长示意图;
图5是本申请的一种目标用户的识别装置实施例的结构框图。
具体实施方式
为使本申请的上述目的、特征和优点能够更加明显易懂,下面结合附图和具体实施方式对本申请作进一步详细的说明。
参照图1,示出了本申请的一种目标用户的识别方法实施例一的步骤流程图,具体可以包括如下步骤:
步骤101,获取待识别用户的用户信息;
在本申请实施例中,所述用户信息可以包括行为信息和关联信息。
行为信息可以是记录的所述用户的某一行为,例如用户浏览某商品的介绍信息,在收藏夹中收藏该商品,或者购买该商品。
关联信息可以是与所述用户直接关联的其他用户,例如,用户的好友信息;也可以是与所述用户间接关联的其他用户,例如,都曾浏览或购买过某一商品的其他用户的信息。
步骤102,针对所述行为信息,确定所述待识别用户的行为指数;
步骤103,针对所述关联信息,确定所述待识别用户的关联指数;
在本申请实施例中,在获取到用户的行为信息和关联信息后,可以分别针对所述行为信息和所述关联信息,确定出待识别用户的行为指数和关联指数。
步骤104,依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
在具体实现中,可以将行为指数与关联指数直接相加,将其结果与预设的判断条件 进行比较,从而识别出所述用户是否是目标用户,还可以对行为指数和关联指数分别赋予不同的权重值,然后对加权求和的结果进行判断,来识别目标用户,也可以分别将行为指数和关联指数与预设的判断条件比较,得到待识别用户是否为目标用户的判断,本领域技术人员可以根据实际需要选择以何种方式来识别目标用户,本申请对此不作具体限定。
在本申请实施例中,通过获取待识别用户的用户信息,然后根据所述用户信息确定出所述待识别用户的行为指数和关联指数,从而确定出所述待识别用户是否为目标用户,可以实现识别方法的标准化和定量化,使得对目标用户的识别更准确、更有效。
参照图2,示出了本申请的一种目标用户的识别方法实施例二的步骤流程图,具体可以包括如下步骤:
步骤201,获取待识别用户的用户信息;
在本申请实施例中,所述用户信息可以包括用户在某一电商平台的行为信息和关联信息。
行为信息可以是记录的所述用户的某一行为,例如用户浏览某商品的介绍信息,在收藏夹中收藏该商品,或者购买该商品。
关联信息可以是与所述用户直接关联的其他用户,例如,用户的好友信息;也可以是与所述用户间接关联的其他用户,例如,都曾浏览或购买过某一商品的其他用户的信息。
如图3所示,是本申请的一种确定用户的行为指数的原理图,在具体实现中,用户信息可以通过所述电商平台的数据仓库获得。数据仓库一般用作数据读写,可以存储用户交易信息表,商品交易信息表,商品DSR(Detail Seller Rating,卖家服务评级系统)信息表等数据。
步骤202,针对所述用户行为次数,确定所述待识别用户的次数指数;
在具体实现中,所述用户行为次数可以是用户交易量数据。
在本申请的一种优选实施例中,所述针对所述用户行为次数,确定所述待识别用户的次数指数的步骤具体可以包括如下子步骤:
子步骤2021,判断在预设时间段内的用户行为次数是否大于第一预设阈值;
子步骤2022,若是,则确定所述次数指数为第一次数指数;
子步骤2023,若否,则确定所述次数指数为第二次数指数。
在本申请实施例中,根据用户行为次数,确定出待识别用户的次数指数可以通过将在预设时间段内的用户行为次数与预设阈值进行比较得到。具体地,预设时间段可以是30天,90天或者180天等等,本领域技术人员可以根据实际需要确定预设时间段的长短,本申请对此不作具体限定。
例如,可以提取所述待识别用户在90天内的交易量数据,然后与第一预设阈值进行比较,若所述交易量大于第一预设阈值,则可以确定次数指数为第一次数指数,若所述交易量小于或等于第一预设阈值,则可以确定次数指数为第二次数指数。第一预设阈值可以根据预设时间段的长短来具体设置,例如,若预设时间段设置为30天,则可相应地将第一预设阈值设置为10次,即在30天内的交易量为10次,若预设时间段设置为90天,则可相应地将第一预设阈值设置为45次,即在90天内的交易量达到45次。
一般地,第一次数指数应该大于第二次数指数,例如,若第一次数指数设置为0.8,则可以相应地设置第二次数指数为0.5。第一次数指数和第二次数指数的大小也可以由本领域技术人员根据实际需要具体决定,本申请对此不作具体限定。
步骤203,针对所述用户行为对象,确定所述待识别用户的对象指数;
在本申请实施例中,所述用户行为对象可以包括第一属性行为对象和第二属性行为对象。例如,在用户行为对象为商品时,所述第一属性行为对象可以是销售量较高的商品(爆品),所述第二属性行为对象则可以是好评率较高的商品(优品)。
在本申请的一种优选实施例中,所述针对所述用户行为对象,确定所述待识别用户的对象指数的步骤具体可以包括如下子步骤:
子步骤2031,针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
在本申请实施例中,在根据所述第一属性行为对象,确定出所述第一属性行为对象指数前,可以首先判断所述用户行为对象中是否包括第一属性行为对象,即所述用户购买的商品中是否包括销售量较高的商品,若是,则可以进一步确定针对所述第一属性行为对象的用户行为的发生时间。用户行为的发生时间可以是用户购买此类销售量较高的商品的具体时间。
因此,所述针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数的子步骤可以进一步包括:
获取所述发生时间早于预设时间的第一属性行为对象的数量;
根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属 性行为对象在所述用户行为对象中所占的比例。
在本申请实施例中,所述预设时间可以是商品销售量增长率达到某一数值时的时间。如图4所示,是本申请的某一商品销售量增长示意图,其中横轴为商品上线销售的时间(单位:天),纵轴为销量。从图中可以看出,其中第1—3天销售量增长较慢,第3天后,由于各种原因(例如商品质量较好,前期买家自发宣传、促销、回头客)等,商品销售量增速明显变快,并在第6—7天增速达到最大值,在第7天后,由于某些原因(例如市场仿品开始出现,竞争对手打击等),销售量增速开始下跌。
若设最大增速为Smax,本申请实施例将增速为αSmax所处的时间点设为临界点T,其中α<1,当用户的购买时间t<=T时,则可以认为该用户是在商品成为爆品之前就发生了购买行为。例如对于图4,假设在第7天时销售量达到最大增速,此时Smax=120%,取α=1/2,则αSmax=60%,假定增速为60%的时间点为第4天,即预设时间为第4天(临界点T=4)。对于不同的商品类目,其预设时间可以是不同的,本领域技术人员也可以根据实际需要确定临界点的具体数值,本申请对此不作具体限定。
然后,可以获取到购买时间早于预设时间的商品数量,然后确定出所述商品在全部商品中所占的比例。
例如,若某一用户一共购买了10件商品,其中有3件为销售量较高的商品(爆品),而在这3件爆品中有1件是在它成为爆品前购买的(即购买时间早有预设时间/临界点),则销售量较高的商品在全部商品中所占的比例为1/10=0.1。
子步骤2032,针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
在本申请实施例中,所述第二属性行为对象可以是好评率较高的商品(优品),例如好评率超过一定数值r(r<=100%)的商品,本领域技术人员可以根据实际需要确定r的具体大小,本申请对此不作具体限定。
在本申请的一种优选实施例中,所述针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数的子步骤可以进一步包括:
获取所述第二属性行为对象的数量;
根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
在具体实现中,若取r=99%,则好评率大于等于99%的商品可以认为是优质商品, 可以从用户信息中获得好评率大于等于99%的商品数量,然后确定出所述商品在全部商品中所占的比例。
例如,若某一用户一共购买了10件商品,其中9件商品的好评率超过99%,则可以确定好评率较高的商品在全部商品中所占的比例为9/10=0.9。
子步骤2033,采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
在具体实现中,可以将所述第一属性行为对象指数和所述第二属性行为对象指数直接相加,得到所述待识别用户的对象指数,还可以对所述第一属性行为对象指数和所述第二属性行为对象指数分别赋予不同的权重值,然后将加权求和的结果作为所述待识别用户的对象指数。本领域技术人员可以根据实际需要选择以何种方式来确定待识别用户的对象指数,本申请对此不作具体限定。
步骤204,采用所述次数指数和对象指数,确定所述行为指数;
在本申请的一种优选实施例中,所述采用所述次数指数和对象指数,确定所述行为指数的步骤具体可以包括如下子步骤:
子步骤2041,将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数加权求和,获得所述行为指数。
在具体实现中,可以采用如下公式计算所述待识别用户的行为指数:
score=w1a+w2b+w3c
其中,a为所述待识别用户的次数指数,b为第一属性行为对象指数,c为第二属性行为对象指数,w1,w2,w3分别为a、b、c的权重值,且w1+w2+w3=100,本领域技术人员可以根据业务具体目标的需要,对上述权重值进行调整,本申请对此不作具体限定。
例如,若所述待识别用户的次数指数a=0.5,第一属性行为对象指数b=0.1,第二属性行为对象指数c=0.9,且w1=30,w2=50,w3=20,可以得到所述待识别用户的行为指数score=30*0.5+50*0.1+20*0.9=38。
步骤205,针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
在本申请实施例中,所述关联用户的数量可以是用户在所述电商平台的好友数量,好友关系体现了用户间的强关系。
在本申请的一种优选实施例中,所述针对所述关联用户的数量,确定所述待识别用 户的用户关联指数的步骤具体可以包括如下子步骤:
子步骤2051,判断所述关联用户的数量是否大于第二预设阈值;
子步骤2052,若是,则确定所述用户关联指数为第一用户关联指数;
子步骤2053,若否,则确定所述用户关联指数为第二用户关联指数。
在具体实现中,可以将待识别用户的好友数量与预设阈值进行比较,从而确定出用户关联指数。具体地,预设阈值可以是100人,150人或者180人等等,本领域技术人员可以根据实际需要确定预设阈值的大小,本申请对此不作具体限定。
例如,可以设定第二预设阈值为150人,然后将待识别用户的好友数量与第二预设阈值进行比较,若所述好友数量大于第二预设阈值,则可以确定用户关联指数为第一用户关联指数,若所述好友数量小于或等于第二预设阈值,则可以确定用户关联指数为第二用户关联指数。
一般地,第一次数指数应该大于第二次数指数,例如,用户A有100个买家好友,而用户B有200个买家好友,在第二预设阈值为150人时,可以将第一用户关联指数设置为0.7,相应地设置第一用户关联指数为0.4,即用户A的用户关联指数为0.4,用户B的用户关联指数为0.8,第一用户关联指数和第二用户关联指数的大小也可以由本领域技术人员根据实际需要具体决定,本申请对此不作具体限定。
步骤206,针对所述关联对象信息,确定所述待识别用户的对象关联指数;
在本申请实施例中,所述关联对象信息可以是根据用户行为对象确定的信息,例如同时购买了某一商品。
在本申请的一种优选实施例中,所述针对所述关联对象信息,确定所述待识别用户的对象关联指数的步骤具体可以包括如下子步骤:
子步骤2061,针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
在本申请实施例中,可以根据用户购买的商品来计算用户间的相似度,当相似度达到一定程度时,可以认为这两个用户间存在弱关系。
在本申请的一种优选实施例中,可以采用如下公式,逐个计算所述待识别用户与其他用户间的相似度:
Figure PCTCN2017073866-appb-000001
其中,|M|是待识别用户的用户行为对象的数量,|N|是其他用户的用户行为对象的数 量;|MN|是待识别用户和其他用户拥有的相同的用户行为对象的数量。
例如,若用户1购买了{a,b,c}三件商品,用户2购买了{a,c,d}三件商品,则用户1与用户2共同购买的商品有两件:a和c,则根据上述公式,有:
Figure PCTCN2017073866-appb-000002
在本申请实施例中,可以扫描所述电商平台的每一个用户,分别计算出待识别用户与其他每一个用户间的相似度。
子步骤2062,根据所述相似度,确定所述待识别用户的对象关联指数。
在本申请的一种优选实施例中所述根据所述相似度,确定所述待识别用户的对象关联指数的子步骤可以进一步包括:
查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
根据所述用户数量,确定所述待识别用户的对象关联指数。
在具体实现中,可以认为当相似度达到第三预设阈值β时(例如β=0.8),用户间建立有弱关系,然后查找出与所述待识别用户的相似度大于0.8的用户数量,然后根据所述用户数量,确定出待识别用户的对象关联指数。本领域技术人员可以根据业务需要确定第三预设阈值β的具体大小,本申请对此不作具体限定。
步骤207,采用所述用户关联指数和对象关联指数,确定所述关联指数;
在本申请的一种优选实施例中,所述采用所述用户关联指数和对象关联指数,确定所述关联指数的步骤具体可以包括如下子步骤:
子步骤2071,将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
在具体实现中,可以采用如下公式计算所述待识别用户的关联指数:
score2=w1P+w2Q
其中,P为所述待识别用户的用户关联指数,Q为对象关联指数,w1,w2分别为P、Q的权重值,且w1+w2=100,本领域技术人员可以根据业务具体目标的需要,对上述权重值进行调整,本申请对此不作具体限定。
例如,若所述待识别用户的用户关联指数P=0.4,Q=0.8,且w1=60,w2=40,可以得到所述待识别用户的关联指数score2=60*0.4+40*0.8=56。
步骤208,依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用 户。
当分别获得所述待识别用户的行为指数和关联指数后,可以依据所述行为指数和关联指数,判断所述待识别用户是否为目标用户。
例如,可以将所述待识别用户的行为指数与关联指数相加,根据获得的结果对待识别用户进行识别,还可以分别根据行为指数或关联指数对所述用户进行识别。
在本申请的一种优选实施例中,所述依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户的步骤具体可以包括如下子步骤:
子步骤2081,将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
子步骤2082,将所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
子步骤2083,根据比较结果,确定所述待识别用户是否为目标用户。
在本申请实施例中,可以将获得的全部用户的行为指数和关联指数分别排序,获得行为指数排序比例和关联指数排序比例,例如,可以按照从大到小的顺序对所述行为指数和关联指数进行排序,若总用户数为M,待识别用户的行为指数在全部用户中排第N位,关联指数在全部用户中排第K位,则所述行为指数排序比例为N/M*100%,所述关联指数排序比例为K/M*100%,然后将所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较,根据比较结果,确定所述待识别用户是否为目标用户。
在具体实现中,可以首先将行为指数排序比例与第一预设比例进行比较,若确定所述行为指数排序比例在第二预设比例范围内,然后再将关联指数排序比例与第二预设比例,判断所述关联指数排序比例是否在第二预设比例范围内,若是,则确定所述待识别用户为目标用户,也可以首先在确定关联指数排序比例在第二预设比例范围内时,再将行为指数排序比例与第一预设比例进行比较,根据比较结果确定待识别用户是否为目标用户。本领域技术人员根据实际需要可以自行确定比较的先后顺序,也可以同时对行为指数排序比例与第一预设比例,和关联指数排序比例与第二预设比例进行比较,本申请对此不作具体限定。
此外,在某些特定情形下,本申请实施例还可以仅仅依靠行为指数或者关联指数的排序比例,来对目标用户进行识别,本申请对此不作具体限定。
在本申请实施例中,所述第一预设比例与第二预设比例可以相同,也可以不同,例 如,第一预设比例与第二预设比例均可以设置为10%,或者,第一预设比例设置为10%,第二预设比例设置为8%。本领域技术人员可以根据实际需要确定第一预设比例与第二预设比例的具体数值,本申请对此不作具体限定。
本申请实施例通过对用户购买商品的相关情况(如好评率)及购买商品的时间点来确定行为指数,判断用户是否权威;通过用户的强关系和弱关系来确定关联指数,判断用户是否具有影响力,其中,弱关系采用商品相似作关联,可以具体表征用户对他人的潜在影响力,进一步明确了用户行为指数和关联指数的计算方式,有助于快速识别出目标用户的,有助于商家基于识别出的目标用户进行社会化营销,运营选品及类目推荐等,使得商家的营销和推荐更具针对性。
需要说明的是,对于方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本申请实施例并不受所描述的动作顺序的限制,因为依据本申请实施例,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作并不一定是本申请实施例所必须的。
参照图5,示出了本申请的一种目标用户的识别装置实施例的结构框图,具体可以包括如下模块:
用户信息获取模块501,用于获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;
行为指数确定模块502,用于针对所述行为信息,确定所述待识别用户的行为指数;以及,
关联指数确定模块503,用于针对所述关联信息,确定所述待识别用户的关联指数;
目标用户识别模块504,用于依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
在本申请实施例中,所述行为信息可以包括用户行为次数和用户行为对象,所述行为指数确定模块502具体可以包括如下子模块:
次数指数确定子模块5021,用于针对所述用户行为次数,确定所述待识别用户的次数指数;
对象指数确定子模块5022,用于针对所述用户行为对象,确定所述待识别用户的对 象指数;
行为指数确定子模块5023,用于采用所述次数指数和对象指数,确定所述行为指数。
在本申请实施例中,所述次数指数确定子模块5021具体可以包括如下子模块:
用户行为次数判断子模块211,用于判断在预设时间段内的用户行为次数是否大于第一预设阈值;
第一次数指数确定子模块212,用于在预设时间段内的用户行为次数大于第一预设阈值时,确定所述次数指数为第一次数指数;
第二次数指数确定子模块213,用于在预设时间段内的用户行为次数小于第一预设阈值时,确定所述次数指数为第二次数指数。
在本申请实施例中,所述用户行为对象可以包括第一属性行为对象和第二属性行为对象,所述对象指数确定子模块5022具体可以包括如下子模块:
第一属性行为对象指数确定子模块221,用于针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
第二属性行为对象指数确定子模块222,用于针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
待识别用户对象指数确定子模块223,用于采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
在本申请实施例中,所述对象指数确定子模块5022还可以包括如下子模块:
第一属性行为对象判断子模块224,用于判断所述用户行为对象中是否包括第一属性行为对象;
用户行为发生时间确定子模块225,用于在所述用户行为对象中包括有第一属性行为对象时,确定针对所述第一属性行为对象的用户行为的发生时间。
在本申请实施例中,所述第一属性行为对象指数确定子模块221具体可以包括如下子模块:
第一属性行为对象数量获取子模块221A,用于获取所述发生时间早于预设时间的第一属性行为对象的数量;
第一属性行为对象比例确定子模块221B,用于根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属性行为对象在所述用户行为对象中所占的比例。
在本申请实施例中,所述第二属性行为对象指数确定子模块222具体可以包括如下 子模块:
第二属性行为对象数量获取子模块222A,用于获取所述第二属性行为对象的数量;
第二属性行为对象比例确定子模块222B,用于根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
在本申请实施例中,所述行为指数确定子模块5023具体可以包括如下子模块:
行为指数加权求和子模块231,用于将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数加权求和,获得所述行为指数。
在本申请实施例中,所述关联信息可以包括关联用户的数量和关联对象信息,所述关联指数确定模块503具体可以包括如下子模块:
用户关联指数确定子模块5031,用于针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
对象关联指数确定子模块5032,用于针对所述关联对象信息,确定所述待识别用户的对象关联指数;
关联指数确定子模块5033,用于采用所述用户关联指数和对象关联指数,确定所述关联指数。
在本申请实施例中,所述用户关联指数确定子模块5031具体可以包括如下子模块:
关联用户数量判断子模块311,用于判断所述关联用户的数量是否大于第二预设阈值;
第一用户关联指数确定子模块312,用于在所述关联用户的数量大于第二预设阈值时,确定所述用户关联指数为第一用户关联指数;
第二用户关联指数确定子模块313,用于在所述关联用户的数量小于第二预设阈值时,确定所述用户关联指数为第二用户关联指数。
在本申请实施例中,所述对象关联指数确定子模块5032具体可以包括如下子模块:
相似度计算子模块321,用于针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
待识别用户对象关联指数确定子模块322,用于根据所述相似度,确定所述待识别用户的对象关联指数。
在本申请实施例中,可以采用如下公式,逐个计算所述待识别用户与其他用户间的相似度:
Figure PCTCN2017073866-appb-000003
其中,|M|是待识别用户的用户行为对象的数量,|N|是其他用户的用户行为对象的数量;|MN|是待识别用户和其他用户拥有的相同的用户行为对象的数量。
在本申请实施例中,所述待识别用户对象关联指数确定子模块322具体可以包括如下单元:
用户数量查找单元322A,用于查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
对象关联指数确定单元322B,用于根据所述用户数量,确定所述待识别用户的对象关联指数。
在本申请实施例中,所述关联指数确定子模块5033具体可以包括如下子模块:
关联指数加权求和子模块331,用于将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
在本申请实施例中,所述目标用户识别模块504具体可以包括如下子模块:
指数排序子模块5041,用于将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
排序比例比较子模块5042,用于对所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
目标用户识别子模块5043,用于根据比较结果,确定所述待识别用户是否为目标用户。
对于装置实施例而言,由于其与方法实施例基本相似,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。
本领域内的技术人员应明白,本申请实施例的实施例可提供为方法、装置、或计算机程序产品。因此,本申请实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请实施例可采用在一个或多个其中包含有计 算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
在一个典型的配置中,所述计算机设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。内存可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括非持续性的电脑可读媒体(transitory media),如调制的数据信号和载波。
本申请实施例是参照根据本申请实施例的方法、终端设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理终端设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理终端设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理终端设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理终端设备上,使得在计算机或其他可编程终端设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程终端设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本申请实施例的优选实施例,但本领域内的技术人员一旦得知了基本 创造性概念,则可对这些实施例做出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本申请实施例范围的所有变更和修改。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的相同要素。
以上对本申请所提供的一种目标用户的识别方法和一种目标用户的识别装置,进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的方法及其核心思想;同时,对于本领域的一般技术人员,依据本申请的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本申请的限制。

Claims (28)

  1. 一种目标用户的识别方法,其特征在于,包括:
    获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;
    针对所述行为信息,确定所述待识别用户的行为指数;以及,
    针对所述关联信息,确定所述待识别用户的关联指数;
    依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
  2. 根据权利要求1所述的方法,其特征在于,所述行为信息包括用户行为次数和用户行为对象,所述针对所述行为信息,确定所述待识别用户的行为指数的步骤包括:
    针对所述用户行为次数,确定所述待识别用户的次数指数;
    针对所述用户行为对象,确定所述待识别用户的对象指数;
    采用所述次数指数和对象指数,确定所述行为指数。
  3. 根据权利要求2所述的方法,其特征在于,所述针对所述用户行为次数,确定所述待识别用户的次数指数的步骤包括:
    判断在预设时间段内的用户行为次数是否大于第一预设阈值;
    若是,则确定所述次数指数为第一次数指数;
    若否,则确定所述次数指数为第二次数指数。
  4. 根据权利要求2或3所述的方法,其特征在于,所述用户行为对象包括第一属性行为对象和第二属性行为对象,所述针对所述用户行为对象,确定所述待识别用户的对象指数的步骤包括:
    针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
    针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
    采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
  5. 根据权利要求4所述的方法,其特征在于,在所述针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数的步骤前,还包括:
    判断所述用户行为对象中是否包括第一属性行为对象;
    若是,则确定针对所述第一属性行为对象的用户行为的发生时间。
  6. 根据权利要求5所述的方法,其特征在于,所述针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数的步骤包括:
    获取所述发生时间早于预设时间的第一属性行为对象的数量;
    根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属性行为对象在所述用户行为对象中所占的比例。
  7. 根据权利要求5或6所述的方法,其特征在于,所述针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数的步骤包括:
    获取所述第二属性行为对象的数量;
    根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
  8. 根据权利要求7所述的方法,其特征在于,所述采用所述次数指数和对象指数,确定所述行为指数的步骤包括:
    将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数加权求和,获得所述行为指数。
  9. 根据权利要求1或2或3或5或6或8所述的方法,其特征在于,所述关联信息包括关联用户的数量和关联对象信息,所述针对所述关联信息,确定所述待识别用户的关联指数的步骤包括:
    针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
    针对所述关联对象信息,确定所述待识别用户的对象关联指数;
    采用所述用户关联指数和对象关联指数,确定所述关联指数。
  10. 根据权利要求9所述的方法,其特征在于,所述针对所述关联用户的数量,确定所述待识别用户的用户关联指数的步骤包括:
    判断所述关联用户的数量是否大于第二预设阈值;
    若是,则确定所述用户关联指数为第一用户关联指数;
    若否,则确定所述用户关联指数为第二用户关联指数。
  11. 根据权利要求10所述的方法,其特征在于,所述针对所述关联对象信息,确定所述待识别用户的对象关联指数的步骤包括:
    针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
    根据所述相似度,确定所述待识别用户的对象关联指数。
  12. 根据权利要求11所述的方法,其特征在于,所述根据所述相似度,确定所述待识别用户的对象关联指数的步骤包括:
    查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
    根据所述用户数量,确定所述待识别用户的对象关联指数。
  13. 根据权利要求12所述的方法,其特征在于,所述采用所述用户关联指数和对象关联指数,确定所述关联指数的步骤包括:
    将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
  14. 根据权利要求10或11或12或13所述的方法,其特征在于,所述依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户的步骤包括:
    将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
    将所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
    根据比较结果,确定所述待识别用户是否为目标用户。
  15. 一种目标用户的识别装置,其特征在于,包括:
    用户信息获取模块,用于获取待识别用户的用户信息,所述用户信息包括行为信息和关联信息;
    行为指数确定模块,用于针对所述行为信息,确定所述待识别用户的行为指数;以及,
    关联指数确定模块,用于针对所述关联信息,确定所述待识别用户的关联指数;
    目标用户识别模块,用于依据所述行为指数和所述关联指数,确定所述待识别用户是否为目标用户。
  16. 根据权利要求15所述的装置,其特征在于,所述行为信息包括用户行为次数和用户行为对象,所述行为指数确定模块包括:
    次数指数确定子模块,用于针对所述用户行为次数,确定所述待识别用户的次数指数;
    对象指数确定子模块,用于针对所述用户行为对象,确定所述待识别用户的对象指数;
    行为指数确定子模块,用于采用所述次数指数和对象指数,确定所述行为指数。
  17. 根据权利要求16所述的装置,其特征在于,所述次数指数确定子模块包括:
    用户行为次数判断子模块,用于判断在预设时间段内的用户行为次数是否大于第一预设阈值;
    第一次数指数确定子模块,用于在预设时间段内的用户行为次数大于第一预设阈值时,确定所述次数指数为第一次数指数;
    第二次数指数确定子模块,用于在预设时间段内的用户行为次数小于第一预设阈值时,确定所述次数指数为第二次数指数。
  18. 根据权利要求16或17所述的装置,其特征在于,所述用户行为对象包括第一属性行为对象和第二属性行为对象,所述对象指数确定子模块包括:
    第一属性行为对象指数确定子模块,用于针对所述第一属性行为对象,确定所述待识别用户的第一属性行为对象指数;
    第二属性行为对象指数确定子模块,用于针对所述第二属性行为对象,确定所述待识别用户的第二属性行为对象指数;
    待识别用户对象指数确定子模块,用于采用所述第一属性行为对象指数和所述第二属性行为对象指数,确定所述待识别用户的对象指数。
  19. 根据权利要求18所述的装置,其特征在于,所述对象指数确定子模块还包括:
    第一属性行为对象判断子模块,用于判断所述用户行为对象中是否包括第一属性行为对象;
    用户行为发生时间确定子模块,用于在所述用户行为对象中包括有第一属性行为对象时,确定针对所述第一属性行为对象的用户行为的发生时间。
  20. 根据权利要求19所述的装置,其特征在于,所述第一属性行为对象指数确定子模块包括:
    第一属性行为对象数量获取子模块,用于获取所述发生时间早于预设时间的第一属性行为对象的数量;
    第一属性行为对象比例确定子模块,用于根据所述第一属性行为对象的数量,确定所述发生时间早于所述预设时间的第一属性行为对象在所述用户行为对象中所占的比例。
  21. 根据权利要求19或20所述的装置,其特征在于,所述第二属性行为对象指数确定子模块包括:
    第二属性行为对象数量获取子模块,用于获取所述第二属性行为对象的数量;
    第二属性行为对象比例确定子模块,用于根据所述第二属性行为对象的数量,确定所述第二属性行为对象在所述用户行为对象中所占的比例。
  22. 根据权利要求21所述的装置,其特征在于,所述行为指数确定子模块包括:
    行为指数加权求和子模块,用于将所述次数指数、所述第一属性行为对象指数,以及,所述第二属性行为对象指数加权求和,获得所述行为指数。
  23. 根据权利要求15或16或17或19或20或22所述的装置,其特征在于,所述关联信息包括关联用户的数量和关联对象信息,所述关联指数确定模块包括:
    用户关联指数确定子模块,用于针对所述关联用户的数量,确定所述待识别用户的用户关联指数;
    对象关联指数确定子模块,用于针对所述关联对象信息,确定所述待识别用户的对象关联指数;
    关联指数确定子模块,用于采用所述用户关联指数和对象关联指数,确定所述关联指数。
  24. 根据权利要求23所述的装置,其特征在于,所述用户关联指数确定子模块包括:
    关联用户数量判断子模块,用于判断所述关联用户的数量是否大于第二预设阈值;
    第一用户关联指数确定子模块,用于在所述关联用户的数量大于第二预设阈值时,确定所述用户关联指数为第一用户关联指数;
    第二用户关联指数确定子模块,用于在所述关联用户的数量小于第二预设阈值时,确定所述用户关联指数为第二用户关联指数。
  25. 根据权利要求24所述的装置,其特征在于,所述对象关联指数确定子模块包括:
    相似度计算子模块,用于针对所述关联对象信息,逐个计算所述待识别用户与其他用户间的相似度;
    待识别用户对象关联指数确定子模块,用于根据所述相似度,确定所述待识别用户的对象关联指数。
  26. 根据权利要求25所述的装置,其特征在于,所述待识别用户对象关联指数确定子模块包括:
    用户数量查找单元,用于查找出与所述待识别用户的相似度大于第三预设阈值的用户数量;
    对象关联指数确定单元,用于根据所述用户数量,确定所述待识别用户的对象关联指数。
  27. 根据权利要求26所述的装置,其特征在于,所述关联指数确定子模块包括:
    关联指数加权求和子模块,用于将所述用户关联指数与所述对象关联指数加权求和,获得所述待识别用户的关联指数。
  28. 根据权利要求24或25或26或27所述的装置,其特征在于,所述目标用户识别模块包括:
    指数排序子模块,用于将所述行为指数和所述关联指数分别排序,获得行为指数排序比例和关联指数排序比例;
    排序比例比较子模块,用于对所述行为指数排序比例与第一预设比例,和/或,所述关联指数排序比例与第二预设比例进行比较;
    目标用户识别子模块,用于根据比较结果,确定所述待识别用户是否为目标用户。
PCT/CN2017/073866 2016-02-29 2017-02-17 一种目标用户的识别方法和装置 WO2017148272A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610112500.3 2016-02-29
CN201610112500.3A CN107133811A (zh) 2016-02-29 2016-02-29 一种目标用户的识别方法和装置

Publications (1)

Publication Number Publication Date
WO2017148272A1 true WO2017148272A1 (zh) 2017-09-08

Family

ID=59721012

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/073866 WO2017148272A1 (zh) 2016-02-29 2017-02-17 一种目标用户的识别方法和装置

Country Status (3)

Country Link
CN (1) CN107133811A (zh)
TW (1) TW201734909A (zh)
WO (1) WO2017148272A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111367955A (zh) * 2019-10-09 2020-07-03 杭州海康威视系统技术有限公司 目标对象的识别方法、装置、电子设备和存储介质
CN112311719A (zh) * 2019-07-24 2021-02-02 阿里巴巴集团控股有限公司 数据处理方法、设备、系统及存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108665312B (zh) * 2018-05-08 2020-09-29 京东数字科技控股有限公司 用于生成信息的方法和装置
CN113468453B (zh) * 2020-03-30 2022-09-09 武汉斗鱼网络科技有限公司 一种目标用户的识别方法、装置、电子设备及存储介质
CN112836899A (zh) * 2020-09-24 2021-05-25 王海宏 基于区块链和云计算的属性预测方法、系统及服务中心

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100088327A1 (en) * 2008-10-02 2010-04-08 Nokia Corporation Method, Apparatus, and Computer Program Product for Identifying Media Item Similarities
CN104239335A (zh) * 2013-06-19 2014-12-24 阿里巴巴集团控股有限公司 特定用户信息获取方法及装置
CN104239338A (zh) * 2013-06-19 2014-12-24 阿里巴巴集团控股有限公司 信息推荐方法及装置

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103514215A (zh) * 2012-06-28 2014-01-15 北京奇虎科技有限公司 生成用户社交影响力信息的方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100088327A1 (en) * 2008-10-02 2010-04-08 Nokia Corporation Method, Apparatus, and Computer Program Product for Identifying Media Item Similarities
CN104239335A (zh) * 2013-06-19 2014-12-24 阿里巴巴集团控股有限公司 特定用户信息获取方法及装置
CN104239338A (zh) * 2013-06-19 2014-12-24 阿里巴巴集团控股有限公司 信息推荐方法及装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112311719A (zh) * 2019-07-24 2021-02-02 阿里巴巴集团控股有限公司 数据处理方法、设备、系统及存储介质
CN111367955A (zh) * 2019-10-09 2020-07-03 杭州海康威视系统技术有限公司 目标对象的识别方法、装置、电子设备和存储介质
CN111367955B (zh) * 2019-10-09 2024-03-26 杭州海康威视系统技术有限公司 目标对象的识别方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN107133811A (zh) 2017-09-05
TW201734909A (zh) 2017-10-01

Similar Documents

Publication Publication Date Title
TWI787196B (zh) 業務對象屬性標識的生成方法、裝置和系統
WO2017148272A1 (zh) 一种目标用户的识别方法和装置
WO2020215751A1 (zh) 一种推荐方法、电子设备及可读存储介质
CN110135915B (zh) 商品推荐方法和系统
JP6134444B2 (ja) 情報を推薦するための方法およびシステム
JP5828958B2 (ja) ソーシャル商品推薦の提供
TWI512653B (zh) Information providing method and apparatus, method and apparatus for determining the degree of comprehensive relevance
US9208437B2 (en) Personalized information pushing method and device
CN106934648B (zh) 一种数据处理方法和装置
CN106557480B (zh) 查询改写的实现方法及装置
WO2018053966A1 (zh) 点击率预估
US20130275269A1 (en) Searching supplier information based on transaction platform
TW201520790A (zh) 個性化資料搜尋方法和裝置
CN107633416B (zh) 一种业务对象的推荐方法、装置和系统
CN105630836A (zh) 搜索结果的排序方法和装置
WO2016107455A1 (zh) 一种信息匹配处理方法和装置
CN113837842A (zh) 一种基于用户行为数据的商品推荐方法及设备
CN104346428A (zh) 信息处理装置、信息处理方法及程序
US20190065611A1 (en) Search method and apparatus
Prasetyo Searching cheapest product on three different e-commerce using k-means algorithm
US20230101928A1 (en) User attribute preference model
CN111951051B (zh) 一种为客户推荐产品的方法、装置和系统
CN110827101A (zh) 一种店铺推荐的方法和装置
CN111339434B (zh) 信息推荐方法、装置、电子设备及计算机存储介质
WO2017219317A1 (zh) 一种基于搜索内容的信息推送方法及装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 17759123

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17759123

Country of ref document: EP

Kind code of ref document: A1