CN108121737B

CN108121737B - Method, device and system for generating business object attribute identifier

Info

Publication number: CN108121737B
Application number: CN201611079471.1A
Authority: CN
Inventors: 郝智恒; 李萍; 张泽聪; 沈晶晶; 李江
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2016-11-29
Filing date: 2016-11-29
Publication date: 2022-04-26
Anticipated expiration: 2036-11-29
Also published as: CN108121737A; TW201820230A; TWI787196B; WO2018099275A1

Abstract

The embodiment of the application provides a method, a device and a system for generating a business object attribute identifier, wherein the method comprises the following steps: determining the association degree between different business objects according to the behavior information of the user aiming at the business objects; generating a business object cluster according to the relevance; extracting a user label of the service object cluster; according to the attribute identification of the corresponding business object cluster generated by the user tag, the embodiment of the application automatically mines and identifies the information which is contained in the business object and related to the user behavior, and the information is not divided by a certain dimensionality of the business object, but is automatically mined and identified by the user aiming at the behavior information of the business object, so that the accuracy of identifying the attribute identification of the business object can be effectively improved.

Description

Method, device and system for generating business object attribute identifier

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and a device for generating a service object attribute identifier, a terminal for generating a service object attribute identifier, a server for generating a service object attribute identifier, and a system for generating a service object attribute identifier.

Background

For e-commerce type websites, each business object has some attributes for characterizing itself, such as the style of a certain brand of goods, the targeted consumer group, and the main push market of the brand.

When a business object is online to promote, a promotion strategy is generally formulated based on the attributes. One way is to define the target user according to the attributes of the business object, for example, the business object party defines the target user as a female living in a city at first line, aged between 28-35 years, and recently browsed a certain commodity according to the attributes of the business object, and then recommends the business object to the defined user. The other mode is that according to a machine learning algorithm, according to the attribute of a business object, the preference of a user to the business object is predicted, then the user with the probability of clicking or searching the business object or other similar business objects exceeding a certain threshold value is selected as a target user, and then directional recommendation is carried out.

In the first method, since the business object side needs to determine the attribute of the business object and then select the target user, the target user cannot be accurately selected due to the influence of the subjective factors of the business object side. In the second mode, due to the use of the machine learning algorithm, under the condition that an algorithm target function (such as a conversion rate) is fixed, and under the condition that the algorithm feature engineering has limitations on the extraction of attributes for and for business objects, a marvelo effect is easily caused, for example, for a new online business object, there is not much user data, and the machine learning algorithm often cannot find a more appropriate target user.

Disclosure of Invention

In view of the above problems, the embodiments of the present application are provided to provide a method for generating a service object attribute identifier, an apparatus for generating a service object attribute identifier, a terminal for generating a service object attribute identifier, a server for generating a service object attribute identifier, and a corresponding system for generating a service object attribute identifier, which overcome the above problems or at least partially solve the above problems.

In order to solve the above problem, the present application discloses a system for generating a service object attribute identifier, including: the system comprises a collecting unit, a display unit and a server;

the acquisition unit acquires behavior information of a user aiming at the business object and sends the behavior information to the server;

the server determines the association degree between different business objects after receiving the behavior information of the user aiming at the business objects, which is sent by the acquisition unit, and generates a business object cluster according to the association degree;

and the display unit extracts the user label of the service object cluster from the server and generates the attribute identifier of the corresponding service object cluster according to the user label.

In order to solve the above problem, the present application further discloses a method for generating a service object attribute identifier, including:

determining the association degree between different business objects according to the behavior information of the user aiming at the business objects;

generating a business object cluster according to the relevance;

extracting a user label of the service object cluster;

and generating an attribute identifier of the corresponding service object cluster according to the user label.

Optionally, the behavior information of the user for the business object is obtained by:

selecting an initial business object cluster;

and extracting the behavior information of the user aiming at the initial service object cluster in a preset time range.

Optionally, the step of determining the association between different business objects according to the behavior information of the user for the business objects includes:

generating a vector expression of each business object according to the behavior information of the user aiming at the business object;

determining similarity and support degree between different business objects by adopting the vector expression;

and determining the association degree between the different business objects by adopting the similarity and the support degree.

Optionally, the step of determining the association between the different business objects by using the similarity and the support includes:

and weighting the similarity and the support degree to obtain the association degrees between different business objects.

Optionally, the behavior information includes first behavior information and/or second behavior information, and the step of determining the similarity between different business objects according to the behavior information of the user for the business objects further includes:

determining a first association degree between different business objects according to the first behavior information;

determining a second association degree between the different business objects according to the second behavior information;

and weighting the first relevance and the second relevance to obtain the relevance between the different business objects.

Optionally, the step of generating a business object cluster according to the association degree includes:

when the association degree between different business objects exceeds a first preset threshold value, identifying that the different business objects have a target relationship;

generating a business object relation map according to the target relation among all the business objects;

and dividing the service object into a plurality of service object clusters by adopting the service object relation map.

Optionally, the step of generating a business object relationship map according to the target relationship among all business objects includes:

and respectively connecting different service objects with the target relationship to obtain a service object relationship map.

Optionally, the step of dividing the service object into a plurality of service object clusters by using the service object relationship map includes:

configuring a label for each service object in the service object relation map;

transmitting the label of each business object to the connected business objects;

selecting one label as an owned label from the labels received by each business object according to the number of the labels;

judging whether the label of each service object in the service object relation map changes or not, or whether the current label is smaller than the preset maximum iteration number or not;

if yes, returning to the step of transmitting the label of each business object to the connected business object;

if not, dividing the business objects with the same label into business object clusters.

Optionally, after the step of dividing the service object into a plurality of service object clusters by using the service object relationship map, the method further includes:

and checking the plurality of business object clusters.

Optionally, the plurality of service object clusters respectively have corresponding text information, and the step of verifying the plurality of service object clusters includes:

extracting key words in the text information of each service object cluster;

determining the text similarity between any two service object clusters according to the keywords;

and combining the two service object clusters with the text similarity exceeding a second preset threshold.

Optionally, the step of extracting the user tag of the business object cluster includes:

acquiring user information of a service object cluster;

identifying core users in the service object cluster by adopting the user information;

and extracting the user label of the core user.

Optionally, the step of identifying the core user in the service object cluster by using the user information includes:

sequencing the users in the service object cluster according to a preset dimension;

a preset number of core users are identified.

Optionally, after the step of generating the attribute identifier of the corresponding service object cluster according to the user tag, the method further includes:

determining a target user of the business object cluster by adopting the attribute identification of the business object cluster;

and recommending the target service object to the target user.

receiving a generation instruction of a business object attribute identifier;

submitting the generating instruction to a server;

receiving an attribute identifier of the service object sent by the server, wherein the attribute identifier of the service object is obtained by the server aiming at the generation instruction by extracting a user tag of a service object cluster to which the service object belongs;

and displaying the attribute identification of the business object.

In order to solve the above problem, the present application further discloses a device for generating a service object attribute identifier, including:

the association degree determining module is used for determining the association degree between different business objects according to the behavior information of the user aiming at the business objects;

the business object cluster generating module is used for generating a business object cluster according to the relevance;

the user label extraction module is used for extracting the user label of the service object cluster;

and the attribute identifier generating module is used for generating the attribute identifier of the corresponding service object cluster according to the user label.

Optionally, the behavior information of the user for the business object is obtained by invoking the following sub-modules:

the selecting submodule is used for selecting an initial business object cluster;

and the extraction submodule is used for extracting the behavior information of the user aiming at the initial service object cluster in a preset time range.

Optionally, the association degree determining module includes:

the vector expression generation submodule is used for generating a vector expression of each business object according to the behavior information of the user aiming at the business object;

the similarity and support degree determining submodule is used for determining the similarity and support degree between different business objects by adopting the vector expression;

and the association degree determining submodule is used for determining the association degree between the different business objects by adopting the similarity degree and the support degree.

Optionally, the association degree determining sub-module includes:

and the similarity and support weighting unit is used for weighting the similarity and support to obtain the association between different business objects.

Optionally, the behavior information includes first behavior information and/or second behavior information, and the association degree determining module further includes:

the first association degree determining submodule is used for determining first association degrees among different business objects according to the first behavior information;

the second association degree determining submodule is used for determining second association degrees among the different business objects according to the second behavior information;

and the relevancy weighting submodule is used for weighting the first relevancy and the second relevancy to obtain the relevancy between the different business objects.

Optionally, the business object cluster generating module includes:

the target relation identification submodule is used for identifying that target relation exists between different business objects when the association degree between the different business objects exceeds a first preset threshold;

the business object relation map generation submodule is used for generating a business object relation map according to the target relation among all the business objects;

and the service object cluster dividing submodule is used for dividing the service object into a plurality of service object clusters by adopting the service object relation map.

Optionally, the business object relationship map generating sub-module includes:

and the service object connection unit is used for respectively connecting different service objects with the target relationship to obtain a service object relationship map.

Optionally, the business object cluster partitioning sub-module includes:

the label configuration unit is used for configuring a label for each service object in the service object relation map;

the label transmitting unit is used for transmitting the label of each business object to the connected business objects;

the tag selecting unit is used for selecting one tag from the tags received by each business object as an owned tag according to the number of the tags;

the judging unit is used for judging whether the label of each service object in the service object relation map changes or not, or whether the current label is smaller than the preset maximum iteration number or not; if yes, calling the label transfer unit;

and the business object cluster dividing unit is used for dividing the business objects with the same label into business object clusters.

Optionally, the business object cluster generating module further includes:

and the service object cluster checking submodule is used for checking the plurality of service object clusters.

Optionally, the plurality of service object clusters respectively have corresponding text information, and the service object cluster verification sub-module includes:

the keyword extraction unit is used for extracting keywords in the text information of each business object cluster;

the text similarity determining unit is used for determining the text similarity between any two service object clusters according to the keywords;

and the service object cluster merging unit is used for merging the two service object clusters of which the text similarity exceeds a second preset threshold.

Optionally, the user tag extraction module includes:

the user information acquisition submodule is used for acquiring the user information of the service object cluster;

the core user identification submodule is used for identifying the core user in the service object cluster by adopting the user information;

and the user tag extraction submodule is used for extracting the user tag of the core user.

Optionally, the core user identification sub-module includes:

the sequencing unit is used for sequencing the users in the service object cluster according to a preset dimension;

and the identification unit is used for identifying the core users with the preset number.

Optionally, the apparatus further comprises:

the target user determining module is used for determining the target user of the business object cluster by adopting the attribute identifier of the business object cluster;

and the target service object recommending module is used for recommending the target service object to the target user.

the first receiving module is used for receiving a generation instruction of the business object attribute identifier;

the submitting module is used for submitting the generating instruction to a server;

a second receiving module, configured to receive an attribute identifier of the service object sent by the server, where the attribute identifier of the service object is obtained by the server by extracting a user tag of a service object cluster to which the service object belongs, for the generation instruction;

and the display module is used for displaying the attribute identification of the business object.

In order to solve the above problem, the present application further discloses a terminal for generating a service object attribute identifier, including:

one or more processors;

a memory; and

one or more programs, wherein the one or more programs are stored in the memory and configured for execution by the one or more processors to include instructions for:

receiving a generation instruction of a business object attribute identifier;

submitting the generating instruction to a server;

and displaying the attribute identification of the business object.

In order to solve the above problem, the present application further discloses a server for generating a service object attribute identifier, including:

one or more processors;

a memory; and

one or more modules stored in the memory and configured to be executed by the one or more processors, wherein the one or more modules have functionality to:

determining similarity between different business objects according to behavior information of a user aiming at the business objects;

generating a business object cluster according to the similarity;

extracting a user label of the service object cluster;

Compared with the background art, the embodiment of the application has the following advantages:

according to the embodiment of the application, the association degree between different business objects can be determined according to the behavior information of the user aiming at the business objects, the business object cluster is generated according to the association degree, then the user label of the business object cluster is extracted, so that the attribute identification of the corresponding business object cluster can be generated according to the user label, the information related to the user behavior, such as the life style, the demand preference, the style attribute and the like, which is contained in the business object is extracted, the information is not divided through a certain dimensionality of the business object, but the information is automatically mined and recognized according to the behavior information of the business object by the user, and the accuracy of recognizing the attribute identification of the business object can be effectively improved.

Secondly, in the embodiment of the application, the target user can be identified and recommended to the target user through the attribute identification of the service object cluster, so that the efficiency of identifying the target user is further improved, more related information about the service object can be provided for the service object party in the process of popularizing the service object, and the user group to be selected can be automatically defined for the service object party in a systematic manner. In addition, for the target user selection method based on machine learning, the attribute identifier of the business object generated in the embodiment of the application can also be used as an effective supplement of the feature engineering, so that the accuracy and the recall rate of the model are effectively improved.

Thirdly, when the business object is a brand of a commodity, the embodiment of the application can divide a plurality of brands into different brand groups according to the association degree by determining the association degree between the brands, and then generate different target user groups by identifying the attribute identification of the brand groups.

Drawings

Fig. 1 is a flowchart illustrating a first step of a first embodiment of a method for generating a service object attribute identifier according to the present application;

FIG. 2 is a schematic diagram of a business object relationship graph of the present application;

fig. 3 is a flowchart illustrating steps of a second embodiment of a method for generating a service object attribute identifier according to the present application;

FIG. 4 is a schematic diagram of the generation of an attribute identifier for a brand of merchandise of the present application;

fig. 5 is a flowchart illustrating a third step of a method for generating a service object attribute identifier according to the present application;

fig. 6A to 6D are block diagrams illustrating a first embodiment of a device for generating a service object attribute identifier according to the present application;

fig. 7 is a block diagram illustrating a second embodiment of a device for generating a service object attribute identifier according to the present application;

fig. 8 is a block diagram of an embodiment of a system for generating a service object attribute identifier according to the present application.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, the present application is described in further detail with reference to the accompanying drawings and the detailed description.

Referring to fig. 1, a flowchart illustrating a first step of a first embodiment of a method for generating a service object attribute identifier according to the present application is shown, which may specifically include the following steps:

step 101, determining the association degree between different business objects according to the behavior information of a user aiming at the business objects;

generally, there may be different business objects in different business fields, for example, in the communication field, the business object may be communication data; in the news media field, the business object may be news data; in the search field, the business object may be a web page; in the field of electronic commerce, a business object may be a good or brand of goods, and so on. The embodiment of the present application does not limit the specific type of the business object. Further, the behavior information of the user for the business object may also be different according to different business fields, for example, for a news data-type business object, the behavior information of the user may be a reading or searching behavior of the user for a certain news, and for a commodity or a commodity brand-type business object, the behavior information of the user may be a purchasing behavior or a browsing behavior of a certain commodity, and so on.

In the embodiment of the application, after behavior information of a user for a business object is collected, the association degree between different business objects can be calculated according to the behavior information, and the association degree can be a measure for the mutual relationship between different business objects.

In a specific implementation, the association degree may be through quantification of a relationship between certain dimensions, for example, similarity or support degree between business objects may be used as the association degree; or quantifying the relationship among multiple dimensions, for example, uniformly processing the similarity and the support degree between the business objects to obtain the association degree. Of course, those skilled in the art may also determine the association degree between the business objects in different manners according to actual needs, which is not limited in the embodiment of the present application.

In the embodiment of the application, the behavior information of the user for the business object can be obtained by selecting the initial business object cluster and extracting the behavior information of the user for the initial business object cluster in a preset time range.

Specifically, the range of the initial business object may be first selected according to specific business target division, for example, if some businesses are specialized for middle and high end brands, only the commodities in the range of the middle and high end brands may be selected, and if some businesses are only specific for women, only the commodities in the class of women's clothing need to be selected. Then, the behavior information of the user for the business object in the range can be extracted in a certain time period, and the behavior information of the user for the business object can be obtained.

In the embodiment of the present application, a vector expression of each business object may be generated according to behavior information of a user for the business object. For example, the behavior information of a certain number of users may be studied, and a vector relational expression between the user behavior and the business object may be formed. Specifically, taking whether 1000 users purchase a brand of goods as an example, if the 1 st, 3 rd and 6 th users purchase the brand of goods and none of the other users have any purchasing behavior, the generated vector expression may be (1,0,1,0,0,1,0, … …, 0).

Then, the expression can be adopted to further calculate the association degree between the business objects. Specifically, the vector expression may be used to determine similarity and support between different business objects.

In the embodiment of the present application, the similarity may be cosine similarity, and the cosine similarity is also called cosine similarity, and the similarity may be evaluated by calculating a cosine value of an included angle between two vectors, that is, an included angle between the two vectors may be obtained, and a cosine value corresponding to the included angle may be obtained, and the cosine value may be used to represent the similarity between the two vectors. In general, cosine values range between [ -1,1], and the more cosine values approach 1, the more directions of two vectors approach 0, the more consistent the directions are, and the corresponding similarity is higher. While the degree of support indicates a degree of support, typically expressed as a percentage. For example, if 400 users purchased a brand of goods among 1000 users, the user may be assumed to have a 40% support for the brand of goods.

In the embodiment of the application, the cosine similarity between different service objects can be calculated directly by adopting a vector expression of the service object according to a known calculation formula of the cosine similarity. For the support degree between the business objects, the specific number of the supporters can be extracted from the vector expression, and the specific value between the actual number of the supporters and the total number of the supporters for different business objects is adopted to determine. For example, for certain brands a and B, where the number of users supporting (e.g., purchasing behavior) brand a is 10, and the number of users supporting brand B is 15, where 5 users support both brand a and brand B, the support between certain brands a and B can be determined to be (10+15-5)/(10+15) × 100% — 80%.

After calculating the similarity and the support degree between different business objects, the similarity and the support degree can be adopted to determine the association degree between the different business objects.

In a specific implementation, different weights may be set for the similarity and the support degree, respectively, and then the similarity and the support degree are weighted to obtain the association degrees between different business objects. The embodiment of the present application does not limit the specific values of the set weights of the similarity and the support degree. Of course, a person skilled in the art may also calculate the association degree between different business objects in other manners according to actual needs, for example, the association degree between different business objects may be obtained by calculating indexes such as confidence degrees or promotion degrees between business objects, which is not limited in this embodiment of the present application.

As another example of the present application, the behavior information of the user for the business object may include a plurality of kinds, for example, the first behavior information and the second behavior information may be included. Taking the business object as a brand product as an example, the first behavior information may be a purchasing behavior of a user for a certain brand, and the second behavior information may be a shopping cart adding behavior of the user for the brand, that is, a behavior that the brand product is put into a shopping cart of an e-commerce website but is not purchased temporarily.

Taking the example that the user behavior information includes first behavior information and second behavior information, when determining the association degree between different business objects, first determining a first association degree between different business objects according to the first behavior information, and determining a second association degree between different business objects according to the second behavior information; and then weighting the first relevance and the second relevance to obtain the relevance between the different business objects. Of course, the behavior information of the user may further include third behavior information, fourth behavior information, and other more types of behavior information, which is not limited in this embodiment of the application.

102, generating a business object cluster according to the relevance;

in the embodiment of the present application, after obtaining the association degrees between different business objects, all the business objects may be divided into a plurality of different business objects for calculation according to the difference of the association degrees, so that the business objects in each business object cluster have higher similarity.

In a specific implementation, a threshold may be preset to distinguish similarity between the business objects, for example, the first preset threshold may be set to 80%, and when the association degree between different business objects exceeds the first preset threshold, a target relationship is identified between the different business objects, where the target relationship may be that when the association degree exceeds the above-mentioned threshold of 80%, two different business objects have higher similarity. Of course, a person skilled in the art may set the specific size of the first preset threshold according to actual needs, which is not limited in the embodiment of the present application.

Then, a business object relationship map, which is a relationship network map formed according to the relationship between business objects, may be generated according to the target relationship between all business objects. Specifically, different business objects having the target relationship may be connected respectively to obtain a business object relationship map. For example, for business object A, B, C, D, if business objects A and C, D have the target relationship, business objects B and D have the target relationship, and business objects C and D also have the target relationship, A and C, A and D, B and D, and C and D may be connected two by two to form a relationship graph.

Finally, the service object may be divided into a plurality of service object clusters by using the service object relationship map.

In a preferred embodiment of the present application, the sub-step of dividing the service object into a plurality of service object clusters by using the service object relationship map may further include:

s11, configuring a label for each business object in the business object relation map;

s12, transmitting the label of each business object to the connected business object;

s13, selecting one label as the owned label according to the number of the labels from the labels received by each business object;

s14, judging whether the label of each service object in the service object relation map changes or not, or whether the current number of iterations is less than the preset maximum number of iterations;

s15, if yes, returning to the step of transmitting the label of each business object to the connected business object;

and S16, if not, dividing the business objects with the same label into business object clusters.

In a specific implementation, for convenience of calculation, the tag configured for the business object may be an ID thereof, and of course, the tag may also be configured in other manners, such as random configuration, as long as the uniqueness of the tag is maintained, which is not limited in this embodiment of the present application.

In the first iteration, the labels can be randomly selected, and because the nodes of the core in the service object relationship graph are connected with other peripheral nodes, the probability that the labels are randomly selected is high, and in the subsequent iteration process, the number of the labels of the nodes of the core can be increased and gradually becomes stable.

When the label is stable or reaches the maximum iteration number, the service objects with the same label can be regarded as belonging to the same service object cluster, and the label of the node can be regarded as the identification label of the service object.

For example, as shown in fig. 2, which is a schematic diagram of a business object relationship graph of the present application, taking names of nodes as labels of business objects, that is, labels of the nodes R, S, T, U are R, S, T, U respectively, then the iterative process is as follows:

after the 3 rd iteration, the labels owned by the business objects are all R, and no change occurs, so that the business objects corresponding to the node R, S, T, U can be considered to belong to the same cluster, and can be divided into the same business object cluster.

The description of the service object relationship graph is simple, and is only used as an example for introducing the service object cluster division according to the embodiment of the present application, and in actual use, the number of service objects included in the service object relationship graph may be very large. Of course, a person skilled in the art may also select other ways to implement the division of the service object relationship graph into a plurality of different service object clusters, for example, a clustering method, a community division algorithm, and the like, which is not limited in this embodiment of the present application.

In the embodiment of the present application, after dividing the service object into a plurality of different service object clusters, in order to determine whether the obtained division result is reasonable and accurate, the plurality of service object clusters may also be verified.

Generally, the business object cluster may include corresponding text information, for example, in case of the business object cluster being some similar brand, the text information may be an advertisement of each business object (i.e. each brand) in the cluster, customer rating information, brand slogan, brand culture information, etc.

In a specific implementation, a keyword in the text information of each service object cluster may be extracted first, and then the text similarity between any two service object clusters is determined according to the keyword.

For example, for the text information "i love in beijing tiananmen", after keyword extraction, the text information may be "i", "love", "beijing", "tiananmen", and many similar texts may be changed into some combinations of keywords in such a manner; for another text message "i love yellow river", after word segmentation and keyword extraction, the text message may be "i", "love", "yellow river", and then two keywords between the two text messages are the same, and three words are different, so that the similarity between the two text messages is 2/5 ═ 0.4.

Of course, the above example is only for explaining the keyword extraction and the text similarity calculation process, and those skilled in the art may also select other ways to do the above process, and the embodiment of the present application does not limit this.

In this embodiment of the present application, a threshold may be set for the text similarity between the service object clusters, for example, a second preset threshold may be set to be 90%, and after the text similarity between any two service object clusters is obtained, the two service object clusters with the text similarity exceeding the second preset threshold may be merged.

103, extracting the user label of the service object cluster;

in this embodiment of the present application, after dividing a plurality of service objects into different service object clusters, user information of a service object cluster to be researched may be further obtained, where the user information may be determined according to a user owned by each service object in the service object cluster, for example, a user group of each service object in the service object cluster may be first identified, and then information of the user groups of all service objects is used as the user information of the service object cluster.

Then, the user information may be used to identify core users in the business object cluster. Specifically, the users in the service object cluster may be sorted according to a preset dimension, and then a preset number of core users may be identified.

For example, if a business object cluster is composed of a plurality of similar product brands, the core users in the business object cluster may be considered as a class of users consuming more products of the brands. Therefore, according to the dimension of the consumption amount, firstly, the consumption amount of each user to all brands in the service object cluster is counted, then, the ranking is carried out according to the consumption amount, and the users with the consumption amount being 20% in the top are identified as core users. Of course, for different types of service objects or service object clusters, the identification standard of the core user may also be different, and a person skilled in the art may set a specific manner for identifying the core user in the service object cluster according to actual needs, which is not limited in the embodiment of the present application.

Typically, a user may have their own user tags, e.g., age, work information, city of residence, consumption preferences, and so forth. After the core user in the service object cluster is identified, the user tag of the core user can be further extracted.

And 104, generating an attribute identifier of the corresponding business object cluster according to the user label.

In the embodiment of the present application, a user tag of a core user in a business object cluster may be used as an attribute identifier of the business object cluster.

In the embodiment of the application, the association between different business objects can be determined according to the behavior information of the user for the business objects, the business object cluster is generated according to the association, then the user label of the business object cluster is extracted, so that the attribute identification of the corresponding business object cluster can be generated according to the user label.

In the embodiment of the application, after the attribute identifier of the service object cluster is generated, the attribute identifier of the service object cluster can be further adopted to determine the target user of the service object cluster, and then the target service object is recommended to the target user.

Generally, there may be a plurality of attribute identifiers of the generated service objects, and therefore, after the attribute identifiers of the service object cluster are generated, one or more attribute identifiers of the attribute identifiers may be used to determine the target user of the service object cluster, for example, for the attribute identifier of the service object cluster being "male, 18-22 years old, a two-line city, or a consumption level with a medium bias", two attribute identifiers of "male, 18-22" may be selected, so as to identify the user with the two attribute identifiers as the target user, and further recommend the target service object to the target user.

In a specific implementation, the target service object may be a certain service object in the service object cluster, or may not be a service object in the service object cluster, but is other service objects having higher similarity to the service object in the service object cluster, which is not limited in this embodiment of the present application.

In the embodiment of the application, the target user can be identified and the target service object is recommended to the target user through the attribute identification of the service object cluster, so that the efficiency of identifying the target user is further improved, more related information about the service object can be provided for a service object party in the process of popularizing the service object, and the user group to be selected can be automatically defined for the service object party in a systematic mode. In addition, for the target user selection method based on machine learning, the attribute identifier of the business object generated in the embodiment of the application can also be used as an effective supplement of the feature engineering, so that the accuracy and the recall rate of the model are effectively improved.

For convenience of understanding, the following describes a method for generating a business object attribute identifier according to the present application, taking a business object as an example of a brand of a commodity.

Referring to fig. 3, a flowchart illustrating steps of a second embodiment of the method for generating a service object attribute identifier according to the present application is shown, and specifically, the method may include the following steps:

step 301, determining the association degree between different business objects according to the behavior information of the user aiming at the business objects;

when the business object is a brand of a commodity, the behavior information of the user for the business object is a purchasing behavior, a browsing behavior, a searching behavior, a shopping cart placement and the like of the user for the commodity of the brand.

Fig. 4 is a schematic diagram illustrating generation of an attribute identifier of a brand of merchandise according to the present invention. In fig. 4, the behavior information of the user includes three categories of purchasing behavior, browsing behavior, and putting in a shopping cart.

In specific implementation, the purchase association degree among different brands can be calculated according to the information of various brands purchased by users; calculating the browsing association degree among different brands aiming at the information of various brands browsed by a user; and calculating the purchasing association degree among different brands according to the information that the user puts various brands of commodities into the shopping cart.

In a specific implementation, taking the purchase association degree as an example, if 500 users purchase a brand a commodity in 1000 users, a vector expression a of the purchase behavior of the brand a by the users may be generated; if 700 users of the 1000 users purchase the brand B, a vector expression B of the purchasing behavior of the user on the brand B may be generated in the same manner, and then the similarity between the brand a and the brand B may be calculated by using the vector expression a and the vector expression B.

In addition, the support degree, the confidence degree, the promotion degree and the like between the brand A and the brand B can be calculated by adopting the vector expression A and the vector expression B, and the similarity, the support degree, the confidence degree and the promotion degree are weighted, so that the purchase association degree between the brand A and the brand B is obtained. Of course, those skilled in the art may specifically select the weighted object according to actual needs, for example, the similarity result may be used as the purchase association degree, or the support degree may be added on the basis of the similarity, and then the weighting is performed, and the like, which is not limited in the embodiment of the present application.

Similarly, the calculation process of the browsing association degree and the purchasing association degree between the brand a and the brand B is similar to the calculation process of the purchasing association degree, and details are not repeated here in the embodiment of the present application. After the three types of similarity are obtained, the purchase association degree, the browsing association degree and the purchase adding association degree can be weighted respectively, so that the association degree between the brand A and the brand B is obtained.

Step 302, generating a service object cluster according to the similarity;

in this embodiment, when the business object is a brand of a commodity, the business object cluster is a different brand group.

In specific implementation, after the similarity between every two brands is obtained through calculation, the two brands with the association degree exceeding a first preset threshold value can be connected to generate a brand relationship map, the brand relationship map can be divided by adopting an iterative hierarchical clustering algorithm or other community discovery algorithms to obtain a plurality of brand groups, and each brand group comprises a plurality of different brands with proper quantity. For example, there may be around 6-8 different brands in each set of brands.

Step 303, checking the plurality of service object clusters;

in this embodiment, after obtaining a plurality of brand groups, each brand group may be checked to determine whether the division of the brand relationship graph as accomplished in step 302 is reasonable and effective.

In specific implementation, the text information of the brand group can be generated according to the advertisement words, the consumer evaluation information, the brand slogan, the brand culture information and the like of each brand in the brand group, and then the text similarity between every two brand groups is calculated by extracting the keywords in the text information. If the text similarity exceeds a preset second threshold, the similarity between the two brand groups can be considered to be high, and the processes can be combined.

Step 304, extracting the user label of the service object cluster;

in the embodiment of the present application, when extracting the user tags of the set of brands, the core users of the set of brands may be identified first, for example, the core users may be users who purchase the first 20% of the consumption amount of the goods of each brand in the set of brands. Then, the user tags of the above-mentioned top 20% of users are extracted as the user tags of the brand group.

305, generating an attribute identifier of a corresponding business object cluster according to the user label;

after the label of the core user of each brand group is obtained, the attribute identifier of the corresponding brand group can be generated according to the user label. For example, for a certain brand group, the attribute identifier may be "young, male, two-line city, middle-high-end consumption, sports, outdoor, fashion", and so on.

Step 306, determining a target user of the business object cluster by adopting the attribute identifier of the business object cluster;

in specific implementation, after the attribute identifier of each brand group is generated, target user population results can be generated according to the actual requirements of brand parties, such as directional putting requirements, and according to the attribute identifiers of the brand groups when brand promotion is performed, for example, "european style furniture population", "smart home feverish people", "light luxury people who reach, high-end mother and infant population", "eat family" and the like can be obtained, and then the target user population suitable for the user can be selected from the results.

Step 307, recommending a target service object to the target user.

In the embodiment of the present application, when the business object is a brand of a commodity, the target business object is a specific brand of a commodity. Specifically, the commodity brand may be a certain commodity brand in the commodity brand group, or may be a commodity brand in the non-commodity brand group, which is not limited in this application embodiment. So that the target brand of goods can be recommended to the target user.

In the embodiment of the application, the similarity between the brands can be determined, so that the brands are divided into different brand groups according to the similarity, different target user groups are generated by identifying the attribute identifiers of the brand groups, and when a brand party carries out brand promotion, the information of the related brands can be obtained, so that users who are most interested in the brands can be found, a brand owner can be facilitated to quickly and conveniently locate the target users, the accuracy of the location of the target users is improved, and the maximization of the benefit of the brand promotion is facilitated.

Referring to fig. 5, a flowchart illustrating a third step of the embodiment of the method for generating a service object attribute identifier according to the present application is shown, which may specifically include the following steps:

step 501, receiving a generation instruction of a business object attribute identifier;

step 502, submitting the generation instruction to a server;

step 503, receiving an attribute identifier of the service object sent by the server, wherein the attribute identifier of the service object is obtained by the server by extracting a user tag of a service object cluster to which the service object belongs, for the generation instruction;

step 504, showing the attribute identification of the business object.

In the embodiment of the application, when the attribute identifier of the service object needs to be generated, a generation instruction of the attribute identifier of the service object may be sent to the terminal, after the terminal receives the generation instruction, the generation instruction may be submitted to the server, the server obtains the attribute identifier of the service object according to the user tag of the service object cluster to which the service object belongs, and then feeds back the attribute identifier to the terminal, and after the terminal receives the attribute identifier of the service object fed back by the server, the terminal may display the attribute identifier on a user interface of the terminal.

Since the process of generating the service object attribute identifier by the server in this embodiment is similar to steps 101 to 104 in the first method embodiment and steps 301 to 305 in the second method embodiment, reference may be made to each other, and details of this embodiment are not repeated.

It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the embodiments are not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the embodiments. Further, those skilled in the art will also appreciate that the embodiments described in the specification are presently preferred and that no particular act is required of the embodiments of the application.

Referring to fig. 6A, a block diagram of a first structure of an embodiment of a device for generating a service object attribute identifier according to the present application is shown, which may specifically include the following modules:

the association degree determining module 601 is configured to determine association degrees between different business objects according to behavior information of a user for the business objects;

a business object cluster generating module 602, configured to generate a business object cluster according to the association degree;

a user tag extracting module 603, configured to extract a user tag of the service object cluster;

an attribute identifier generating module 604, configured to generate an attribute identifier of a corresponding service object cluster according to the user tag.

Referring to fig. 6B, a second structural block diagram of a first embodiment of the apparatus for generating a service object attribute identifier according to the present application is shown, where the behavior information of the user for the service object may be obtained by invoking the following sub-modules:

a selecting submodule 6011 configured to select an initial service object cluster;

the extracting submodule 6012 is configured to extract behavior information of the user for the initial service object cluster within a preset time range.

In this embodiment of the application, the association degree determining module 601 may further include the following sub-modules:

a vector expression generation submodule 6013, configured to generate a vector expression for each service object according to behavior information of a user for the service object;

a similarity and support degree determining submodule 6014, configured to determine, by using the vector expression, a similarity and support degree between different service objects;

and an association degree determining submodule 6015, configured to determine, by using the similarity and the support degree, an association degree between the different business objects.

In this embodiment of the present application, the relevancy determination submodule 6015 may specifically include the following units:

In this embodiment of the application, the behavior information may include first behavior information and/or second behavior information, and the association degree determining module 601 may further include the following sub-modules:

a first association degree determining submodule 6016, configured to determine, according to the first behavior information, a first association degree between different business objects;

a second association degree determining submodule 6017, configured to determine, according to the second behavior information, a second association degree between the different business objects;

and an association weighting submodule 6018, configured to weight the first association and the second association, and obtain an association between the different business objects.

Referring to fig. 6C, a third structural block diagram of a first embodiment of a device for generating a service object attribute identifier according to the present application is shown, where the service object cluster generating module 602 may specifically include the following sub-modules:

the target relationship identification submodule 6021 is configured to identify that a target relationship exists between different business objects when the association degree between the different business objects exceeds a first preset threshold;

a business object relationship map generation submodule 6022 for generating a business object relationship map according to the target relationship among all the business objects;

and a business object cluster division submodule 6023, configured to divide the business object into a plurality of business object clusters by using the business object relationship map.

In this embodiment of the present application, the business object relationship map generating sub-module 6022 may specifically include the following units:

In this embodiment of the present application, the business object cluster partitioning submodule 6023 may specifically include the following units:

In this embodiment of the present application, the business object cluster generating module 602 may further include the following sub-modules:

and the business object cluster checking submodule 6024 is configured to check the multiple business object clusters.

In this embodiment of the present application, the plurality of service object clusters may respectively have corresponding text information, and the service object cluster verification sub-module 6024 may specifically include the following units:

Referring to fig. 6D, a fourth structural block diagram of a first embodiment of the apparatus for generating a service object attribute identifier according to the present application is shown, where the user tag extraction module 603 specifically includes the following sub-modules:

a user information obtaining sub-module 6031, configured to obtain user information of the service object cluster;

a core user identification submodule 6032, configured to identify a core user in the service object cluster by using the user information;

and the user tag extraction sub-module 6033 is configured to extract the user tag of the core user.

In this embodiment, the core subscriber identity submodule 6032 may specifically include the following units:

In this embodiment, the apparatus may further include the following modules:

Referring to fig. 7, a block diagram of a second embodiment of the apparatus for generating a service object attribute identifier according to the present application is shown, and specifically, the second embodiment of the apparatus may include the following modules:

a first receiving module 701, configured to receive a generation instruction of a service object attribute identifier;

a submitting module 702, configured to submit the generating instruction to a server;

a second receiving module 703, configured to receive an attribute identifier of the service object sent by the server, where the attribute identifier of the service object is obtained by the server by extracting a user tag of a service object cluster to which the service object belongs, for the generating instruction;

and a display module 704, configured to display the attribute identifier of the service object.

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

Referring to fig. 8, a block diagram of a system for generating a service object attribute identifier according to the present application is shown, where the system specifically includes: an acquisition unit 801, a server 802, and a display unit 803;

the acquisition unit 801 acquires behavior information of a user for a business object, and sends the behavior information to the server 802;

the server 802, after receiving the behavior information of the user for the business object sent by the acquisition unit 801, determines the association degree between different business objects, and generates a business object cluster according to the association degree;

the display unit 803 extracts the user tag of the service object cluster from the server 802, and generates an attribute identifier of the corresponding service object cluster according to the user tag.

The embodiment of the present application further discloses a terminal for generating a service object attribute identifier, including:

one or more processors;

a memory; and

receiving a generation instruction of a business object attribute identifier;

submitting the generating instruction to a server;

and displaying the attribute identification of the business object.

The embodiment of the present application further discloses a server for generating a service object attribute identifier, including:

one or more processors;

a memory; and

generating a business object cluster according to the similarity;

extracting a user label of the service object cluster;

The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

As will be appreciated by one of skill in the art, embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, 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 aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

In a typical configuration, the computer device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium. Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, 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 disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (fransitory media), such as modulated data signals and carrier waves.

Embodiments of the present application are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present application have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the true scope of the embodiments of the application.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

Detailed descriptions are given above to a method for generating a service object attribute identifier, a device for generating a service object attribute identifier, a terminal for generating a service object attribute identifier, a server for generating a service object attribute identifier, and a system for generating a service object attribute identifier, where a specific example is applied in the description of the principles and embodiments of the present application, and the description of the above embodiments is only used to help understand the method and core ideas of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A system for generating a business object attribute identifier, comprising: the system comprises a collecting unit, a display unit and a server;

the display unit extracts the user label of the service object cluster from the server and generates the attribute identification of the corresponding service object cluster according to the user label;

wherein, the generating the service object cluster according to the relevancy comprises:

2. A method for generating a service object attribute identifier is characterized by comprising the following steps:

generating a business object cluster according to the relevance;

extracting a user label of the service object cluster;

generating attribute identification of a corresponding business object cluster according to the user label;

wherein the step of generating the service object cluster according to the relevance comprises the following steps:

3. The method according to claim 2, wherein the behavior information of the user for the business object is obtained by:

selecting an initial business object cluster;

4. The method according to claim 2, wherein the step of determining the association degree between different business objects according to the behavior information of the user for the business objects comprises:

5. The method according to claim 4, wherein the step of determining the association between the different business objects using the similarity and the support comprises:

6. The method according to any one of claims 2 to 5, wherein the behavior information includes first behavior information and/or second behavior information, and the step of determining the association degree between different business objects according to the behavior information of the user for the business objects further includes:

7. The method of claim 2, wherein the step of generating a business object relationship graph according to the target relationship among all business objects comprises:

8. The method of claim 7, wherein the step of dividing the service objects into a plurality of service object clusters using the service object relationship graph comprises:

configuring a label for each service object in the service object relation map;

9. The method according to any one of claims 2, 7 or 8, wherein after the step of dividing the service object into a plurality of service object clusters by using the service object relationship graph, the method further comprises:

and checking the plurality of business object clusters.

10. The method of claim 9, wherein the plurality of business object clusters respectively have corresponding text information, and the step of verifying the plurality of business object clusters comprises:

extracting key words in the text information of each service object cluster;

11. The method of claim 2, wherein the step of extracting the user tag of the business object cluster comprises:

acquiring user information of a service object cluster;

and extracting the user label of the core user.

12. The method of claim 11, wherein the step of using the user information to identify core users in the business object cluster comprises:

a preset number of core users are identified.

13. The method according to claim 2, wherein after the step of generating the attribute identifier of the corresponding business object cluster according to the user tag, the method further comprises:

and recommending the target service object to the target user.

14. A method for generating a service object attribute identifier is characterized by comprising the following steps:

receiving a generation instruction of a business object attribute identifier;

submitting the generating instruction to a server;

receiving an attribute identifier of the service object sent by the server, wherein the attribute identifier of the service object is obtained by the server aiming at the generation instruction by extracting a user tag of a service object cluster to which the service object belongs; the service object cluster to which the service object belongs determines the association degree between different service objects according to the behavior information of a user for the service object, when the association degree between different service objects exceeds a first preset threshold value, the service object cluster identifies that target relations exist between different service objects, a service object relation map is generated according to the target relations between all service objects, and the service object is divided by adopting the service object relation map;

and displaying the attribute identification of the business object.

15. An apparatus for generating attribute identifiers of business objects, comprising:

the attribute identifier generating module is used for generating the attribute identifier of the corresponding business object cluster according to the user label;

the service object cluster generating module comprises:

16. An apparatus for generating attribute identifiers of business objects, comprising:

a second receiving module, configured to receive an attribute identifier of the service object sent by the server, where the attribute identifier of the service object is obtained by the server by extracting a user tag of a service object cluster to which the service object belongs, for the generation instruction; the service object cluster to which the service object belongs determines the association degree between different service objects according to the behavior information of a user for the service object, when the association degree between different service objects exceeds a first preset threshold value, the service object cluster identifies that target relations exist between different service objects, a service object relation map is generated according to the target relations between all service objects, and the service object is divided by adopting the service object relation map;

17. A terminal for generating a service object attribute identifier, comprising:

one or more processors;

a memory; and

receiving a generation instruction of a business object attribute identifier;

submitting the generating instruction to a server;

and displaying the attribute identification of the business object.

18. A server for generating a business object attribute identifier, comprising:

one or more processors;

a memory; and

generating a business object cluster according to the relevance;

extracting a user label of the service object cluster;